This document summarizes a research paper on sentiment analysis using fuzzy logic. It discusses how sentiment analysis of consumer reviews can help both producers and consumers make effective decisions by understanding opinions on products and services. It also describes how fuzzy logic can be used to analyze sentiment values more precisely. The document outlines different approaches to sentiment analysis, including classifying reviews as positive or negative and determining the degree of sentiment. Twitter is presented as a source of consumer opinions that can be analyzed for sentiment using techniques like identifying emoticons, hashtags and targets.
Review on Opinion Mining for Fully Fledged Systemijeei-iaes
Humans communication is generally under the control of emotions and full of opinions. Emotions an d their opinions plays an important role in thinking process of mind, influences the human actions too. Sentiment analysis is one of the ways to explore user’s opinion made on any social media and networking site for various commercial applications in number of fields. This paper takes into account the basis requirements of opinion mining to explore the present techniques used to develop a fully fledged system. Is highlights the opportunities or deployment and research of such systems. The available tools used for building such applications have even presented with their merits and limitations.
This document provides an overview of opinion mining and sentiment analysis. It discusses how opinion mining is used to analyze sentiments expressed by people on the web through reviews and opinions. It covers the basic terminology of opinion mining, different data sources like blogs and reviews sites, techniques for sentiment classification including machine learning and lexicon-based approaches, challenges in opinion mining, tools used, and applications. Key aspects of opinion mining discussed include identifying opinion holders, objects, determining document, phrase and sentence level sentiment polarity, and using sentiment lexicons and classifiers.
The document discusses analyzing sentiment towards employee stock ownership plans (ESOP) on social media using sentiment analysis and clustering algorithms. It collects feedback on ESOP from four social monitoring tools - Social Mention, Trackur, Twendz, and Twitratr. It then uses K-means clustering, Expectation Maximization clustering, and VAR K-Means algorithms in the Tanagra1.4 data mining tool to cluster the results. The analysis finds that cluster 2 consistently indicates more negative sentiment than clusters 1 and 3, regardless of the clustering algorithm used.
IRJET- Real Time Sentiment Analysis of Political Twitter Data using Machi...IRJET Journal
This document summarizes a research paper that analyzed sentiments of political tweets related to the Ayodhya issue in India using machine learning. It collected tweets using keywords and preprocessed them by removing URLs, usernames, stop words, and irrelevant data. It then extracted sentiment-bearing words as features. It classified the polarity of each tweet as positive, negative, or neutral using the Vader sentiment analysis tool and calculated overall sentiment scores. It aimed to analyze public opinion on the Ayodhya issue expressed on Twitter.
This document summarizes a research paper on opinion mining from Twitter data. It discusses the challenges of sentiment analysis on short Twitter posts, including named entity recognition, anaphora resolution, parsing, and detecting sarcasm. It also reviews several papers on related topics, such as frameworks for Twitter opinion mining using classification techniques, using Twitter as a corpus for sentiment analysis, and analyzing opinions during the 2012 Korean presidential election on Twitter. Overall, it covers key techniques in opinion mining like identifying opinion targets and orientation. It proposes future work to develop a web application to compare Twitter opinion mining performance and use supervised learning to improve accuracy.
Sentiment Analysis is the process of finding the sentiments from different classes of words.
Generally speaking, sentiment analysis aims to determine the attitude of a speaker or a writer with
respect to some topic or the overall contextual polarity of a document. The attitude may be his or
her judgment or evaluation, affective state, or the intended emotional communication. In this case,
‘tweets’! Given a micro-blogging platform where official, verified tweets are available to us, we
need to identify the sentiments of those tweets. A model must be constructed where the sentiments
are scored, for each product individually and then they are compared with, diagrammatically,
portraying users’ feedback from the producers stand point.
There are many websites that offer a comparison between various products or services based on
certain features of the article such as its predominant traits, price, and its welcome in the market and
so on. However not many provide a juxtaposing of commodities with user review as the focal point.
Those few that do work with Naïve Bayes Machine Learning Algorithms, that poses a disadvantage
as it mandatorily assumes that the features, in our project, words, are independent of each other.
This is a comparatively inefficient method of performing Sentiment Analysis on bulk text, for
official purposes, since sentences will not give the meaning they are supposed to convey, if each
word is considered a separate entity. Maximum Entropy Classifier overcomes this draw back by
limiting the assumptions it makes of the input data feed, which is what we use in the proposed
system.
Sentiment Analysis on Twitter Dataset using R Languageijtsrd
Sentiment Analysis involves determining the evaluative nature of a piece of text. A product review can express a positive, negative, or neutral sentiment or polarity . Automatically identifying sentiment expressed in text has a number of applications, including tracking sentiment towards Movie reviews and Automobile reviews improving customer relation models, detecting happiness and well being, and improving automatic dialogue systems. The evaluative intensity for both positive and negative terms changes in a negated context, and the amount of change varies from term to term. To adequately capture the impact of negation on individual terms, here proposed to empirically estimate the sentiment scores of terms in negated context from movie review and auto mobile review, and built two lexicons, one for terms in negated contexts and one for terms in affirmative non negated contexts. By using these Affirmative Context Lexicons and Negated Context Lexicons were able to significantly improve the performance of the overall sentiment analysis system on both tasks. This thesis have proposed a sentiment analysis system that detects the sentiment of corpus dataset using movie review and Automobile review as well as the sentiment of a term a word or a phrase within a message term level task using R language. B. Nagajothi | Dr. R. Jemima Priyadarsini "Sentiment Analysis on Twitter Dataset using R Language" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-3 | Issue-6 , October 2019, URL: https://www.ijtsrd.com/papers/ijtsrd28071.pdf Paper URL: https://www.ijtsrd.com/computer-science/data-miining/28071/sentiment-analysis-on-twitter-dataset-using-r-language/b-nagajothi
IRJET- Interpreting Public Sentiments Variation by using FB-LDA TechniqueIRJET Journal
This document discusses sentiment analysis techniques for classifying tweets based on their positive, negative, or neutral sentiment. It proposes two Latent Dirichlet Allocation (LDA) based models - Foreground and Background LDA (FB-LDA) and Reason Candidate and Background LDA (RCB-LDA) - to analyze sentiment variation in tweets. FB-LDA can filter background topics and extract foreground topics to identify possible explanations for sentiment changes. RCB-LDA can rank reason candidates expressed in tweets to provide sentence-level sentiment explanations. The proposed techniques are intended to classify tweets and evaluate public sentiment variations by extracting possible reasons for those variations.
Review on Opinion Mining for Fully Fledged Systemijeei-iaes
Humans communication is generally under the control of emotions and full of opinions. Emotions an d their opinions plays an important role in thinking process of mind, influences the human actions too. Sentiment analysis is one of the ways to explore user’s opinion made on any social media and networking site for various commercial applications in number of fields. This paper takes into account the basis requirements of opinion mining to explore the present techniques used to develop a fully fledged system. Is highlights the opportunities or deployment and research of such systems. The available tools used for building such applications have even presented with their merits and limitations.
This document provides an overview of opinion mining and sentiment analysis. It discusses how opinion mining is used to analyze sentiments expressed by people on the web through reviews and opinions. It covers the basic terminology of opinion mining, different data sources like blogs and reviews sites, techniques for sentiment classification including machine learning and lexicon-based approaches, challenges in opinion mining, tools used, and applications. Key aspects of opinion mining discussed include identifying opinion holders, objects, determining document, phrase and sentence level sentiment polarity, and using sentiment lexicons and classifiers.
The document discusses analyzing sentiment towards employee stock ownership plans (ESOP) on social media using sentiment analysis and clustering algorithms. It collects feedback on ESOP from four social monitoring tools - Social Mention, Trackur, Twendz, and Twitratr. It then uses K-means clustering, Expectation Maximization clustering, and VAR K-Means algorithms in the Tanagra1.4 data mining tool to cluster the results. The analysis finds that cluster 2 consistently indicates more negative sentiment than clusters 1 and 3, regardless of the clustering algorithm used.
IRJET- Real Time Sentiment Analysis of Political Twitter Data using Machi...IRJET Journal
This document summarizes a research paper that analyzed sentiments of political tweets related to the Ayodhya issue in India using machine learning. It collected tweets using keywords and preprocessed them by removing URLs, usernames, stop words, and irrelevant data. It then extracted sentiment-bearing words as features. It classified the polarity of each tweet as positive, negative, or neutral using the Vader sentiment analysis tool and calculated overall sentiment scores. It aimed to analyze public opinion on the Ayodhya issue expressed on Twitter.
This document summarizes a research paper on opinion mining from Twitter data. It discusses the challenges of sentiment analysis on short Twitter posts, including named entity recognition, anaphora resolution, parsing, and detecting sarcasm. It also reviews several papers on related topics, such as frameworks for Twitter opinion mining using classification techniques, using Twitter as a corpus for sentiment analysis, and analyzing opinions during the 2012 Korean presidential election on Twitter. Overall, it covers key techniques in opinion mining like identifying opinion targets and orientation. It proposes future work to develop a web application to compare Twitter opinion mining performance and use supervised learning to improve accuracy.
Sentiment Analysis is the process of finding the sentiments from different classes of words.
Generally speaking, sentiment analysis aims to determine the attitude of a speaker or a writer with
respect to some topic or the overall contextual polarity of a document. The attitude may be his or
her judgment or evaluation, affective state, or the intended emotional communication. In this case,
‘tweets’! Given a micro-blogging platform where official, verified tweets are available to us, we
need to identify the sentiments of those tweets. A model must be constructed where the sentiments
are scored, for each product individually and then they are compared with, diagrammatically,
portraying users’ feedback from the producers stand point.
There are many websites that offer a comparison between various products or services based on
certain features of the article such as its predominant traits, price, and its welcome in the market and
so on. However not many provide a juxtaposing of commodities with user review as the focal point.
Those few that do work with Naïve Bayes Machine Learning Algorithms, that poses a disadvantage
as it mandatorily assumes that the features, in our project, words, are independent of each other.
This is a comparatively inefficient method of performing Sentiment Analysis on bulk text, for
official purposes, since sentences will not give the meaning they are supposed to convey, if each
word is considered a separate entity. Maximum Entropy Classifier overcomes this draw back by
limiting the assumptions it makes of the input data feed, which is what we use in the proposed
system.
Sentiment Analysis on Twitter Dataset using R Languageijtsrd
Sentiment Analysis involves determining the evaluative nature of a piece of text. A product review can express a positive, negative, or neutral sentiment or polarity . Automatically identifying sentiment expressed in text has a number of applications, including tracking sentiment towards Movie reviews and Automobile reviews improving customer relation models, detecting happiness and well being, and improving automatic dialogue systems. The evaluative intensity for both positive and negative terms changes in a negated context, and the amount of change varies from term to term. To adequately capture the impact of negation on individual terms, here proposed to empirically estimate the sentiment scores of terms in negated context from movie review and auto mobile review, and built two lexicons, one for terms in negated contexts and one for terms in affirmative non negated contexts. By using these Affirmative Context Lexicons and Negated Context Lexicons were able to significantly improve the performance of the overall sentiment analysis system on both tasks. This thesis have proposed a sentiment analysis system that detects the sentiment of corpus dataset using movie review and Automobile review as well as the sentiment of a term a word or a phrase within a message term level task using R language. B. Nagajothi | Dr. R. Jemima Priyadarsini "Sentiment Analysis on Twitter Dataset using R Language" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-3 | Issue-6 , October 2019, URL: https://www.ijtsrd.com/papers/ijtsrd28071.pdf Paper URL: https://www.ijtsrd.com/computer-science/data-miining/28071/sentiment-analysis-on-twitter-dataset-using-r-language/b-nagajothi
IRJET- Interpreting Public Sentiments Variation by using FB-LDA TechniqueIRJET Journal
This document discusses sentiment analysis techniques for classifying tweets based on their positive, negative, or neutral sentiment. It proposes two Latent Dirichlet Allocation (LDA) based models - Foreground and Background LDA (FB-LDA) and Reason Candidate and Background LDA (RCB-LDA) - to analyze sentiment variation in tweets. FB-LDA can filter background topics and extract foreground topics to identify possible explanations for sentiment changes. RCB-LDA can rank reason candidates expressed in tweets to provide sentence-level sentiment explanations. The proposed techniques are intended to classify tweets and evaluate public sentiment variations by extracting possible reasons for those variations.
This document discusses using Twitter data for sentiment analysis and influence tracking. It describes how Twitter data was collected using its APIs and preprocessed by removing links, usernames and stopwords. N-grams and part-of-speech tags were then extracted as features from the tweets. The tweets were classified into positive, negative, neutral or irrelevant categories. Sentiment analysis was performed at the entity level to determine sentiment towards specific topics mentioned in tweets, like products. Influence was tracked using algorithms that rank users based on retweets, followers and mentions.
Sentiment analysis of comments in social media IJECEIAES
Social media platforms are witnessing a significant growth in both size and purpose. One specific aspect of social media platforms is sentiment analysis, by which insights into the emotions and feelings of a person can be inferred from their posted text. Research related to sentiment analysis is acquiring substantial interest as it is a promising filed that can improve user experience and provide countless personalized services. Twitter is one of the most popular social media platforms, it has users from different regions with a variety of cultures and languages. It can thus provide valuable information for a diverse and large amount of data to be used to improve decision making. In this paper, the sentiment orientation of the textual features and emoji-based components is studied targeting “Tweets” and comments posted in Arabic on Twitter, during the 2018 world cup event. This study also measures the significance of analyzing texts including or excluding emojis. The data is obtained from thousands of extracted tweets, to find the results of sentiment analysis for texts and emojis separately. Results show that emojis support the sentiment orientation of the texts and those texts or emojis cannot separately provide reliable information as they complement each other to give the intended meaning.
IRJET- Improved Real-Time Twitter Sentiment Analysis using ML & Word2VecIRJET Journal
This document discusses improving real-time Twitter sentiment analysis using machine learning and Word2Vec. It begins by introducing sentiment analysis and its importance for product analysis and business. Next, it describes extracting Twitter data using APIs, preprocessing it, and applying machine learning algorithms like Naive Bayes, logistic regression, and decision trees to classify tweets as expressing positive, negative or neutral sentiment. It also reviews literature on using techniques like linguistic analysis and ensemble models to improve sentiment analysis from social media sources.
This document summarizes a research paper that proposes using a logistic regression classifier trained with stochastic gradient descent to predict Twitter users' personalities from their tweets. It begins with an abstract of the paper and an introduction on personality prediction from social media. It then provides more detail on the anatomy of the research, including defining personality prediction from Twitter, its applications, and the general process of using machine learning for the task. Next, it reviews several previous studies on personality prediction from Twitter and social networks, noting their approaches, findings and limitations. It identifies remaining research gaps, such as the need for improved linguistic analysis of tweets and more robust/scalable predictive models. Finally, it proposes using a logistic regression classifier as the personality prediction model to address
A RELIABLE ARTIFICIAL INTELLIGENCE MODEL FOR FALSE NEWS DETECTION MADE BY PUB...caijjournal
The quick access to information on social media networks as well as its exponential rise also made it
difficult to distinguish among fake information or real information. The fast dissemination by way of
sharing has enhanced its falsification exponentially. It is also important for the credibility of social media
networks to avoid the spread of fake information. So it is emerging research challenge to automatically
check for misstatement of information through its source, content, or publisher and prevent the
unauthenticated sources from spreading rumours. This paper demonstrates an artificial intelligence based
approach for the identification of the false statements made by social network entities. Two variants of
Deep neural networks are being applied to evalues datasets and analyse for fake news presence. The
implementation setup produced maximum extent 99% classification accuracy, when dataset is tested for
binary (true or false) labeling with multiple epochs.
A MODEL BASED ON SENTIMENTS ANALYSIS FOR STOCK EXCHANGE PREDICTION - CASE STU...csandit
Predicting the behavior of shares in the stock market is a complex problem, that involves variables not always known and can undergo various influences, from the collective emotion to high-profile news. Such volatility, can represent considerable financial losses for investors. In order to anticipate such changes in the market, it has been proposed various mechanisms to try to predict the behavior of an asset in the stock market, based on previously existing information.
Such mechanisms include statistical data only, without considering the collective feeling. This article, is going to use natural language processing algorithms (LPN) to determine the collective mood on assets and later with the help of the SVM algorithm to extract patterns in an attempt to predict the active behavior. Nevertheless it is important to note that such approach is not intended to be the main factor in the decision making process, but rather an aid tool, which combined with other information, can provide higher accuracy for the solution of this problem
This document reviews research on predicting personality from Twitter users' tweets using machine learning algorithms. It discusses how tweets have attracted research interest from diverse fields. Different techniques have been used to predict personality from tweets, but there are still shortcomings to address. The aim is to consider the current state of this research area and explore personality prediction from tweets by reviewing past literature and discussing approaches to issues researchers face. It provides an overview of machine learning methods used for personality prediction from tweets, including data collection, preprocessing, model training and evaluation.
Design of Automated Sentiment or Opinion Discovery System to Enhance Its Perf...idescitation
In today’s social networking era, if one has to make
decision about any product, service or individual performance,
the availability of various comments, suggestions, ratings,
and feedbacks are abundant. The required decision support
data can be collected through different sources of Medias like
newspapers, blogs, and discussion forums and from internet
too. So surely, it leads to the selection of best product, service
or individual if it is analyzed efficiently. In leading and
competitive world, this is huge and practical need of industries,
organizations to empower their qualities. In the recent years,
the significant study is done in the field of sentiment analysis.
However, the earlier work focused the implementation and
evaluation of individual sub technique of sentiment analysis.
Though these implementations produces significant results
of sentiment or opinion analysis, the trust of decision makers
is still in dangling to accept the results of such analysis. In
this paper, initially, we have been described the brief review
about the sentiment or opinion analysis system. Then the
details are provided about the design and about how to build
an automated opinion discovery system to enhance
performance of sentiment or opinion analysis based on feature
extraction sentiment analysis sub technique, natural language
processing and data mining techniques in an integrated way
Extraction of Emoticons with Sentimental Barvivatechijri
The latest generation of emoticons which are called as emojis that is largely being used in mobile
communications as well as in social media. In past few years, more than ten billion emojis were used on Twitter.
Emojis which are known as the Unicode graphic symbols, which are basically used as shorthand to express the
concepts and ideas of the people. For smaller number of well-known emoticons, their meanings or sentiments
are well known but there are thousands of emojis so extracting their sentiments is difficult. The Emoji Sentiment
Ranking method which is used to evaluate a sentiment mapping of emojis by using sentiment polarity such as
negative, neutral, or positive. The sentimental classification of tweets with and without emoticons are very much
different.Finally, the method also gives representation of sentiments and a better visualization in the form of a
sentimental Bar.
A Hybrid Approach for Supervised Twitter Sentiment Classification ....................................................1
K. Revathy and Dr. B. Sathiyabhama
A Survey of Dynamic Duty Cycle Scheduling Scheme at Media Access Control Layer for Energy
Conservation .....................................................................................................................................1
Prof. M. V. Nimbalkar and Sampada Khandare
A Survey on Privacy Preserving Data Mining Techniques ....................................................................1
A. K. Ilavarasi, B. Sathiyabhama and S. Poorani
An Ontology Based System for Predicting Disease using SWRL Rules ...................................................1
Mythili Thirugnanam, Tamizharasi Thirugnanam and R. Mangayarkarasi
Performance Evaluation of Web Services in C#, JAVA, and PHP ..........................................................1
Dr. S. Sagayaraj and M. Santhosh Kumar
Semi-Automated Polyhouse Cultivation Using LabVIEW......................................................................1
Prathiba Jonnala and Sivaji Satrasupalli
Performance of Biometric Palm Print Personal Identification Security System Using Ordinal Measures 1
V. K. Narendira Kumar and Dr. B. Srinivasan
MIMO System for Next Generation Wireless Communication..............................................................1
Sharif, Mohammad Emdadul Haq and Md. Arif Rana
A Baseline Based Deep Learning Approach of Live Tweetsijtsrd
In this scenario social media plays a vital role in influencing the life of people. Twitter , Facebook, Instagram etc are the major social media platforms . They act as a platform for users to raise their opinions on things and events around them. Twitter is one such micro blogging site that allows the user to tweet 6000 tweets per day each of 280 characters long. Data analyst rely on this data to reach conclusion on the events happening around and also to rate a product. But due to massive volume of reviews the analysts find it difficult to go through them and reach at conclusions. In order to solve this problem we adopt the method of sentiment analysis. Sentiment analysis is an approach to classify the sentiment of user reviews, documents etc in terms of positive good , negative bad , neutral surprise . I suggest an enhanced twitter sentiment analysis that retrieves data based on a baseline in a particular pre defined time span and performs sentiment analysis using Textblob . This scheme differs from the traditional and existing one which performs sentiment analysis on pre saved data by performing sentiment analysis on real time data fetched via Twitter API . Thereby providing a much recent and relevant conclusion. Anjana Jimmington ""A Baseline Based Deep Learning Approach of Live Tweets"" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-3 | Issue-4 , June 2019, URL: https://www.ijtsrd.com/papers/ijtsrd23918.pdf
Paper URL: https://www.ijtsrd.com/computer-science/other/23918/a-baseline-based-deep-learning-approach-of-live-tweets/anjana-jimmington
IRJET - Unauthorized Terror Attack Tracking System using Web Usage MiningIRJET Journal
This document describes a system for tracking unauthorized terror attacks using web usage mining and sentiment analysis of social media data. It discusses collecting tweets on a topic using keywords, preprocessing the tweets, analyzing the sentiment of each tweet using TextBlob and VADER sentiment analysis tools, and visualizing the results through graphs and tables. The system aims to help detect terror-related activities by analyzing opinions and sentiments expressed on social media platforms like Twitter.
Twitter, has fast emerged as one of the most powerful social media sites which can
sway opinions. Sentiment or opinion analysis has of late emerged one of the most
researched and talked about subject in Natural Language Processing (NLP), thanks
mainly to sites like Twitter. In the past, sentiment analysis models using Twitter data have
been built to predict sales performance, rank products and merchants, public opinion
polls, predict election results, political standpoints, predict box-office revenues for movies
and even predict the stock market. This study proposes a general frame in R programming
language to act as a gateway for the analysis of the tweets that portray emotions in a
short and concentrated format. The target tweets include brief emotion descriptions and
words that are not used with a proper format or grammatical structure. Majority of the
work constituted in Turkish includes the data scope and the aim of preparing a data-set.
There is no concrete and usable work done on Turkish Tweet sentiment analysis as a
software client/web application. This study is a starting point on building up the next
steps. The aim is to compare five different common machine learning methods (support
vector machines, random forests, boosting, maximum entropy, and artificial neural
networks) to classify Twitters sentiments
Current trends of opinion mining and sentiment analysis in social networkseSAT Publishing House
IJRET : International Journal of Research in Engineering and Technology is an international peer reviewed, online journal published by eSAT Publishing House for the enhancement of research in various disciplines of Engineering and Technology. The aim and scope of the journal is to provide an academic medium and an important reference for the advancement and dissemination of research results that support high-level learning, teaching and research in the fields of Engineering and Technology. We bring together Scientists, Academician, Field Engineers, Scholars and Students of related fields of Engineering and Technology
IRJET - Election Result Prediction using Sentiment AnalysisIRJET Journal
This document proposes a method to predict election results using sentiment analysis of social media data. It involves collecting data from Twitter, Facebook, and Instagram using their APIs. The data will then be preprocessed by removing special characters and URLs. Popular machine learning algorithms like Naive Bayes and SVM will be trained on the preprocessed data to classify tweets as positive, negative, or neutral sentiment toward political parties. The classified tweets will then be analyzed to predict the outcome of elections.
Sentimental Emotion Analysis using Python and Machine LearningYogeshIJTSRD
Sentiment analysis is used in opinion mining. It helps businesses understand the customers’ reviews with a particular product by analyzing their emotional from the product reviews they post, the online recommendations they make, their survey responses and other forms of social media text. Businesses can get feedback on how happy or sad the customer is, and use this insight to gain a competitive edge. In this article, we explore how to conduct sentiment analysis on a piece of text using some machine learning techniques. Python happens to be one of the best programming language, when it comes to machine learning as it is easy to learn, is open source, and is effective in catering to machine learning requirements like processing big datasets and performing mathematical computations. Natural Language ToolKit NLTK is one of the popular packages in Python that can use for in sentiment analysis. Mohit Chaudhari "Sentimental Emotion Analysis using Python and Machine Learning" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-5 | Issue-4 , June 2021, URL: https://www.ijtsrd.compapers/ijtsrd41198.pdf Paper URL: https://www.ijtsrd.comengineering/computer-engineering/41198/sentimental-emotion-analysis-using-python-and-machine-learning/mohit-chaudhari
This document summarizes 4 papers on sentiment analysis of tweets. It discusses how the papers preprocess tweets by removing URLs, usernames, repeated characters, and applying part-of-speech tagging. It also discusses how the papers classify sentiment at the document, sentence, and entity levels. Classification algorithms discussed include Naive Bayes, SVM, maximum entropy. The papers achieve accuracies between 67-80% for binary and multi-class sentiment classification of tweets.
Determine the sentiment of sentence that is positive or negative based on the presence of part of
speech tag, the emoticons present in the sentences. For this research we use the most popular microblogging sit
twitter for sentiment orientation. In this paper we want to extract tweets form the twitter related to the product
like mobile phones, home appliances, vehicle etc. After retrieving tweets we perform some preprocessing on it
like remove retweets, remove tweets containing few words with minimum threshold of length five, remove tweets
containing only urls. After this the remaining tweets are pre-processed like that transform all letters of the
tweets to the lower case then remove punctuation from the tweets because it reduces the accuracy of result.
After this remove extra white spaces from the tweets, then we apply a pos tagger to tag each word. The tuple
after the applying above steps contain (word, pos tag, English-word, stop-word). We are interested in only
tweets that contain opinion and eliminate the remaining non-opinion tweets from the data set. For this we use
the Naïve Bays classification algorithm. After this we use short text classification on tweets i.e., the word having
different meaning in different domain. In order to solve this problem we use two different feature selection
algorithms the mutual information (MI) and the X2 feature selection. At final stage predicting the orientation of
an opinion sentence that is positive or negative as we mentioned above. For this we use two model like unigram
model and opinion miner.
Humans communication is generally under the control of emotions and full of opinions. Emotions and their opinions plays an important role in thinking process of mind, influences the human actions too. Sentiment analysis is one of the ways to explore user’s opinion made on any social media and networking site for various commercial applications in number of fields. This paper takes into account the basis requirements of opinion mining to explore the present techniques used to developed an full fledge system. Is highlights the opportunities or deployment and research of such systems. The available tools used for building such applications have even presented with their merits and limitations.
This document summarizes a research project on sentiment analysis of tweets about news. The researchers collected tweets related to news articles from various sources and analyzed the sentiment of the tweets to determine the overall public sentiment toward that news. They first preprocessed the tweet text through tokenization, removed stopwords, and calculated term frequencies. Next, they analyzed term co-occurrences to understand context. They also created visualizations of frequent terms. Finally, they used a naive Bayes classifier trained on labeled data to classify tweets in real-time as positive, negative, or neutral sentiment toward the news. The system aimed to provide a score indicating overall public sentiment toward each news article based on related tweets.
With the rise of social networking epoch, there has been a surge of user generated content. Micro blogging sites have millions of people sharing their thoughts daily because of its characteristic short and simple manner of expression. We propose and investigate a paradigm to mine the sentiment from a popular real-time micro blogging service, Twitter, where users post real time reactions to and opinions about “everything”. In this paper, we expound a hybrid approach using both corpus based and dictionary based methods to determine the semantic orientation of the opinion words in tweets. A case study is presented to illustrate the use and effectiveness of the proposed system.
This document discusses using Twitter data for sentiment analysis and influence tracking. It describes how Twitter data was collected using its APIs and preprocessed by removing links, usernames and stopwords. N-grams and part-of-speech tags were then extracted as features from the tweets. The tweets were classified into positive, negative, neutral or irrelevant categories. Sentiment analysis was performed at the entity level to determine sentiment towards specific topics mentioned in tweets, like products. Influence was tracked using algorithms that rank users based on retweets, followers and mentions.
Sentiment analysis of comments in social media IJECEIAES
Social media platforms are witnessing a significant growth in both size and purpose. One specific aspect of social media platforms is sentiment analysis, by which insights into the emotions and feelings of a person can be inferred from their posted text. Research related to sentiment analysis is acquiring substantial interest as it is a promising filed that can improve user experience and provide countless personalized services. Twitter is one of the most popular social media platforms, it has users from different regions with a variety of cultures and languages. It can thus provide valuable information for a diverse and large amount of data to be used to improve decision making. In this paper, the sentiment orientation of the textual features and emoji-based components is studied targeting “Tweets” and comments posted in Arabic on Twitter, during the 2018 world cup event. This study also measures the significance of analyzing texts including or excluding emojis. The data is obtained from thousands of extracted tweets, to find the results of sentiment analysis for texts and emojis separately. Results show that emojis support the sentiment orientation of the texts and those texts or emojis cannot separately provide reliable information as they complement each other to give the intended meaning.
IRJET- Improved Real-Time Twitter Sentiment Analysis using ML & Word2VecIRJET Journal
This document discusses improving real-time Twitter sentiment analysis using machine learning and Word2Vec. It begins by introducing sentiment analysis and its importance for product analysis and business. Next, it describes extracting Twitter data using APIs, preprocessing it, and applying machine learning algorithms like Naive Bayes, logistic regression, and decision trees to classify tweets as expressing positive, negative or neutral sentiment. It also reviews literature on using techniques like linguistic analysis and ensemble models to improve sentiment analysis from social media sources.
This document summarizes a research paper that proposes using a logistic regression classifier trained with stochastic gradient descent to predict Twitter users' personalities from their tweets. It begins with an abstract of the paper and an introduction on personality prediction from social media. It then provides more detail on the anatomy of the research, including defining personality prediction from Twitter, its applications, and the general process of using machine learning for the task. Next, it reviews several previous studies on personality prediction from Twitter and social networks, noting their approaches, findings and limitations. It identifies remaining research gaps, such as the need for improved linguistic analysis of tweets and more robust/scalable predictive models. Finally, it proposes using a logistic regression classifier as the personality prediction model to address
A RELIABLE ARTIFICIAL INTELLIGENCE MODEL FOR FALSE NEWS DETECTION MADE BY PUB...caijjournal
The quick access to information on social media networks as well as its exponential rise also made it
difficult to distinguish among fake information or real information. The fast dissemination by way of
sharing has enhanced its falsification exponentially. It is also important for the credibility of social media
networks to avoid the spread of fake information. So it is emerging research challenge to automatically
check for misstatement of information through its source, content, or publisher and prevent the
unauthenticated sources from spreading rumours. This paper demonstrates an artificial intelligence based
approach for the identification of the false statements made by social network entities. Two variants of
Deep neural networks are being applied to evalues datasets and analyse for fake news presence. The
implementation setup produced maximum extent 99% classification accuracy, when dataset is tested for
binary (true or false) labeling with multiple epochs.
A MODEL BASED ON SENTIMENTS ANALYSIS FOR STOCK EXCHANGE PREDICTION - CASE STU...csandit
Predicting the behavior of shares in the stock market is a complex problem, that involves variables not always known and can undergo various influences, from the collective emotion to high-profile news. Such volatility, can represent considerable financial losses for investors. In order to anticipate such changes in the market, it has been proposed various mechanisms to try to predict the behavior of an asset in the stock market, based on previously existing information.
Such mechanisms include statistical data only, without considering the collective feeling. This article, is going to use natural language processing algorithms (LPN) to determine the collective mood on assets and later with the help of the SVM algorithm to extract patterns in an attempt to predict the active behavior. Nevertheless it is important to note that such approach is not intended to be the main factor in the decision making process, but rather an aid tool, which combined with other information, can provide higher accuracy for the solution of this problem
This document reviews research on predicting personality from Twitter users' tweets using machine learning algorithms. It discusses how tweets have attracted research interest from diverse fields. Different techniques have been used to predict personality from tweets, but there are still shortcomings to address. The aim is to consider the current state of this research area and explore personality prediction from tweets by reviewing past literature and discussing approaches to issues researchers face. It provides an overview of machine learning methods used for personality prediction from tweets, including data collection, preprocessing, model training and evaluation.
Design of Automated Sentiment or Opinion Discovery System to Enhance Its Perf...idescitation
In today’s social networking era, if one has to make
decision about any product, service or individual performance,
the availability of various comments, suggestions, ratings,
and feedbacks are abundant. The required decision support
data can be collected through different sources of Medias like
newspapers, blogs, and discussion forums and from internet
too. So surely, it leads to the selection of best product, service
or individual if it is analyzed efficiently. In leading and
competitive world, this is huge and practical need of industries,
organizations to empower their qualities. In the recent years,
the significant study is done in the field of sentiment analysis.
However, the earlier work focused the implementation and
evaluation of individual sub technique of sentiment analysis.
Though these implementations produces significant results
of sentiment or opinion analysis, the trust of decision makers
is still in dangling to accept the results of such analysis. In
this paper, initially, we have been described the brief review
about the sentiment or opinion analysis system. Then the
details are provided about the design and about how to build
an automated opinion discovery system to enhance
performance of sentiment or opinion analysis based on feature
extraction sentiment analysis sub technique, natural language
processing and data mining techniques in an integrated way
Extraction of Emoticons with Sentimental Barvivatechijri
The latest generation of emoticons which are called as emojis that is largely being used in mobile
communications as well as in social media. In past few years, more than ten billion emojis were used on Twitter.
Emojis which are known as the Unicode graphic symbols, which are basically used as shorthand to express the
concepts and ideas of the people. For smaller number of well-known emoticons, their meanings or sentiments
are well known but there are thousands of emojis so extracting their sentiments is difficult. The Emoji Sentiment
Ranking method which is used to evaluate a sentiment mapping of emojis by using sentiment polarity such as
negative, neutral, or positive. The sentimental classification of tweets with and without emoticons are very much
different.Finally, the method also gives representation of sentiments and a better visualization in the form of a
sentimental Bar.
A Hybrid Approach for Supervised Twitter Sentiment Classification ....................................................1
K. Revathy and Dr. B. Sathiyabhama
A Survey of Dynamic Duty Cycle Scheduling Scheme at Media Access Control Layer for Energy
Conservation .....................................................................................................................................1
Prof. M. V. Nimbalkar and Sampada Khandare
A Survey on Privacy Preserving Data Mining Techniques ....................................................................1
A. K. Ilavarasi, B. Sathiyabhama and S. Poorani
An Ontology Based System for Predicting Disease using SWRL Rules ...................................................1
Mythili Thirugnanam, Tamizharasi Thirugnanam and R. Mangayarkarasi
Performance Evaluation of Web Services in C#, JAVA, and PHP ..........................................................1
Dr. S. Sagayaraj and M. Santhosh Kumar
Semi-Automated Polyhouse Cultivation Using LabVIEW......................................................................1
Prathiba Jonnala and Sivaji Satrasupalli
Performance of Biometric Palm Print Personal Identification Security System Using Ordinal Measures 1
V. K. Narendira Kumar and Dr. B. Srinivasan
MIMO System for Next Generation Wireless Communication..............................................................1
Sharif, Mohammad Emdadul Haq and Md. Arif Rana
A Baseline Based Deep Learning Approach of Live Tweetsijtsrd
In this scenario social media plays a vital role in influencing the life of people. Twitter , Facebook, Instagram etc are the major social media platforms . They act as a platform for users to raise their opinions on things and events around them. Twitter is one such micro blogging site that allows the user to tweet 6000 tweets per day each of 280 characters long. Data analyst rely on this data to reach conclusion on the events happening around and also to rate a product. But due to massive volume of reviews the analysts find it difficult to go through them and reach at conclusions. In order to solve this problem we adopt the method of sentiment analysis. Sentiment analysis is an approach to classify the sentiment of user reviews, documents etc in terms of positive good , negative bad , neutral surprise . I suggest an enhanced twitter sentiment analysis that retrieves data based on a baseline in a particular pre defined time span and performs sentiment analysis using Textblob . This scheme differs from the traditional and existing one which performs sentiment analysis on pre saved data by performing sentiment analysis on real time data fetched via Twitter API . Thereby providing a much recent and relevant conclusion. Anjana Jimmington ""A Baseline Based Deep Learning Approach of Live Tweets"" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-3 | Issue-4 , June 2019, URL: https://www.ijtsrd.com/papers/ijtsrd23918.pdf
Paper URL: https://www.ijtsrd.com/computer-science/other/23918/a-baseline-based-deep-learning-approach-of-live-tweets/anjana-jimmington
IRJET - Unauthorized Terror Attack Tracking System using Web Usage MiningIRJET Journal
This document describes a system for tracking unauthorized terror attacks using web usage mining and sentiment analysis of social media data. It discusses collecting tweets on a topic using keywords, preprocessing the tweets, analyzing the sentiment of each tweet using TextBlob and VADER sentiment analysis tools, and visualizing the results through graphs and tables. The system aims to help detect terror-related activities by analyzing opinions and sentiments expressed on social media platforms like Twitter.
Twitter, has fast emerged as one of the most powerful social media sites which can
sway opinions. Sentiment or opinion analysis has of late emerged one of the most
researched and talked about subject in Natural Language Processing (NLP), thanks
mainly to sites like Twitter. In the past, sentiment analysis models using Twitter data have
been built to predict sales performance, rank products and merchants, public opinion
polls, predict election results, political standpoints, predict box-office revenues for movies
and even predict the stock market. This study proposes a general frame in R programming
language to act as a gateway for the analysis of the tweets that portray emotions in a
short and concentrated format. The target tweets include brief emotion descriptions and
words that are not used with a proper format or grammatical structure. Majority of the
work constituted in Turkish includes the data scope and the aim of preparing a data-set.
There is no concrete and usable work done on Turkish Tweet sentiment analysis as a
software client/web application. This study is a starting point on building up the next
steps. The aim is to compare five different common machine learning methods (support
vector machines, random forests, boosting, maximum entropy, and artificial neural
networks) to classify Twitters sentiments
Current trends of opinion mining and sentiment analysis in social networkseSAT Publishing House
IJRET : International Journal of Research in Engineering and Technology is an international peer reviewed, online journal published by eSAT Publishing House for the enhancement of research in various disciplines of Engineering and Technology. The aim and scope of the journal is to provide an academic medium and an important reference for the advancement and dissemination of research results that support high-level learning, teaching and research in the fields of Engineering and Technology. We bring together Scientists, Academician, Field Engineers, Scholars and Students of related fields of Engineering and Technology
IRJET - Election Result Prediction using Sentiment AnalysisIRJET Journal
This document proposes a method to predict election results using sentiment analysis of social media data. It involves collecting data from Twitter, Facebook, and Instagram using their APIs. The data will then be preprocessed by removing special characters and URLs. Popular machine learning algorithms like Naive Bayes and SVM will be trained on the preprocessed data to classify tweets as positive, negative, or neutral sentiment toward political parties. The classified tweets will then be analyzed to predict the outcome of elections.
Sentimental Emotion Analysis using Python and Machine LearningYogeshIJTSRD
Sentiment analysis is used in opinion mining. It helps businesses understand the customers’ reviews with a particular product by analyzing their emotional from the product reviews they post, the online recommendations they make, their survey responses and other forms of social media text. Businesses can get feedback on how happy or sad the customer is, and use this insight to gain a competitive edge. In this article, we explore how to conduct sentiment analysis on a piece of text using some machine learning techniques. Python happens to be one of the best programming language, when it comes to machine learning as it is easy to learn, is open source, and is effective in catering to machine learning requirements like processing big datasets and performing mathematical computations. Natural Language ToolKit NLTK is one of the popular packages in Python that can use for in sentiment analysis. Mohit Chaudhari "Sentimental Emotion Analysis using Python and Machine Learning" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-5 | Issue-4 , June 2021, URL: https://www.ijtsrd.compapers/ijtsrd41198.pdf Paper URL: https://www.ijtsrd.comengineering/computer-engineering/41198/sentimental-emotion-analysis-using-python-and-machine-learning/mohit-chaudhari
This document summarizes 4 papers on sentiment analysis of tweets. It discusses how the papers preprocess tweets by removing URLs, usernames, repeated characters, and applying part-of-speech tagging. It also discusses how the papers classify sentiment at the document, sentence, and entity levels. Classification algorithms discussed include Naive Bayes, SVM, maximum entropy. The papers achieve accuracies between 67-80% for binary and multi-class sentiment classification of tweets.
Determine the sentiment of sentence that is positive or negative based on the presence of part of
speech tag, the emoticons present in the sentences. For this research we use the most popular microblogging sit
twitter for sentiment orientation. In this paper we want to extract tweets form the twitter related to the product
like mobile phones, home appliances, vehicle etc. After retrieving tweets we perform some preprocessing on it
like remove retweets, remove tweets containing few words with minimum threshold of length five, remove tweets
containing only urls. After this the remaining tweets are pre-processed like that transform all letters of the
tweets to the lower case then remove punctuation from the tweets because it reduces the accuracy of result.
After this remove extra white spaces from the tweets, then we apply a pos tagger to tag each word. The tuple
after the applying above steps contain (word, pos tag, English-word, stop-word). We are interested in only
tweets that contain opinion and eliminate the remaining non-opinion tweets from the data set. For this we use
the Naïve Bays classification algorithm. After this we use short text classification on tweets i.e., the word having
different meaning in different domain. In order to solve this problem we use two different feature selection
algorithms the mutual information (MI) and the X2 feature selection. At final stage predicting the orientation of
an opinion sentence that is positive or negative as we mentioned above. For this we use two model like unigram
model and opinion miner.
Humans communication is generally under the control of emotions and full of opinions. Emotions and their opinions plays an important role in thinking process of mind, influences the human actions too. Sentiment analysis is one of the ways to explore user’s opinion made on any social media and networking site for various commercial applications in number of fields. This paper takes into account the basis requirements of opinion mining to explore the present techniques used to developed an full fledge system. Is highlights the opportunities or deployment and research of such systems. The available tools used for building such applications have even presented with their merits and limitations.
This document summarizes a research project on sentiment analysis of tweets about news. The researchers collected tweets related to news articles from various sources and analyzed the sentiment of the tweets to determine the overall public sentiment toward that news. They first preprocessed the tweet text through tokenization, removed stopwords, and calculated term frequencies. Next, they analyzed term co-occurrences to understand context. They also created visualizations of frequent terms. Finally, they used a naive Bayes classifier trained on labeled data to classify tweets in real-time as positive, negative, or neutral sentiment toward the news. The system aimed to provide a score indicating overall public sentiment toward each news article based on related tweets.
With the rise of social networking epoch, there has been a surge of user generated content. Micro blogging sites have millions of people sharing their thoughts daily because of its characteristic short and simple manner of expression. We propose and investigate a paradigm to mine the sentiment from a popular real-time micro blogging service, Twitter, where users post real time reactions to and opinions about “everything”. In this paper, we expound a hybrid approach using both corpus based and dictionary based methods to determine the semantic orientation of the opinion words in tweets. A case study is presented to illustrate the use and effectiveness of the proposed system.
This document summarizes a research project on sentiment analysis of tweets about news. The researchers collected tweets related to news articles from various sources and analyzed the sentiment of the tweets to determine the overall public sentiment toward that news. They first preprocessed the tweet text through tokenization, removed stopwords, and calculated term frequencies. Next, they analyzed term co-occurrences to understand context. They also created visualizations of frequent terms. Finally, they used a naive Bayes classifier trained on labeled data to classify tweets in real-time as positive, negative, or neutral sentiment toward the news. The system aimed to provide a score indicating overall public sentiment toward each news article based on related tweets.
This document discusses using sentiment analysis on social media data to extract useful information for businesses and customers. It proposes a methodology that uses three modules: an extractor to access social media APIs and obtain raw data, a preprocessor to clean the raw data, and an analyzer using naive Bayes classification to categorize the preprocessed data into positive, negative, or neutral sentiments. The categorized sentiment data can then be used by businesses for decision making and by customers to inform their purchasing decisions. The methodology is demonstrated by implementing sentiment analysis on tweets from Twitter.
This document provides a high-level and low-level description of a sentiment analysis system. At the high level, it collects text data, splits it into sentences, assigns polarity, checks for repeated words, and extracts sentiment. The low-level description details how it collects data from Facebook using APIs, processes the data by tagging parts of speech, analyzes polarity vs neutral sets, lists features, and builds a classifier using naive Bayes and dependencies between n-grams and parts of speech. The system aims to analyze sentiment from social media texts at both the document and sentence level.
A MODEL BASED ON SENTIMENTS ANALYSIS FOR STOCK EXCHANGE PREDICTION - CASE STU...cscpconf
Predicting the behavior of shares in the stock market is a complex problem, that involves variables not always known and can undergo various influences, from the collective emotion to the high-profile news. Such volatility can represent considerable financial losses for investors. In order to anticipate such changes in the market, it has been proposed various mechanisms to try to predict the behavior of an asset in the stock market, based on previously existing information. Such mechanisms include statistical data only, without considering the collective feeling. This article is going to use natural language processing algorithms (LPN) to determine the collective mood on assets and later with the help of the SVM algorithm to extract patterns in an
attempt to predict the active behavior. Nevertheless it is important to note that such approach is not intended to be the main factor in the decision making process, but rather an aid tool, which combined with other information, can provide higher accuracy for the solution of this problem.
THE ANALYSIS FOR CUSTOMER REVIEWS THROUGH TWEETS, BASED ON DEEP LEARNINGIRJET Journal
The document describes a study that analyzes customer reviews on Twitter about hotels using deep learning techniques. Twitter data is collected using Python's Tweepy library and preprocessed by removing noise like retweets, URLs and hashtags. The data is then split using scikit-learn into training, validation and testing sets. Tokenization is performed to convert text to vectors and sentiment analysis is done using techniques like Bi-Sense Emoji Embedding (BSEE), Random Forest (RF) and Support Vector Machine (SVM). The performance of BSEE is compared based on accuracy, recall, precision and time taken and is found to provide better results.
Social Sentiment Analysis and Its Use in Communication Campaignssvladovic
This document discusses social sentiment analysis and how it can be used in communication campaigns. It defines social sentiment analysis and evaluates three free online tools for sentiment analysis: Topsy, Sentiment140, and Social Mention. It then provides an example analysis of sentiments toward Super Bowl 2015 commercials using these tools. The analysis found the tools can consistently track message numbers and detect prevailing sentiments, making them useful for marketers despite challenges in social media analysis.
This document discusses classifying toxic comments on online platforms using machine learning algorithms. It begins with an abstract discussing how analyzing user data on social media can help organizations understand public sentiment. It then discusses how machine learning classifiers can classify tweets as positive or negative. The introduction discusses how online comments can sometimes be abusive and spread negativity, so it is important for platforms to filter toxic comments. The existing system section outlines how sentiment analysis is commonly done using machine learning approaches on Twitter data. The proposed system would gather labeled tweet data, preprocess it, extract relevant features, and apply machine learning classifiers to categorize tweets as having high/moderate/low levels of positive or negative sentiment.
Sentiment Analysis in Hindi Language : A SurveyEditor IJMTER
With recent development in web technologies and mobile technologies, with increasing
user-generated content in Hindi on the internet is the motivation behind the sentiment analysis
Research that is growing up at a lightning speed. This information can prove to be very useful for
researchers, governments and organization to learn what’s on public mind, to make sound decisions.
Opinion Mining or Sentiment Analysis is a natural language processing task that mine information
from various text forms such as reviews, news, and blogs and classify them on the basis of their
polarity as positive, negative or neutral. But, from the last few years, enormous increase has been seen
in Hindi language on the Web. Research in opinion mining mostly carried out in English language
but it is very important to perform the opinion mining in Hindi language also as large amount
of information in Hindi is also available on the Web. This paper gives an overview of the work that
has been done Hindi language.
Expectations for Electronic Debate Platforms as a Function of Application DomainIJERA Editor
Electronic debate (or commenting) platforms are used with many types of online applications, as a way to
engage the users or to provide enhancements, e.g., based on some type of collaborative filtering [1], [2]. The
applications enhanced with such debate platforms range widely : news, products, sport, religion, politics, etc.
Therefore, the emerging question is whether it is possible to make one electronic debate mechanism good for all
applications, and whether the studies on the success of a debate mechanism in one domain do automatically
apply to other application domains. Here we compare two traditional application domains of electronic debate
platforms: product evaluation and commented news. We exploit the fact that most users are very familiar with
both types of such applications, and therefore surveys can be designed to gauge reliably subtle differences
between expectations and properties of these domains. Based on over 1000 responses to surveys described here,
we are able to report statistically significant differences between the user behavior and expectations in the
studied domains.
Expectations for Electronic Debate Platforms as a Function of Application DomainIJERA Editor
Electronic debate (or commenting) platforms are used with many types of online applications, as a way to engage the users or to provide enhancements, e.g., based on some type of collaborative filtering [1], [2]. The applications enhanced with such debate platforms range widely : news, products, sport, religion, politics, etc. Therefore, the emerging question is whether it is possible to make one electronic debate mechanism good for all applications, and whether the studies on the success of a debate mechanism in one domain do automatically apply to other application domains. Here we compare two traditional application domains of electronic debate platforms: product evaluation and commented news. We exploit the fact that most users are very familiar with both types of such applications, and therefore surveys can be designed to gauge reliably subtle differences between expectations and properties of these domains. Based on over 1000 responses to surveys described here, we are able to report statistically significant differences between the user behavior and expectations in the studied domains.
23 may 2015 monitoring & analyzing social media Mats Björe
Our approach to social media analytics focuses on analyzing social media data to better understand interactions and influence. We aim to identify, monitor, track, and measure influence and reach to build dynamic dossiers and provide decision support. Some key challenges include fragmented information lacking context, use of symbols and emoticons, similar messages from many sources, and constantly changing platforms and user preferences. We use proven software like Silobreaker to monitor social media at scale and provide analytical dashboards and tools to examine specific stakeholders, anomalies, timelines, exposure, locations, and networks.
POLITICAL OPINION ANALYSIS IN SOCIAL NETWORKS: CASE OF TWITTER AND FACEBOOK dannyijwest
The 21st century has been characterized by an increased attention to social networks. Nowadays, going 24
hours without getting in touch with them in some way has become difficult. Facebook and Twitter, these
social platforms are now part of everyday life. Thus, these social networks have become important sources
to be aware of frequently discussed topics or public opinions on a current issue. A lot of people write
messages about current events, give their opinion on any topic and discuss social issues more and more.
IRJET - Social Media Intelligence ToolsIRJET Journal
This document discusses social media intelligence tools and analyzing social media data. It begins with an abstract that introduces social media as a way for information exchange and discusses building a tool to find suspects who have stolen social media users' data. It then provides more details in the following sections:
1. An introduction to social media and the large amounts of data stored, requiring monitoring tools. Examples of commonly used tools are provided.
2. Sentiment analysis is discussed as a way to analyze emotions from social media data through preprocessing, feature extraction, and classification.
3. Applications like analyzing word of mouth are mentioned.
4. Previous literature on topics like usage of social media and its impacts is reviewed.
This document discusses techniques for classifying sentiments and mining opinions from text data. It begins with defining key terminology in opinion mining like opinion feature, sentiment, polarity, holder and time. It then discusses various data sources for opinion mining like blogs, reviews sites, datasets, microblogs and other text. It describes the granularity of opinion mining tasks at the document level, sentence level and feature level. Finally, it outlines approaches to opinion mining including supervised learning techniques like Naive Bayes, SVM and unsupervised learning techniques that use lexical resources without prior training. Evaluation metrics for sentiment classification systems like accuracy, precision, recall and F1 measure are also discussed.
The document discusses utilizing weight allocation in a term frequency-inverse document frequency (TF-IDF) environment to identify and remove noisy data from social media for improved customer segmentation and targeted advertising. Specifically, it aims to recognize keywords that can help cluster social media users based on demographics and behaviors while eliminating uninfluential data. The approach assigns higher weight to words that frequently appear in a document but rarely in the entire collection compared to common words.
Similar to Sentiment Analysis using Fuzzy logic (20)
End-to-end pipeline agility - Berlin Buzzwords 2024Lars Albertsson
We describe how we achieve high change agility in data engineering by eliminating the fear of breaking downstream data pipelines through end-to-end pipeline testing, and by using schema metaprogramming to safely eliminate boilerplate involved in changes that affect whole pipelines.
A quick poll on agility in changing pipelines from end to end indicated a huge span in capabilities. For the question "How long time does it take for all downstream pipelines to be adapted to an upstream change," the median response was 6 months, but some respondents could do it in less than a day. When quantitative data engineering differences between the best and worst are measured, the span is often 100x-1000x, sometimes even more.
A long time ago, we suffered at Spotify from fear of changing pipelines due to not knowing what the impact might be downstream. We made plans for a technical solution to test pipelines end-to-end to mitigate that fear, but the effort failed for cultural reasons. We eventually solved this challenge, but in a different context. In this presentation we will describe how we test full pipelines effectively by manipulating workflow orchestration, which enables us to make changes in pipelines without fear of breaking downstream.
Making schema changes that affect many jobs also involves a lot of toil and boilerplate. Using schema-on-read mitigates some of it, but has drawbacks since it makes it more difficult to detect errors early. We will describe how we have rejected this tradeoff by applying schema metaprogramming, eliminating boilerplate but keeping the protection of static typing, thereby further improving agility to quickly modify data pipelines without fear.
ViewShift: Hassle-free Dynamic Policy Enforcement for Every Data LakeWalaa Eldin Moustafa
Dynamic policy enforcement is becoming an increasingly important topic in today’s world where data privacy and compliance is a top priority for companies, individuals, and regulators alike. In these slides, we discuss how LinkedIn implements a powerful dynamic policy enforcement engine, called ViewShift, and integrates it within its data lake. We show the query engine architecture and how catalog implementations can automatically route table resolutions to compliance-enforcing SQL views. Such views have a set of very interesting properties: (1) They are auto-generated from declarative data annotations. (2) They respect user-level consent and preferences (3) They are context-aware, encoding a different set of transformations for different use cases (4) They are portable; while the SQL logic is only implemented in one SQL dialect, it is accessible in all engines.
#SQL #Views #Privacy #Compliance #DataLake
Global Situational Awareness of A.I. and where its headedvikram sood
You can see the future first in San Francisco.
Over the past year, the talk of the town has shifted from $10 billion compute clusters to $100 billion clusters to trillion-dollar clusters. Every six months another zero is added to the boardroom plans. Behind the scenes, there’s a fierce scramble to secure every power contract still available for the rest of the decade, every voltage transformer that can possibly be procured. American big business is gearing up to pour trillions of dollars into a long-unseen mobilization of American industrial might. By the end of the decade, American electricity production will have grown tens of percent; from the shale fields of Pennsylvania to the solar farms of Nevada, hundreds of millions of GPUs will hum.
The AGI race has begun. We are building machines that can think and reason. By 2025/26, these machines will outpace college graduates. By the end of the decade, they will be smarter than you or I; we will have superintelligence, in the true sense of the word. Along the way, national security forces not seen in half a century will be un-leashed, and before long, The Project will be on. If we’re lucky, we’ll be in an all-out race with the CCP; if we’re unlucky, an all-out war.
Everyone is now talking about AI, but few have the faintest glimmer of what is about to hit them. Nvidia analysts still think 2024 might be close to the peak. Mainstream pundits are stuck on the wilful blindness of “it’s just predicting the next word”. They see only hype and business-as-usual; at most they entertain another internet-scale technological change.
Before long, the world will wake up. But right now, there are perhaps a few hundred people, most of them in San Francisco and the AI labs, that have situational awareness. Through whatever peculiar forces of fate, I have found myself amongst them. A few years ago, these people were derided as crazy—but they trusted the trendlines, which allowed them to correctly predict the AI advances of the past few years. Whether these people are also right about the next few years remains to be seen. But these are very smart people—the smartest people I have ever met—and they are the ones building this technology. Perhaps they will be an odd footnote in history, or perhaps they will go down in history like Szilard and Oppenheimer and Teller. If they are seeing the future even close to correctly, we are in for a wild ride.
Let me tell you what we see.
06-04-2024 - NYC Tech Week - Discussion on Vector Databases, Unstructured Data and AI
Discussion on Vector Databases, Unstructured Data and AI
https://www.meetup.com/unstructured-data-meetup-new-york/
This meetup is for people working in unstructured data. Speakers will come present about related topics such as vector databases, LLMs, and managing data at scale. The intended audience of this group includes roles like machine learning engineers, data scientists, data engineers, software engineers, and PMs.This meetup was formerly Milvus Meetup, and is sponsored by Zilliz maintainers of Milvus.
4th Modern Marketing Reckoner by MMA Global India & Group M: 60+ experts on W...Social Samosa
The Modern Marketing Reckoner (MMR) is a comprehensive resource packed with POVs from 60+ industry leaders on how AI is transforming the 4 key pillars of marketing – product, place, price and promotions.
06-04-2024 - NYC Tech Week - Discussion on Vector Databases, Unstructured Data and AI
Round table discussion of vector databases, unstructured data, ai, big data, real-time, robots and Milvus.
A lively discussion with NJ Gen AI Meetup Lead, Prasad and Procure.FYI's Co-Found
Enhanced Enterprise Intelligence with your personal AI Data Copilot.pdfGetInData
Recently we have observed the rise of open-source Large Language Models (LLMs) that are community-driven or developed by the AI market leaders, such as Meta (Llama3), Databricks (DBRX) and Snowflake (Arctic). On the other hand, there is a growth in interest in specialized, carefully fine-tuned yet relatively small models that can efficiently assist programmers in day-to-day tasks. Finally, Retrieval-Augmented Generation (RAG) architectures have gained a lot of traction as the preferred approach for LLMs context and prompt augmentation for building conversational SQL data copilots, code copilots and chatbots.
In this presentation, we will show how we built upon these three concepts a robust Data Copilot that can help to democratize access to company data assets and boost performance of everyone working with data platforms.
Why do we need yet another (open-source ) Copilot?
How can we build one?
Architecture and evaluation
Predictably Improve Your B2B Tech Company's Performance by Leveraging DataKiwi Creative
Harness the power of AI-backed reports, benchmarking and data analysis to predict trends and detect anomalies in your marketing efforts.
Peter Caputa, CEO at Databox, reveals how you can discover the strategies and tools to increase your growth rate (and margins!).
From metrics to track to data habits to pick up, enhance your reporting for powerful insights to improve your B2B tech company's marketing.
- - -
This is the webinar recording from the June 2024 HubSpot User Group (HUG) for B2B Technology USA.
Watch the video recording at https://youtu.be/5vjwGfPN9lw
Sign up for future HUG events at https://events.hubspot.com/b2b-technology-usa/
1. International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol. 4,No. 1, February 2014
DOI : 10.5121/ijcseit.2014.4104 33
SENTIMENT ANALYSIS BY USING FUZZY LOGIC
Md. Ansarul Haque1
, Tamjid Rahman 2
1
Department of Computer Science and Engineering, Stamford University, Bangladesh
2
Department of Computer Science and Engineering, Stamford University, Bangladesh
ABSTRACT
How could a product or service is reasonably evaluated by anyone in the shortest time? A million dollar
question but it is having a simple answer: Sentiment analysis. Sentiment analysis is consumers review on
products and services which helps both the producers and consumers (stakeholders) to take effective and
efficient decision within a shortest period of time. Producers can have better knowledge of their products
and services through the sentiment analysis (ex. positive and negative comments or consumers likes and
dislikes) which will help them to know their products status (ex. product limitations or market status).
Consumers can have better knowledge of their interested products and services through the sentiment
analysis (ex. positive and negative comments or consumers likes and dislikes) which will help them to know
their deserving products status (ex. product limitations or market status). For more specification of the
sentiment values, fuzzy logic could be introduced. Therefore, sentiment analysis with the help of fuzzy logic
(deals with reasoning and gives closer views to the exact sentiment values) will help the producers or
consumers or any interested person for taking the effective decision according to their product or service
interest.
KEYWORDS
Market status, Producer or consumer reviews, Sentiment analysis, Stakeholder.
1. INTRODUCTION
Now-a-days, time and reliable source is very much needed to gather the deserving information
related to any specific matter. Web in one sense can provide those deserving information
maintaining the less time and reliable source. Opinion is the vital type of information on the web.
These opinions are expressed in some user generated contents such as customer reviews of
products, micro-blogs, and forum posts. So, this is referred as online ‘word-of-mouth’.
Social media refers to the web-based technologies which turns the communication into an
interactive dialogue. These media are usually used for social interaction. These provide a huge
information about different individual’s interest and behaviors and also retrieve all the
information related to certain events. After retrieving, we can distinguish what is important and
what is negligible. [1] [2]
Among the top-ranked social networking sites, twitter (which launched in 2006) is very popular
for its micro-blogging features. Its information helps to answer the technological and sociological
queries. In this modern age, it is too expensive and time consuming to proceed without this type
of social network. Within a short period, around 160 million users are cope up with its service,
specifically saying with its allocation of 140 characters. Just for an example we could refer that
during the period of earthquake in Indonesia Twitter has given its feedback and played the key
role with the performance which was greater than the other electronic media such as television,
2. International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol. 4,No. 1, February 2014
34
newspapers and so on. Its vast flow of information helps to measure and analyze the users’
opinions regarding technological, social, environmental and other issues. [3]
1.1 Sentiment Analysis
Everyday millions of comments or opinions are posted in websites that provide the facilities for
micro-blogging such as Twitter or Facebook. The creators of the comments share their opinions
on different topics, discuss current issues even spot accidents or any flu outbreaks. These are the
valuable source of opinions and sentiments as huge amount of posts are posted by the users
according to their used products and services, or express their different views on different
perspectives. Researchers are using these posts to measure the public sentiment and to do
sentiment analysis. They are trying to determine the “PN-polarity” of subjective terms i.e,
identifies whether a term expresses the opinion which could have positive or negative
connotation.
The purpose of sentiment analysis is based on the two sectors:
i. Classifying Documents
Classifying documents or any passages according to sentiment orientation such as positive vs.
negative.
ii.Gathering Information
Extracting information of opinions which contains information of particular aspects of interest
and the corresponding sentiment orientation in a structured form from a set of unstructured data.
The tasks of classifying documents of the sentiment analysis can be divided into three sub-tasks:
i. Identifying SO polarity:
Whether the comment or post is referring a situation or event without disclosing the
subjectivity (positive or negative opinion) on it or expressing opinion on its subject matter.
Briefly, it means that identify the subjective or objective polarity of a post or comment.
ii. Identifying PN-polarity:
Whether a subjective post or comment is expressing positive or negative.
iii.Identifying the Degree of PN-polarity:
This step gives the impression of the degree of positivity or negativity on that opinion.
Positivity could be weakly positive, mildly positive or strongly positive and same could be for the
negative opinion.
"I think this is going to be one of the most important datasets of this era, because we are looking
at what people are talking about in real time at the scale of an entire society," says Mislove, an
assistant professor of computer science at Northeastern University. So, sentiment analysis is
doing the right thing as he told. [13][14] [2].
3. International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol. 4,No. 1, February 2014
35
1.2 SentiWordNet
SentiWordNet is the lexical (converts a sequence of characters into a sequence of tokens)
resource for sentiment analysis in which three numerical scores are maintained by Pos(), Neg() &
Obj() which represents how much positivity, negativity & objectivity are contained in those
opinions. This is called sentiment classification which determines the subjectivity of a given text.
It is the process of deciding whether a given text expresses a positive or negative opinion about its
“subject matter” and “subject attributes”, which also known as ‘product’ and ‘features’. It focuses
on the quantitative analysis. It is very much popular and free for the research works. It reflects a
nice outlook in its graphical user interface.
Considering the SentiWordNet (version 1.0), the synset are containing three numerical scores
Pos(s), Neg(s) & Obj(s). These numerical scores have the range from 0 to 1 and the sum of all
these scores is 1 for each synset. [14].
1.3 Twitter as a Micro-blogging Website
Micro-blogging is very popular communication tool among the internet users and one kind of
information center. In this micro-blog, people posts their real-time comments about their opinions
on a variety of topics, complain, ideas, discuss different issues and feel free to express their
sentiments regarding any products and services. Micro-blogging is getting popular and replacing
the traditional blogs as traditional blogs or mailing lists are not providing the free format of
messages & easy accessibility as the micro-blogging platforms are used to provide. For this
reason, twitter as a micro-blogging website, it is sometimes called as “The SMS of the internet”.
At present twitter is very popular micro-blogging websites considering the duration, impression
and popularity acquired by itself. It has launched on July, 2006. It’s a social networking service
that allows users to post real time messages and their opinions, called tweets. It is having
restriction of 140 characters in length and having no headache of misspellings, slangs or
abbreviations. It mainly focuses on individuation and characterization of opinions in a text. For
this reason, it is very much efficient in the field of sentiment analysis or opinion mining. It
contains a very large number of very short messages and the contents vary from the personal
thoughts to public expression. It classifies the tweets into three sentiments: positive, negative and
neutral. [11] [12].
The terminology is associated with tweets are as follows:
i. Emoticons: Facial expressions are the pictorial representations which express the user’s
mood. It also contains the punctuations and letters. Ex:
4. International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol. 4,No. 1, February 2014
36
three are representing positivity, next three are representing negativity and last two are
representing neutrality.
ii. Target: Users can use ‘@’ symbol to refer any other twitter user on the micro-blog and
Twitter automatically alerts them. Ex: I love to thank you for proposing that iPhone@John
iii. Hash tags: Users can use hash tags to mark topics to increase the visibility of their tweets. Ex:
Results for #electronic-media. [11].
Mostly used abbreviations or short-terms in twitter are:
1.4 Twitter API
API stands for Application Programming Interface and it is a defined way for a program to
establish a task especially by retrieving or modifying data. Twitter API helps the programmer to
make projects, applications or websites that interact with Twitter. As a user we use browsers
through HTTP (Hyper Text Transfer Protocol) to visit and interact with the websites where
programs talk to the Twitter API over HTTP. In our project we have used Twitter 4J as for the
Twitter API which is the unofficial Java library.
Accessing twitter through API is ten times more efficient than the web interface. For this purpose,
different users can use different API, such as:
i. Desktop users: twitterrific & twhirl
ii.Cell users: TinyTwitter, PocketTweets & iTweets [15]
have accessed these databases through the programming commands.
2. Proposed Techniques of Sentiment Analysis
2.1 Introducing Required Tools
i. SentiWordNet
In our project we have used the SentiWordNet 3.0.0 for having the values of the user’s opinions
or sentiments. Basically it is using the positive and negative values. For the zeros of both
(positive and negative), we are considering the neutral opinion. We have described about the
SentiWordNet in the earlier sections (section 3.2).
ii. Twitter Website
From the twitter (www.twitter.com) we have extracted the tweets to analyze them. As for
example, in twitter we can search for any specific topic or product such as iPhone. Different
individuals have different thoughts of their products and they share their positive and negative
opinions. These tweets are useful for the analysis and for our project.
Abbreviations Meaning
Lol laughing out loud
gr8 Great
Bff best friend forever
Rotf rolling on the floor
5. International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol. 4,No. 1, February 2014
37
iii. Java (Eclipse 3.7.1)
For developing applications in Java, Eclipse is the well-suited software development
environment. It comprises with Integrated Development Environment (IDE) and various plug-ins.
We have chosen to work with the Eclipse 3.7.1 among the different versions.
iv. Databases
We are using the databases in storing tweets after extracting and also the SentiWordNet words
with the values. Databases are in the form of Microsoft Excel Files (such as .xls or .txt).
Our Project Procedure
2.2.1 Searching Tweets
We have used the website www.twitter.com to have the frequent tweets from the users. As per we
know that tweets are the comments or posting of the users in the Twitter. The users can post their
tweets depend on their ideas, concepts or their likings about any product or service. We have used
Twitter 4j for retrieving tweets from the Twitter website and at a time we are extracting 100
tweets by using this package.
As our concentrating item of the project is “iPhone”, we went through that term. Whatever we
get, we can classify them in two forms
a. Subjective and
b. Objective
Subjective means it is representation of the mixture of positive and negative values or sets and on
the other hand, the objective means representation of neutral values (0) or sets. In our project, we
6. International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol. 4,No. 1, February 2014
38
have skipped the neutral values or sets. Later we used the POS-tagging where we used the parts-
of-speech for tagging the tokens. Subjective texts tend to use base form of verbs (VB) and also
simple past tense (VBD) instead of the past participle (VBN). Adverb (RB) is mostly used in
subjective texts to give an emotional color to a verb.
2.2.2 Saving the Extracted Tweets
After having the tweets, we have saved those tweets in text files (such as “TempTweets.txt”) by
using FileWriter application. At present, we have so many assisting tools to extract tweets from
different sources. As for example, through programming we could use API which is facilitated
by popular environment such as Java or by using well-established software such as Archivist.
2.2.3 Reading the File (where Tweets are stored)
FileInputStream fstream = new FileInputStream("C:/Users/stalin/Desktop/Temptweets.txt");
By using the FileInputStream, we accessed the Temptweets.txt which one was created after
extracting and saving the extracted tweets. Here we constructed the object of the FileInputStream
which extract data from the excel file. Then we used BufferReader for reading line by line.
2.2.4 Translating the Tweets to English
Translate.setKey("2768f0575d056bb86c91a4b0cf588e1d7382c15a");String translatedText =
Translate.execute(strLine,Language.ENGLISH);
The tweets could be from different users of different nationalities in different languages. Those
tweets of different languages except English are also important enough to analyze sentiments.
There are so many translators with API to translated in English language in our real life. For our
project, we have used Microsoft Translator to get those tweets in English and we have used the
API key .
("2768f0575d056bb86c91a4b0cf588e1d7382c15a") to translate tweets.
2.2.5 Tagging the Tweets
MaxentTagger tagger = new MaxentTagger("D:/java prog/FLproject/taggers/bidirectional-distsim-wsj-0-
18.tagger");
String tagged = tagger.tagString(translatedText);
For the sake of our sentiment analysis, at first we tagged the tweets according to the parts-of-
speech such as noun, verb, adjective, etc. As in our database (SentiWordNet), we have different
focus on different parts-of-speech. As prepositions and conjunctions are too common to be used
among statements, we can remove them easily after tagging as well as proper nouns which
usually don’t have an affective content. After removing the unaffecting content, we usually have
four POS: adjective, verb, adverbs and noun which are known as opinion words.
POS-tagger is an application that helps to read text in some languages and assign part-of-speech
to individual terms, words or tokens. The tagger what we have used here is written by The
Stanford Natural Language Processing Group in Java. Among its three models, we have used the
English tagger model. POS tagger which we used is named as “Penn Treebank Tagset”:
7. International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol. 4,No. 1, February 2014
39
Here, WP$ Possessive wh-pronoun
POS Possessive Ending
NNS Noun Plural
IN Preposition
VBN Verb, past participle
VBZ Verb, 3rd
person singular present
JJRAdjective, comparative
MD Modal
RBS Adverb, superlative
TO to
NNNoun, singular or mass
JJSAdjective, superlative
JJ Adjective
DTDeterminer
VBG Verb, gerund
VBD Verb, past tense
WDTWh-determiner
WP Wh-pronoun
RP Participle
VBVerb, base form RBR
RBR Adverb, comparative
CC Coordinating Conjunction
Ex Extential there
VBP Verb, non-3rd
person
WRB Wh-adverb
RB Adverb
PDTPre-determiner
UHInterjection
We have given some cases for the above figure:
i. Objective texts contain more common and proper nouns.
ii. Verbs in objective texts are usually in the third person and used more often in past
participle.
iii.Superlative adjectives are used more for expressing emotions and opinions, comparative
adjective are used for starting facts.
iv. Verbs in base form are used with modal verbs to express emotions.
v. Authors of subjective texts usually write about themselves (verbs in first person)or
address the audience (second person) and tend to use simple past tense.
vi. Subjective texts contain more personal pronouns.
vii.Utterances are strong indicators of a subjective text.
8. International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol. 4,No. 1, February 2014
40
2.2.6 Removing Punctuation Marks
String pure = translatedText.replaceAll("w+@w+.[w]{3} ","");
String strippedInput = pure.replaceAll("[!@#(:]","");
As the punctuation mark will not have any effect on the sentiment analysis, we could remove
those marks whatever belongs to the tweets. It can also be called as filtering. Here in our project
we tried to apply some punctuation marks such as @, #, !, ‘,’, (: and so on. These removing of the
punctuation marks would help to tokenize our tweets perfectly and efficiently. Such example of
tweets:
iPhone made me friendly @david
Here, we would like to remove the twitter user name (david) which is having the symbol @
giving nothing about opinions only tagged that named person.
Hashtags are words or phrases used by users to group posts together by topic or type, and they are
prefixed by “#” as in example #iphone4s. As they are not used to express opinions, we can
remove them before the analysis.
2.2.7 Tokenization
String[] a=strippedInput.split(" ");
According to our project, tokens are the meaningful terms or elements of text. Tokens can be any
words, symbols or phrases. Tokenization is the process of splitting a stream of text up into those
forms of tokens. This process is very much related to sentiment analysis as our database which is
referring as SentiWordNet works with those tokens. For our simplification, we have changed all
the tokens into lower-case.
2.2.8 Removing Stop Words
if(aList1.contains(stopwordArr));
for(int i=0;i<stopwordArr.length;i++)
{
aList1.remove(stopwordArr[i]);
}
Stop words are those which are non-relevant of our sentiment values. If any tweet is having those
stop words, it will be removed from the list. This is applied for all the extracting tweets. The list
words of stop words we have used in our project are so many. Among them we would like to
pick-up some of them: a, about, above, after, again, against, all, am, an, and, are, aren’t, at, be,
because, been, before, being , below, between, both, but, by, can’t, cant, could, couldn’t, do, does,
9. International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol. 4,No. 1, February 2014
41
doesn’t, did, didn’t, doing, down, during, each, below, for, further, has, hasn’t, have, haven’t,
he’ll, he’s, her, here, here’s, hers, herself, himself, him, himself, it’s, its, let’s, me, more, mustn’t,
my, myself, no, nor, off, on, of, what, what’s, when, where, where’s, which, who’s, why, whom,
why’s, with and so on
2.2.9 Creating Database Connection with the File
Class.forName("sun.jdbc.odbc.JdbcOdbcDriver");
Connection con = DriverManager.getConnection("jdbc:odbc:TempData");
Statement stmt = con.createStatement();
We have created connection with our database (TempData which is having SentiWords with
positive and negative values) which works as SentiWordNet.
2.2.10Storing Splitting Tweets in Array
String[] s1 = aList1.toArray(new String[aList1.size()]);
All the splitting tweets should be listed in array for simplification of calculating scores.
2.2.11SQL for Extracting Scores
SQL = "SELECT PosScore,NegScore FROM Sheet1 WHERE SynsetTerms='" + s1[j] + "'" ;
stmt.execute(SQL);
ResultSet rs = stmt.getResultSet();
The database has two columns for the positive and negative scores of the sentiment terms.
Through the SQL (Structured Query Language), we accessed the positive and negative scores of
the each term of the tweets.
2.2.12 Calculating Total Scores
Score = (Positivescore - Negativescore);
After getting positive and negative scores of the each term of the tweet, we calculated the total
positive score and total negative scores. Then we deducted each other and got the final score.
2.2.13 Normalization
Normalization helps to organize data and it also minimizes redundancy. It produces well-
structured relations in the database.
10. International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol. 4,No. 1, February 2014
42
2.2.14 Implementing Weights
Weights have given to the frequently used terms in our project. As we have concentrated on
‘iPhone’, we have given such weights to the mostly used terms. For example, 0.95 for iPhone, 0.9
for iPhone4s, 0.85 for iPhone4g.
2.2.15 Identifying and Categorizing the Positive and Negative Tweets
If the total score (positive score-negative score) is greater than 0, we will consider it as a positive
tweet and if it is smaller than 0, we will consider it as an negative tweet. In this approach, we will
count the number of positive and negative tweets. But the positivity and negativity depends on
their tweet scores.
For the further classification which depends on tweet score, we have six different classes of
tweets.
a. Strong Positive Tweets
b. Positive Tweets
c. Weak Positive Tweets
d. Weak Negative Tweets
e. Negative Tweets
f. Strong Negative Tweets
2.2.16 Computing the Results
In our project, we tried to analyze sentiments on the tweets of ‘iPhone’ and at last we calculated
the following results:
a. No. of Tweets
Counting the number of tweets to be processed.
b. Total no. of Positive Tweets
Let, t is the token or words and sent(t) is the sentiment value. Then we can take the tweet as the
positive tweet if
{(t € Tweets) && (sent(t) > 0)}
c. Total no. of Negative Tweets
Let, t is the token or words and sent(t) is the sentiment value. Then we can take the tweet as the
negative tweet if
{(t € Tweets) && (sent(t) < 0)}
d. Weighted Mean
Weighted Mean = ( iai ) / i )
Here, wi = weight and ai = value of the tweet
11. International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol. 4,No. 1, February 2014
43
If we want to give more important values to the main features of any product or service, we could
use weighted mean. Descriptive statistics uses this mean very precisely. It is given priority more
focus to the main characteristics more than other usual characteristics.
e. Arithmetic Mean
Arithmetic Mean = (Total Values of the Tweets / Total No. of Tweets)
Arithmetic mean is giving equal importance to all data and this is the basic difference between
weighted and arithmetic mean.
f. Positive Sentiment by Percentage
PS (%) = (Number of Positive Sentiments/Total Number of Tweets) *100
g. Negative Sentiment by Percentage
NS (%) = (Number of Negative Sentiments/Total Number of Tweets) *100
Our analysis could give such results as in the following diagram:
3. RESULT DISCUSSION
Here for the testing purpose of our project, we have extracted six (10) tweets regarding ‘’iPhone’’
and they are displaying in our output screen in the following form:
Input:
@jasonenriquez:J'aime mon Iphone4S.
@stalin:iphone 4s is lovely!!
@YasminScott98:Touch-screen of iphone@ is lovely http://t.co/HY1zqtzq
and attractive
@JessMarieFrench:naked iphone(: is catchy and shiny
@dgrey1986:iphone4s is sloppy in battery
@hlouisewagg:Damn iphone
@BxDiimegambler:iphone is Not bad
12. International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol. 4,No. 1, February 2014
44
@SabrinaHu5:iphone is not good
@nash711:nokia 4 is good
@GersForum:So I just got the iPhone 4s and it's amazing :)
Here, we have extracted 10 tweets as sample. The tweets are initialized with the username
followed by their given comments. From the first tweet, we can see its in French and others are in
English. We have shown our output below and we can find out how our project is working for
different languages, its tagging, calculating the tweet values and also showing the tweets status
according to their degree of positivity and negativity.
Output:
Loading default properties from trained tagger D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger
Reading POS tagger model from D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger ... done
[12.9 sec].
@/SYM jasonenriquez/NN :/: I/PRP love/VBP my/PRP$ Iphone4S/NNS ./.
0.25 positive
Loading default properties from trained tagger D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger
Reading POS tagger model from D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger ... done
[3.2 sec].
@/IN stalin/NN :/: iphone/NN 4s/NNS is/VBZ lovely/JJ !!/NN
0.25 positive
Loading default properties from trained tagger D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger
Reading POS tagger model from D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger ... done
[2.5 sec].
@/IN YasminScott98/NNP :/: Touch-screen/NN of/IN iphone/NN @/SYM is/VBZ
lovely/JJ http://t.co/HY1zqtzq/JJ and/CC attractive/JJ
0.375 positive
Loading default properties from trained tagger D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger
Reading POS tagger model from D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger ... done
[3.3 sec].
@/IN JessMarieFrench/NNP :/: naked/JJ iphone/NN -LRB-/-LRB- :/: is/VBZ
catchy/JJ and/CC shiny/JJ
0.375 positive
Loading default properties from trained tagger D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger
13. International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol. 4,No. 1, February 2014
45
Reading POS tagger model from D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger ... done
[3.4 sec].
@/IN dgrey1986/CD :/: iphone4s/NNS is/VBZ sloppy/JJ in/IN battery/NN
0.1875 weak_positive
Loading default properties from trained tagger D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger
Reading POS tagger model from D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger ... done
[3.4 sec].
@/SYM hlouisewagg/NN :/: Damn/JJ iphone/NN
-0.75 negative
Loading default properties from trained tagger D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger
Reading POS tagger model from D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger ... done
[3.1 sec].
@/IN BxDiimegambler/NNP :/: iphone/NN is/VBZ Not/RB bad/JJ
0.375 positive
Loading default properties from trained tagger D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger
Reading POS tagger model from D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger ... done
[3.3 sec].
@/IN SabrinaHu5/NNP :/: iphone/NN is/VBZ not/RB good/JJ
-1.0 negative
Loading default properties from trained tagger D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger
Reading POS tagger model from D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger ... done
[3.2 sec].
@/IN nash711/CD :/: nokia/NN 4/CD is/VBZ good/JJ
0.625 positive
Loading default properties from trained tagger D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger
Reading POS tagger model from D:/java
prog/FLproject/taggers/bidirectional-distsim-wsj-0-18.tagger ... done
[3.3 sec].
@/IN GersForum/NNP :/: So/IN I/PRP just/RB got/VBD the/DT iPhone/NNP
4s/NNS and/CC it/PRP 's/VBZ amazing/JJ :/: -RRB-/-RRB-
0.6875 positive
Total no of tweets is:10.0
Total no of positive tweets:8.0
14. International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol. 4,No. 1, February 2014
46
Total no of negative tweets:2.0
Arithmetic mean is:0.1375
Sentiment by Percent
Positive sentiment % is:80.0
Negative sentiment % is:20.0
First of all, the POS tagger is generated and starts tagging the extracted tweets. Then computing
the sentiment values of the tweets are done and given their sentiment status. By having the values
of the tweets and the weights, we can compute the weighted and arithmetic mean. Then we can
have the percentage of the individual sentiments (positive and negative).
Fig 3: Percentage of the Sentiments
A big problem regarding the word-level analysis is that detecting the use of irony and sarcasm is
very hard, as a whole understanding of the sentence is required for that. Still, we have tried to
improve this word-level analysis and trying to detect negation particles before adjectives. For
example, if we analyze the words “iphone is not bad”, it will get an overall negative value for
‘not’ and ‘bad’. But the combination of those words is supposed to have a positive meaning. So,
if we can detect those negation particles, we can invert the value of the following adjective.
4. ANALYSIS
In this paper, we did analysis on 100 tweets. But as we could recommend that, the more the
sentiments the more bold the analysis result. For our future works, we would like to extend our
work with more tweets and have more robust result. This result would benefit the interested users
with strong beliefs.
One of the limitations of our project is that it is focused on the sentiment classification and
concentrating to manipulate the results according to those classifications (positive or negative). It
is not focusing on the feature based classification. In our further steps, we will try to include this
limitation and make stronger in this context.
We would like to concentrate on the analysis of context and domain, as both of them has the
capability to influence the word or sentiment or opinion’s attitude. As different individual could
define their concepts or ideas in different manner, so it is really meaningful if we could cope up
with that matter.
15. International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol. 4,No. 1, February 2014
47
Sentiment analysis is really a challenging work for some critical perspective such as colloquial
languages or opinions or for the ironic words from different reviewers. Our future works will
have some steps on those obligations and try to get proper solutions.
For the ranking perspective, we could rank the reviewers for their most positive or negative posts
or sentiments. We mean that have some feedbacks from the users normally and we would rank
their feedbacks by manipulating the values of positivity or negativity of the sentiments. Then the
interested entities could have the like or dislike portions on which they should concentrate
hierarchically.
5. CONCLUSIONS
Sentiment analysis is nothing but special field of text analysis. In short, focus and analyze the
extracted opinions (sentiments or emotional contents) from the posted comments. Our project
goal is to analyze the sentiments on a topic which are extracted from the Twitter and conclude a
remark (positive/negative) of the defined topics. We have implemented an easier procedure to
analyze sentiments on any interested field or topic. Hope this project would helpful for anyone in
any way to meet up their interests or what they deserve. This is our major goal of this project and
waiting to provide much more worthy works in our future work.
ACKNOWLEDGEMENT
We would have pleasure to thank the referees, who provided very useful and efficient resources.
Without their insight resources and information, It would be very difficult to have our such
position and project. Its really hard to express our gratitude level to them for their immense
contributions.
Special thanks to David L. Hicks for his timely supervision and his benevolent guidelines. He was
so caring and ready to help regarding the project, its objectives and aim, report writing and
editorial guidance and so on.
We are also very much thankful to Henrik Legind Larsen. Very much effective part of our project
is to have him and have proper direction in such manner to fulfill this project in the scheduled
time with huge positive feedback.
No comments for the friends of other groups especially Damien, Emanuel, Sayeedi and many
others.
At last we worked as a good communicating folk who interact and discuss each other to make this
project properly. We invested our efforts and hard labor. We tried to make it more perfect and
robust without any error or fault.
All of the remaining errors and faults are, of course, our own.
.
REFERENCES
[1] “A Holistic Lexicon-Based Approach to Opinion Mining”, Xiaowen Ding, Bing Liu, Philips S. Yu,
University of Illinois at Chicago
[2] “A Review of Opinion Mining and Sentiment Classification Framework in Social Networks”, Yee W.
Lo, Vidyasagar Potdar, ISBN: 978-1-4244-2345-3
16. International Journal of Computer Science, Engineering and Information Technology (IJCSEIT), Vol. 4,No. 1, February 2014
48
[3] “Twitter as Medium and Message”, Neil Savage, March 2011
[4] ”Hybrid Controller based Intelligent Speed Control of Induction Motor”, Vinod Kumar, R.R. Joshi,
http://jatit.org/volumes/research-papers/Intelligent_Speed_Control_3_1.pdf
[5] “Fuzzy Logic Systems”, James Vernon, http://www.control-systems-
principles.co.uk/whitepapers/fuzzy-logic-systems.pdf
[6] “Cahier Technique n0 191, Fuzzy Logic”, F. Chevrie, F. Guely, http://www2.schneider-
electric.com/documents/technical-publications/en/shared/automation/automation-information-
networks/ect191.pdf
[7] “Fuzzy rule-based system for prediction of direct action avalanches”, accepted 31st January, 2004,
Lalit Mohan Pant, Ashwagosha Ganju, http://www.ias.ac.in/currsci/jul102004/99.pdf
[8] ”Fundamentals of fuzzy sets and fuzzy logic”, Henrik Legind Larsen, Aalborg University Esbjerg.
[9] “Introduction to Fuzzy Sets, Fuzzy Logic, and Fuzzy Control Systems”, Guanrong Chen, Trung Tat
Pham, ISBN 0-8493-1658-8
[10] “Fuzzy Sets and Fuzzy Logic, Theory and Applications”, George J. Klir/Bo Yuan, ISBN 0-13-
101171-5
[11] “Sentiment Analysis of Twitter Data”, Apoorv Agarwal, Boyi Xie, Ilia Vovsha, Owen Rambow,
Rebecca Passonneau, 23 June, 2011
[12] “Mining Opinions on the Basis of Their Affectivity”, Domenico Potena, Claudia Diamantini, 978-1-
4244-6622-1/10/$26.00 2010 IEEE
[13] “Twitter as a Corpus for Sentiment Analysis and Opinion Mining”, Alexander Pak, Patrick Paroubek,
[14] “SentiWordNet: A Publicly Available Lexical Resource for Opinion Mining”, Andrea Esuli, Fabrizio
Sebastiani.
[15] “Programming to the Twitter API”, Bar Camp Boston 3, May 18, 2008, John Eckman,
http://www.openparenthesis.org/wp-content/uploads/2008/05/bcb3-retweeter.pdf
[16] http://nlp.stanford.edu/software/tagger.shtml
[17] http://www.computing.dcu.ie/~acahill/tagset.html, Penn Treebank Tagset
[18] ”Sentence Factorization for Opinion Feature Mining”, Chun-hung Li, 978-0-7695-3740-5/09 2009
IEEE
Authors
Md. Ansarul Haque is a faculty of Stamford University, Bangladesh. He received his
M.Sc degree in IT, Halmstad University, Sweden and B.Sc in CSE from Shahjalal
University, Bangladesh. His research interest lies in wireless communication, networking
and fuzzy systems.
Tamjid Rahman is a faculty of Stamford University, Bangladesh and he received his BS
degree in Computer Science and Engineering from Stamford University Bangladesh in
2010. He is a MS student of North South University. His research interest lies in artificial
intelligence, software engineering, programming.