Analysis of Student Learning Experience by Mining Social Media Data

Abstract

Many issues like depression, suicide, anger, are increasing among students. These issues are necessary to seek out and analyze, but students never discuss their issues with anyone. Today Social media is very popular medium where individuals share their feeling and opinion. Students also terribly active on social sites like Face book and Twitter. Their unceremonious discussion on social media (e.g. Twitter, Face book) illuminates light on their educational experiences—vote, sentiment, opinions, feelings, and concerns about the learning process. Data from such environments can supply valuable information which is helpful knowledge to understand student learning experiences. Analyzing such data can be challenging. The augmenting scale of data demands automatic data analysis techniques. This paper depicts a workflow to integrate both qualitative analysis and large-scale data mining techniques. This Paper emphasized on student’s twitter posts to learn problems in student life as well as positive things occurred in their educational life. First conducted a qualitative analysis on sample tweets related to student’s college life. Students face issues such as heavy work load of study, lack of social engagement, and sleep deprivation, employment issue, etc. In this paper “positive things” happen in student’s life is also taken in to consideration. To classify tweets reflecting student’s problem multilabel classification algorithms is implemented. Naïve Bayes and Linear Support Vector Machine Learning algorithms are used. The performance of these algorithms is compared in terms of accuracy, precision, recall and F1-Measure. Support Vector Machine learning algorithm have more accuracy than Naïve Bayes Algorithm.