3. Introduction
• Social media plays powerful role in today’s era .Social media provides
stage to share happiness, struggle, sentiment, stress and acquire social
support.
• Students share their happiness and sorrows related to studies on social
media in the form of judgmental comments, tweets, posts etc. Analyze
these data from such environment require classification techniques
5. Proposed System
• This approach uses totally different machine learning classifiers and
has feature extractors.
• The existing system uses classification algorithms finds only negative
emotions of students learning, whereas the proposed system will use to
discover positive as well as negative emotions.
6. Implementation
1. In the step one data collection is done from social sites.
2. Inductive Content analysis procedure is performed and categories are
identified.
3. The preprocessing is done.
4. Naive Bayes classifier, Linear SVM is applied on dataset in order to
demonstrate its application in detecting student’s issues.
7. DATA COLLECTION: Data searching is based on different Boolean
combinations of possible keywords such as engineer, students, exams,
campus, class, homework, professor, and lab.
INDUCTIVE CONTENT ANALYSIS :Because social media content
like tweets contain a large amount of informal language , sarcasm ,
acronyms, and miss-spellings, meaning is often ambiguous and subject to
human interpretation.
This topic classified the student tweets in to seven categories: heavy study
load, lack of social engagement, negative emotions, sleep problem,
diversity issues, positive things and struggle
We found that many tweets could belong to more than one category. For
example, “Why am I not in business school?? Hate being in Engineering
school. Too much stuff. Way too complicated. No fun” falls into heavy
study load, and negative emotion at the same time, So one tweet can be
labeled with multiple categories which is a multi-label classification
8. PREPROCESSING: we pre-processed the texts before training the
classifier using methods like- Stemming and stop word cleaning
NAIVE BAYES Multi-label Classification Algorithm:
The following are the basic procedures of the multi-label Naïve Bayes
classifier: The probability of word in in specific category is-
The probability of category c is -
The purpose is to classify this document into category c or not c,
Therefore, according to Bayes’Theorem, the probability that d(i) belongs
to category c is –
9. Support Vector Machine Algorithm :
Transforming a multi-label classification problem into a set of independent
binary classification problems via the “one-vs-all” scheme is a simple and
computationally efficient solution for multi-label classification.
The binary SVM training is a standard quadratic optimization problem:
10. Conclusion
The research goal of this learning are:
a) To make the enormous amount of data useful for educational purpose,
as well as to combine both qualitative analysis and large-scale data
mining techniques.
b) To examine student’s informal tweets on social media in order to
investigate the problems and issues faced by students in their life.
11. Reference
[1]. Xin Chen, Mihaela Vorvoreanu, and Krishna Madhavan, “Mining
social media data for understanding students' learning experiences”, IEEE
Transaction, 2014.
[2]. A. Go, R. Bhayani, and L. Huang, “Twitter Sentiment Classification
Using Distant Supervision,” CS224N Project Report, Stanford pp. 1-12,
2009.
[3]. E. Goffman, The Presentation of Self in Everyday Life. Lightning
Source Inc., 1959.