This document summarizes an unsupervised model called Author Interaction Topic Viewpoint (AITV) Model for discovering viewpoints in online debates. The model leverages both the post contents and interactions between authors. It achieved state-of-the-art performance in identifying viewpoints at the post level and was competitive with supervised models in identifying viewpoints at the author level. The model represents each post as a mixture of topics and viewpoints and models the interactions between authors' viewpoints to improve viewpoint identification. Experiments on several online debate datasets showed the model outperformed other unsupervised baselines.
Rsqrd AI: Recent Advances in Explainable Machine Learning ResearchSanjana Chowdhury
In this talk, Bernease Herman speaks about recent explainable ML research
Presented on 06/06/2019
**These slides are from a talk given at Rsqrd AI. Learn more at rsqrdai.org**
Entity-oriented sentiment analysis of tweets: results and problemsYuliya Rubtsova
This is summarization of the results of the reputation-oriented
Twitter task, which was held as part of SentiRuEval evaluation of Russian sentiment-analysis systems. The tweets in two domains: telecom companies and banks - were included in the evaluation. The task was to determine if an author of a tweet has a positive or negative attitude to a company mentioned in the message. The main issue of this paper is to analyze the current state and problems of approaches applied by the participants.
Rsqrd AI: Recent Advances in Explainable Machine Learning ResearchSanjana Chowdhury
In this talk, Bernease Herman speaks about recent explainable ML research
Presented on 06/06/2019
**These slides are from a talk given at Rsqrd AI. Learn more at rsqrdai.org**
Entity-oriented sentiment analysis of tweets: results and problemsYuliya Rubtsova
This is summarization of the results of the reputation-oriented
Twitter task, which was held as part of SentiRuEval evaluation of Russian sentiment-analysis systems. The tweets in two domains: telecom companies and banks - were included in the evaluation. The task was to determine if an author of a tweet has a positive or negative attitude to a company mentioned in the message. The main issue of this paper is to analyze the current state and problems of approaches applied by the participants.
Sentiment Analysis also known as opinion mining and Emotional AI
Refers to the use of natural language processing, text analysis, computational linguistics and biometrics to systematically identify, extract, quantify and study affective states and subjective information.
widely used in
Reviews
Survey responses
Online and social media
Health care
SentiTweet is a sentiment analysis tool for identifying the sentiment of the tweets as positive, negative and neutral.SentiTweet comes to rescue to find the sentiment of a single tweet or a set of tweets. Not only that it also enables you to find out the sentiment of the entire tweet or specific phrases of the tweet.
Predictions of links in graphs based on content and information propagations.
Lecture for the M. Sc. Data Science, Sapienza University of Rome, Spring 2016.
A Framework for Arabic Concept-Level Sentiment Analysis using SenticNet IJECEIAES
Arabic Sentiment analysis research field has been progressing in a slow pace compared to English and other languages. In addition to that most of the contributions are based on using supervised machine learning algorithms while comparing the performance of different classifiers with different selected stylistic and syntactic features. In this paper, we presented a novel framework for using the Concept-level sentiment analysis approach which classifies text based on their semantics rather than syntactic features. Moreover, we provided a lexicon dataset of around 69 k unique concepts that covers multi-domain reviews collected from the internet. We also tested the lexicon on a test sample from the dataset it was collected from and obtained an accuracy of 70%. The lexicon has been made publicly available for scientific purposes.
I created this presentation to present my research work to the committee. My research was on extracting tweets and analyzing it with an previously created ontology model. The results of the ontology model will help in identifying the domain area of the problem for which use had shared negative sentiments on tweeter. This system along with the ontology model developed for Postal service domain. The next step in research will be to generate automated responses on twitter to the user who shares negative sentiments.
Improving Sentiment Analysis of Short Informal Indonesian Product Reviews usi...TELKOMNIKA JOURNAL
Sentiment analysis in short informal texts like product reviews is more challenging. Short texts are
sparse, noisy, and lack of context information. Traditional text classification methods may not be suitable
for analyzing sentiment of short texts given all those difficulties. A common approach to overcome these
problems is to enrich the original texts with additional semantics to make it appear like a large document of
text. Then, traditional classification methods can be applied to it. In this study, we developed an automatic
sentiment analysis system of short informal Indonesian texts using Naïve Bayes and Synonym Based
Feature Expansion. The system consists of three main stages, preprocessing and normalization, features
expansion and classification. After preprocessing and normalization, we utilize Kateglo to find some
synonyms of every words in original texts and append them. Finally, the text is classified using Naïve
Bayes. The experiment shows that the proposed method can improve the performance of sentiment
analysis of short informal Indonesian product reviews. The best sentiment classification performance using
proposed feature expansion is obtained by accuracy of 98%.The experiment also show that feature
expansion will give higher improvement in small number of training data than in the large number of them.
In these slides we present a model that was intended to discriminate creative from non-creative news articles. In order to build the classifier, we have combined nine different measures using a stepwise logistic regression model. The obtained model was tested in two experiments: the first one tried to discriminate between news articles about the US 2012 Elections from different newspapers versus articles taken from The Onion (a website providing satiric news) on the same subject, while the second one evaluated the capacity of the model to generalize over different topics and text genres. The experiments showed that the system achieves 80% accuracy, but the lack of true positives from the second experiment raised the question of whether we really identified creativity or in fact we detected satire (as the assumption for the training corpus was that the satiric news from The Onion were also creative).
Sentiment mining- The Design and Implementation of an Internet PublicOpinion...Prateek Singh
Sentiment mining paper presentation, database mining and business intelligence.
The Design and Implementation of an Internet PublicOpinion Monitoring and Analysing System
Supervised Sentiment Classification using DTDP algorithmIJSRD
Sentiment analysis is the process widely used in all fields and it uses the statistical machine learning approach for text modeling. The primarily used approach is Bag-of-words (BOW). Though, this technique has some limitations in polarity shift problem. Thus, here we propose a new method called Dual sentiment analysis (DSA) which resolves the polarity shift problem. Proposed method involves two approaches such as dual training and dual prediction (DPDT). First, we propose a data expansion technique by creating a reversed review for training data. Second, dual training and dual prediction algorithm is developed for doing analysis on sentiment data. The dual training algorithm is used for learning a sentiment classifier and the dual prediction algorithm is developed for classifying the review by considering two sides of one review.
The (standard) Boolean model of information retrieval (BIR) is a classical information retrieval (IR) model and, at the same time, the first and most-adopted one. ... The BIR is based on Boolean logic and classical set theory in that both the documents to be searched and the user's query are conceived as sets of terms.
International Journal of Engineering Research and Applications (IJERA) is an open access online peer reviewed international journal that publishes research and review articles in the fields of Computer Science, Neural Networks, Electrical Engineering, Software Engineering, Information Technology, Mechanical Engineering, Chemical Engineering, Plastic Engineering, Food Technology, Textile Engineering, Nano Technology & science, Power Electronics, Electronics & Communication Engineering, Computational mathematics, Image processing, Civil Engineering, Structural Engineering, Environmental Engineering, VLSI Testing & Low Power VLSI Design etc.
Sentiment Analysis also known as opinion mining and Emotional AI
Refers to the use of natural language processing, text analysis, computational linguistics and biometrics to systematically identify, extract, quantify and study affective states and subjective information.
widely used in
Reviews
Survey responses
Online and social media
Health care
SentiTweet is a sentiment analysis tool for identifying the sentiment of the tweets as positive, negative and neutral.SentiTweet comes to rescue to find the sentiment of a single tweet or a set of tweets. Not only that it also enables you to find out the sentiment of the entire tweet or specific phrases of the tweet.
Predictions of links in graphs based on content and information propagations.
Lecture for the M. Sc. Data Science, Sapienza University of Rome, Spring 2016.
A Framework for Arabic Concept-Level Sentiment Analysis using SenticNet IJECEIAES
Arabic Sentiment analysis research field has been progressing in a slow pace compared to English and other languages. In addition to that most of the contributions are based on using supervised machine learning algorithms while comparing the performance of different classifiers with different selected stylistic and syntactic features. In this paper, we presented a novel framework for using the Concept-level sentiment analysis approach which classifies text based on their semantics rather than syntactic features. Moreover, we provided a lexicon dataset of around 69 k unique concepts that covers multi-domain reviews collected from the internet. We also tested the lexicon on a test sample from the dataset it was collected from and obtained an accuracy of 70%. The lexicon has been made publicly available for scientific purposes.
I created this presentation to present my research work to the committee. My research was on extracting tweets and analyzing it with an previously created ontology model. The results of the ontology model will help in identifying the domain area of the problem for which use had shared negative sentiments on tweeter. This system along with the ontology model developed for Postal service domain. The next step in research will be to generate automated responses on twitter to the user who shares negative sentiments.
Improving Sentiment Analysis of Short Informal Indonesian Product Reviews usi...TELKOMNIKA JOURNAL
Sentiment analysis in short informal texts like product reviews is more challenging. Short texts are
sparse, noisy, and lack of context information. Traditional text classification methods may not be suitable
for analyzing sentiment of short texts given all those difficulties. A common approach to overcome these
problems is to enrich the original texts with additional semantics to make it appear like a large document of
text. Then, traditional classification methods can be applied to it. In this study, we developed an automatic
sentiment analysis system of short informal Indonesian texts using Naïve Bayes and Synonym Based
Feature Expansion. The system consists of three main stages, preprocessing and normalization, features
expansion and classification. After preprocessing and normalization, we utilize Kateglo to find some
synonyms of every words in original texts and append them. Finally, the text is classified using Naïve
Bayes. The experiment shows that the proposed method can improve the performance of sentiment
analysis of short informal Indonesian product reviews. The best sentiment classification performance using
proposed feature expansion is obtained by accuracy of 98%.The experiment also show that feature
expansion will give higher improvement in small number of training data than in the large number of them.
In these slides we present a model that was intended to discriminate creative from non-creative news articles. In order to build the classifier, we have combined nine different measures using a stepwise logistic regression model. The obtained model was tested in two experiments: the first one tried to discriminate between news articles about the US 2012 Elections from different newspapers versus articles taken from The Onion (a website providing satiric news) on the same subject, while the second one evaluated the capacity of the model to generalize over different topics and text genres. The experiments showed that the system achieves 80% accuracy, but the lack of true positives from the second experiment raised the question of whether we really identified creativity or in fact we detected satire (as the assumption for the training corpus was that the satiric news from The Onion were also creative).
Sentiment mining- The Design and Implementation of an Internet PublicOpinion...Prateek Singh
Sentiment mining paper presentation, database mining and business intelligence.
The Design and Implementation of an Internet PublicOpinion Monitoring and Analysing System
Supervised Sentiment Classification using DTDP algorithmIJSRD
Sentiment analysis is the process widely used in all fields and it uses the statistical machine learning approach for text modeling. The primarily used approach is Bag-of-words (BOW). Though, this technique has some limitations in polarity shift problem. Thus, here we propose a new method called Dual sentiment analysis (DSA) which resolves the polarity shift problem. Proposed method involves two approaches such as dual training and dual prediction (DPDT). First, we propose a data expansion technique by creating a reversed review for training data. Second, dual training and dual prediction algorithm is developed for doing analysis on sentiment data. The dual training algorithm is used for learning a sentiment classifier and the dual prediction algorithm is developed for classifying the review by considering two sides of one review.
The (standard) Boolean model of information retrieval (BIR) is a classical information retrieval (IR) model and, at the same time, the first and most-adopted one. ... The BIR is based on Boolean logic and classical set theory in that both the documents to be searched and the user's query are conceived as sets of terms.
International Journal of Engineering Research and Applications (IJERA) is an open access online peer reviewed international journal that publishes research and review articles in the fields of Computer Science, Neural Networks, Electrical Engineering, Software Engineering, Information Technology, Mechanical Engineering, Chemical Engineering, Plastic Engineering, Food Technology, Textile Engineering, Nano Technology & science, Power Electronics, Electronics & Communication Engineering, Computational mathematics, Image processing, Civil Engineering, Structural Engineering, Environmental Engineering, VLSI Testing & Low Power VLSI Design etc.
A SURVEY PAPER ON EXTRACTION OF OPINION WORD AND OPINION TARGET FROM ONLINE R...ijiert bestjournal
Opinion mining is nothing but mining opinion target s and opinion words from online reviews. To find op inion relation among them partially supervised word align ment model have used. To find confidence of each candidate graph based co-ranking algorithm have used. Further candidates having confidence higher than threshold value are extracted as opinion word or opinion targets. Compa red to previous approach syntax-based method this m ethod can give correct results by eliminating parsing errors and can work on reviews in informal language. Compa red to nearest neighbor method this method can give more p recise results and can find relations within a long span. Also to decrease error propagation graph based co-r anking algorithm is used to collectively extract op inion targets and opinion words. Also to decrease probability of error generation penetration of high degree vertice s is done and decrease effect of random walk.
Evolving Swings (topics) from Social Streams using Probability ModelIJERA Editor
Evolving swings from social streams is receiving renewed interest and it is motivated by the growth of social
media and social streams. Non-conventional based approaches can be appropriate which include text, images,
URLs and videos. The focus is on evolving topics by social aspects of the networks and the mentions of user
links between users which are generated intentionally or unintentionally through replies, mentions and retweets.
A probability model of the mentioning behavior is proposed and the proposed model detects the evolving topic
from the anomalies measured. After a several experiments, it shows that mention anomaly based approaches
detects the evolving swing as early as text anomaly based approa0ches.
opinion feature extraction using enhanced opinion mining technique and intrin...INFOGAIN PUBLICATION
Mining patterns are the main source of opinion feature extraction techniques, which was individually evaluated corpus mostly belong to evaluated corpus. A measure called Domain Relevance is used to identify candidate features from domain dependent and domain independent corpora both. Opinion Features originated are relevant to a domain. For every extracted candidate feature its individual Intrinsic Domain Relevance and Extrinsic Domain Relevance values are registered. Threshold has been compared with these values and recognizes as best candidate features. In this thesis, By applying feature filter creation the features from online reviews can be identified .
Graph-based Analysis and Opinion Mining in Social NetworkKhan Mostafa
This is the final report for Networks & Data Mining Techniques project focusing on mining social network to estimate public opinion about entities and associated keywords. This project mines Twitter for recent feeds and analyzes them to estimate sentiment score, discussed entity and describing keywords in each tweet. This data is then exploited to elicit overall sentiment associated with each entity. Entities and keywords extracted is also used to form an entity-keyword bigraph. This graph is further used to detect entity communities and keywords found within those communities. Presented implementation works in linear time.
EMBERS AutoGSR: Automated Coding of Civil Unrest EventsParang Saraf
EMBERS AutoGSR is a novel, web based framework that generates a comprehensive database of validated civil unrest events using minimal human effort. AutoGSR is a deployed system for the past 6 months that is continually processing data 24X7 in an automated fashion. The system extracts civil unrest events of type "who protested where, when and why?" from news articles published in over 7 languages, and collected from 16 countries.
For more information, please visit: http://people.cs.vt.edu/parang/ or contact parang at firstname at cs vt edu
A comparative sentiment analysis of human, Gary FisherSEDA
What role can generative AI, such as ChatGPT, play in producing academic content that can be taught to students? This session explores the results of a mixed-methods study
evaluating the comparative performance of human-generated and AI-generated educational materials. Through a mixture of psycholinguistic analysis of AI- and human-generated teaching content and a quantitative survey of their impact on students, we examine the capabilities and limitations of generative AI as a tool to deliver higher education.
One fundamental problem in sentiment analysis is categorization of sentiment polarity. Given a piece of written text, the problem is to categorize the text into one specific sentiment polarity, positive or negative (or neutral). Based on the scope of the text, there are three distinctions of sentiment polarity categorization, namely the document level, the sentence level, and the entity and aspect level. Consider a review “I like multimedia features but the battery life sucks.†This sentence has a mixed emotion. The emotion regarding multimedia is positive whereas that regarding battery life is negative. Hence, it is required to extract only those opinions relevant to a particular feature (like battery life or multimedia) and classify them, instead of taking the complete sentence and the overall sentiment. In this paper, we present a novel approach to identify pattern specific expressions of opinion in text.
Design patterns are acknowledged as powerful conceptual tools to improve design quality and to reduce the time and cost of design
by effect of the reuse of “good” solutions. In many fields such as software engineering, web engineering, and interface design,
patterns are widely used by practitioners and are also investigated from a research perspective. Still, the concept of design pattern
has received marginal attention in the arena of user interfaces (UIs) for Recommender Systems (RSs). To our knowledge, a little
is known about the use of patterns in this specific class of applications, in spite of their increasing popularity, and no RS
specific interface pattern is available in existing pattern languages. We have performed a systematic analysis of 28 real-world RSs in
a variety of sectors, in order to: (i) discover occurrences of existing general (i.e., domain independent) UI patterns; (ii)
identify recurrent UI design solutions for RS specific features; (iii) elicit a set of new UI patterns for RS interfaces. The analysis
of patterns occurrences highlights the degree at which “good” UI design solutions are adopted in RSs for the different sectors. The
new patterns can be used by UI designers of RSs to improve the UX of their systems.
International Journal of Engineering Research and Applications (IJERA) is an open access online peer reviewed international journal that publishes research and review articles in the fields of Computer Science, Neural Networks, Electrical Engineering, Software Engineering, Information Technology, Mechanical Engineering, Chemical Engineering, Plastic Engineering, Food Technology, Textile Engineering, Nano Technology & science, Power Electronics, Electronics & Communication Engineering, Computational mathematics, Image processing, Civil Engineering, Structural Engineering, Environmental Engineering, VLSI Testing & Low Power VLSI Design etc.
Unlock TikTok Success with Sociocosmos..SocioCosmos
Discover how Sociocosmos can boost your TikTok presence with real followers and engagement. Achieve your social media goals today!
https://www.sociocosmos.com/product-category/tiktok/
Project Serenity is an innovative initiative aimed at transforming urban environments into sustainable, self-sufficient communities. By integrating green architecture, renewable energy, smart technology, sustainable transportation, and urban farming, Project Serenity seeks to minimize the ecological footprint of cities while enhancing residents' quality of life. Key components include energy-efficient buildings, IoT-enabled resource management, electric and autonomous transportation options, green spaces, and robust waste management systems. Emphasizing community engagement and social equity, Project Serenity aspires to serve as a global model for creating eco-friendly, livable urban spaces that harmonize modern conveniences with environmental stewardship.
Your LinkedIn Success Starts Here.......SocioCosmos
In order to make a lasting impression on your sector, SocioCosmos provides customized solutions to improve your LinkedIn profile.
https://www.sociocosmos.com/product-category/linkedin/
This tutorial presentation provides a step-by-step guide on how to use Facebook, the popular social media platform. In simple and easy-to-understand language, this presentation explains how to create a Facebook account, connect with friends and family, post updates, share photos and videos, join groups, and manage privacy settings. Whether you're new to Facebook or just need a refresher, this presentation will help you navigate the features and make the most of your Facebook experience.
Your Path to YouTube Stardom Starts HereSocioCosmos
Skyrocket your YouTube presence with Sociocosmos' proven methods. Gain real engagement and build a loyal audience. Join us now.
https://www.sociocosmos.com/product-category/youtube/
Improving Workplace Safety Performance in Malaysian SMEs: The Role of Safety ...AJHSSR Journal
ABSTRACT: In the Malaysian context, small and medium enterprises (SMEs) experience a significant
burden of workplace accidents. A consensus among scholars attributes a substantial portion of these incidents to
human factors, particularly unsafe behaviors. This study, conducted in Malaysia's northern region, specifically
targeted Safety and Health/Human Resource professionals within the manufacturing sector of SMEs. We
gathered a robust dataset comprising 107 responses through a meticulously designed self-administered
questionnaire. Employing advanced partial least squares-structural equation modeling (PLS-SEM) techniques
with SmartPLS 3.2.9, we rigorously analyzed the data to scrutinize the intricate relationship between safety
behavior and safety performance. The research findings unequivocally underscore the palpable and
consequential impact of safety behavior variables, namely safety compliance and safety participation, on
improving safety performance indicators such as accidents, injuries, and property damages. These results
strongly validate research hypotheses. Consequently, this study highlights the pivotal significance of cultivating
safety behavior among employees, particularly in resource-constrained SME settings, as an essential step toward
enhancing workplace safety performance.
KEYWORDS :Safety compliance, safety participation, safety performance, SME
The Evolution of SEO: Insights from a Leading Digital Marketing AgencyDigital Marketing Lab
Explore the latest trends in Search Engine Optimization (SEO) and discover how modern practices are transforming business visibility. This document delves into the shift from keyword optimization to user intent, highlighting key trends such as voice search optimization, artificial intelligence, mobile-first indexing, and the importance of E-A-T principles. Enhance your online presence with expert insights from Digital Marketing Lab, your partner in maximizing SEO performance.
EASY TUTORIAL OF HOW TO USE G-TEAMS BY: FEBLESS HERNANEFebless Hernane
Using Google Teams (G-Teams) is simple. Start by opening the Google Teams app on your phone or visiting the G-Teams website on your computer. Sign in with your Google account. To join a meeting, click on the link shared by the organizer or enter the meeting code in the "Join a Meeting" section. To start a meeting, click on "New Meeting" and share the link with others. You can use the chat feature to send messages and the video button to turn your camera on or off. G-Teams makes it easy to connect and collaborate with others!
Surat Digital Marketing School is created to offer a complete course that is specifically designed as per the current industry trends. Years of experience has helped us identify and understand the graduate-employee skills gap in the industry. At our school, we keep up with the pace of the industry and impart a holistic education that encompasses all the latest concepts of the Digital world so that our graduates can effortlessly integrate into the assigned roles.
This is the place where you become a Digital Marketing Expert.
Exploring The Dimensions and Dynamics of Felt Obligation: A Bibliometric Anal...AJHSSR Journal
ABSTARCT: This study presents, to our knowledge, the first bibliometric analysis focusing on the concept of
"felt obligation," examining 120 articles published between 1986 and 2024. The aim of the study is to deepen our
understanding of the existing knowledge in the field of "felt obligation" and to provide guidance for further
research. The analysis is centered around the authors, countries, institutions, and keywords of the articles. The
findings highlight prominent researchers in this field, leading universities, and influential journals. Particularly,
it is identified that China plays a leading role in "felt obligation" research. The analysis of keywords emphasizes
the thematic focuses of these studies and provides a roadmap for future research. Finally, various
recommendations are presented to deepen the knowledge in this area and promote applied research. This study
serves as a foundation to expand and advance the understanding of "felt obligation" in the field.
KEYWORDS: Felt Obligation, Bibliometric Analysis, Research Trends
Telegram is a messaging platform that ushers in a new era of communication. Available for Android, Windows, Mac, and Linux, Telegram offers simplicity, privacy, synchronization across devices, speed, and powerful features. It allows users to create their own stickers with a user-friendly editor. With robust encryption, Telegram ensures message security and even offers self-destructing messages. The platform is open, with an API and source code accessible to everyone, making it a secure and social environment where groups can accommodate up to 200,000 members. Customize your messenger experience with Telegram's expressive features.
This tutorial presentation offers a beginner-friendly guide to using THREADS, Instagram's messaging app. It covers the basics of account setup, privacy settings, and explores the core features such as close friends lists, photo and video sharing, creative tools, and status updates. With practical tips and instructions, this tutorial will empower you to use THREADS effectively and stay connected with your close friends on Instagram in a private and engaging way.
HOW TO USE THREADS an Instagram App_ by Clarissa Credito
Research on AITV
1. Unsupervised Model for Topic Viewpoint
Discovery in Online Debates Leveraging
Author Interactions
2018/11/8
SAKATA-MORI LAB M1
SYO KYOJIN
2. Summary
2
【Title and Authors】
- Unsupervised Model for Topic Viewpoint Discovery in
Online Debates Leveraging Author Interactions
- Amine Trabelsi, Osmar R. Za ̈ıane
- ICWSM 2018
【Data】
- Six datasets about four controversial issues, extracted from
4Forums.com (Abbott et al. 2016) and CreateDebate.com
(Hasan and Ng 2014).
【Contributions】
- Propose a purely unsupervised model to detect the
viewpoints at the post and author levels from the post
contents and the authors’ interactions.
- Achieve SOTA in the post level viewpoint identification and
close performance to the supervised SOTA approach in the
author level viewpoint identification.
- Discuss about author level viewpoint clustering.
3. Backgrounds
3
【Motivation】
- Recently, more and more people become able to to express their opinions on different political and social issues at
social media and online forums.
- Contentious issues : legalization of abortion, same-sex marriage and Donald Trump
- Those data are really useful for decision makers and politicians, but the amount of data is too huge. So an automatic
tool to provide a contrasting overview of the main viewpoints and reasons given by both sides is necessary.
- identify the viewpoints at the post and author’s levels
- detect the relevant discourse used to express opinion and reasoning themes.
4. Backgrounds
4
【Challenges】
- Applying sentiment analysis techniques on contentious documents is not sufficient to produce an effective solution
to the problem (Hasan and Ng 2013; Walker et al. 2012b).
- Mohammad, Sobhani, and Kiritchenko(2017) show that both positive and negative lexicons are used, in
contentious text, to express the same stance.
- Moreover, the stance can be implicitly conveyed through reasons and arguments, and not necessarily expressed
through polarity sentiment words.
- Challenges to accurate viewpoint detection can arise because of the dialogic nature of online debate (Hasan and Ng
2014; Boltuzˇic ́ and Sˇnajder 2014).
- There are many unstructured and colloquial language in the dialog of online debate i.e., containing non-
argumentative portions and irrelevant dialogs
- The similarity in words’ usage between authors leads to clustering errors when the model is entirely based on textual
features.
- This is often the case when a post rephrases the opposing side’s premise while attacking it or when asking a
rhetorical question.
- It has been shown that complementary features like the nature of the authors’ interactions at post level can
enhance pure text-based approaches in viewpoint distinguishing (Walker et al. 2012b).
5. Related Works
5
【Category】
- Evaluation series 2016 (SemEval-16) propose a shared task and datasets
for stance detection in Twitter (Mohammad et al. 2016).
- Viewpoint identification can be classified in terms of the type of
- Data : Twitter, Online Debate site
- Feature : Text, Authors’ interaction
- Task : Post level, Author level (Discourse level)
- Methods : Supervised, Unsupervised
【Supervised Learning】
- Sobhani, Inkpen, and Matwin (2015) tackle the viewpoint identification
for news comments using Topic Modeling.
- Hasan and Ng (2013) identify the stance at the post and the sentence
levels of online debates corpora. They construct a rich feature set of
linguistic and semantic features
- Sridhar et al. (2015) model disagreement and collectively predict
stances at the post and the author levels on online debate corpora.
They try different modeling approaches based on Probabilistic Soft Logic
(PSL).
【Examples of Legalization of Abortion】
<AGAINST>
Life, what a beautiful choice. Adoption. Not
abortion. #SemST
<AGAINST>
If you don't want your kid, put it up for
adoption. #sorrynotsorry #SemST
<FAVOR>
Lmao my school thinks giving women the
right to an abortion is against feminism.
They don't know feminism #wtf #SemST
<FAVOR>
@AsaSoltan putting abortion in with choices
for women just won my heart over.
Yaaaaaas! #shahs #SemST
6. Related works
6
【Unsupervised Learning】
- The most Popular approach is Topic-Viewpoint modeling, which hypothesizes the existence of underlying topic and
viewpoint variables that influence the author’s word choice. There are many different applied targets.
- Opinion polls, editorials (Paul, Zhai, and Girju 2010)
- Online forums (Trabelsi and Za ̈ıane 2016)
- Twitter (Joshi, Bhattacharyya, and Carman 2016)
- Parliamentary debates (Vilares and He 2017).
- Qiu and Jiang (2013) exploit the authors’ interactions to discover stances of posts and cluster authors with different
viewpoints. In their work, the polarity of the interactions between the authors, positive or negative, is guided and
determined using lexicon-based methods.
- Thonet et al. (2017) propose extended versions of Social Network-LDA (SN-LDA) (Sachan et al. 2014) that model the
viewpoint discovery in SNS : the Social Network Viewpoint Discovery Models (SNVDMs) in political Twitter data.
- Dong et al. (2017) focuse on a predicting the author’s stance and propose a weakly guided Stance-based Text
Generative Model with Link Regularization (STML) using the text content and the authors’ interactions in news
comments and online debates. The number of a discussion’s turns, the presence of agreement or disagreement
signals are used to guide.
7. Method
7
【Model】
- Propose Author Interaction Topic Viewpoint (AITV) Model
- For each word 𝑤 𝑑𝑎 in the post, the author draws a topic 𝑧 𝑛𝑑 from
𝜃 𝑑𝑎, then, samples each word 𝑤 𝑛𝑑 from the topic-viewpoint
distribution corresponding to topic 𝑧 𝑛𝑑 and viewpoint 𝜈 𝑑𝑎 ,
𝜙 𝑧 𝑛𝑑 𝜈 𝑑𝑎
.
- A : one author
- 𝐷 𝑎 : the number of the author’s post
- 𝑁𝑑𝑎 : the number of words in each 𝑑 𝑎
- K : the total number of topics
- L : the total number of viewpoints( = 2)
- 𝜃 𝑑𝑎 : the probability distribution of K topics under a post 𝑑 𝑎
- 𝜓 𝑎 : the probability distributions of viewpoints for an author a
- 𝜈 𝑑𝑎 : a viewpoint chosen from the distribution 𝜓 𝑎
- 𝜙 𝑎 : the multinomial probability distribution over words
associated with a topic k and a viewpoint l.
8. Method
8
【Model】
- 𝑅𝑏𝑖𝑑 : If the reply attacks the previous author then 𝑅𝑏𝑖𝑑 is set to 1
else if takes 0
- 𝑖𝑑 : a current post with index
- Parent posts : the posts which is the current post’s author is
replying to.
- 𝑅𝑏𝑖𝑑 depends on the degree of opposition between the viewpoint
𝜈𝑖𝑑 of the current post and the viewpoints 𝜈𝑖𝑑
𝑝𝑎𝑟
of its parent posts.
- where I(condition) equals 1 if the condition is true and η is a
smoothing parameter. This modeling of authors interactions is
similar to the users interactions setting presented in (Qiu and Jiang
2013).
9. Method
9
【Experiment Set Up】
- Remove identical portions of text in replying posts. We remove
stop and rare words. We consider working with the stemmed
version of the words.
- The 4Forums datasets contain the ground truth stance labels at the
author level, while the CreateDebate haven’t.
- Assign to each author the majority label of their
corresponding posts (Sridhar et al. 2015).
- All the reported results of AITV in this section correspond to
aggregation measures on 10 runs or repeats.
- The number of Topics K is 30 unless specified otherwise.
- Hyper parameters are set as follows: α = 0.1; β = 1; γ = 1; η =
0.01.
- The number of the Gibbs Sampling iterations is 1500.
- The words occurring less than 20 times are considered rare
words and are removed.
10. Results
10
【post level identification】
- AITV clearly outperforms PSL on each of the datasets.
- the best performances are recorded on the largest and
highest connected datasets (% reply posts).
- Compared AITV’s performance against its degenerate version
“AITV-Rebuttal Known” to compare AITV to a close version to
the weakly-guided work of Qiu and Jiang (2013).
- The difference of the knowledge of rebuttal is not
significant.
- This may be due to the automatic initialization based on
authors’ interactions which helps in reducing the variance
of the non-deterministic outputs, due to the Gibbs
Sampling. .
11. Results
11
【author level identification】
- AITV outperforms its rival unsupervised method SNVDM,
specifically for the datasets containing many interactions. It has
also close to comparable performance with the weakly guided
STML. However, AITV’s performance remains far from that of
the supervised PSL.
- We notice a big drop in accuracies between the post level
and the author level for AITV.
- We evaluate the two unspervised methods with the BCubed F-
Measure. the performances of AITV are overall better than
SNVDM-GPU.
12. Results
12
【author level identification】
- We evaluate the AITV’s Topic-Viewpoint clustering of the
unigram words.
- inferring the topic of the cluster by the unigram is
difficult.
- the used vocabulary may be very similar.
- In order to perform a better evaluation we choose to use
a representative sentence retrieved from a original data
by each topic-view unigram.
- Even in the same position, example 1 and example 2
are standing on different viewpoints.
- Similary, example 5 and example 7 are standing on
different viewpoints.
Propose a sharedSemEval-16ではデータ
All these described methods extensively rely on human annotations, which are expensive to obtain, and on supervision which does not guar- antee scaling to different domains and types of data.
世論調査や社説欄
Most of the Topic-Viewpoint models do not necessarily attempt to cluster the posts or their authors.
different extended versions of Social Network-LDA (SN-LDA) (Sachan et al. 2014)
ポリアのツボ
The SNVDM-GPU, the version based on Generalized Polya Urn sampling, is performing the best among all degenerate versions. It aims to overcome the sparsity of interactions.