Abstract: Social media sites such as Twitter, Facebook, You-tube are very popular sites in higher educational student’s like engineering, medical, pharmacy, trainees and other than students too. It is a platform where one can share their ideas, views and discuss experiences with others in a formal and casual manner. Now a day’s gregarious media provides opportunities for understanding human behavior from the large aggregate data sets that their operation collects. Data Mining is very useful in the field of education, especially while examining students’ learning behavior. Student’s informal discussion on social media (Twitter) elucidates their educational experiences, opinions, feelings, mind-set and concerns about the cognition process. Data from such un instrumented environments can provide valuable knowledge to apprise student problem. Examining such data, however, can be arduous. The problem of students’ experiences reflected from social media content requires human interpretation. However, the growing scale of data demands automatic data analysis techniques. In this paper we propose a workflow to bridge together both qualitative analysis and large-scale data mining techniques. We fixated on engineering students Twitter posts to understand issues and quandaries in their learning experiences. First a sample is taken from student’s tweets and then qualitative analysis is conducted on the sample which is associated to engineering student’s educational life. It is found that engineering students encounter problems such as heavy learning load, lack of social meeting, sleep deficiency etc. Based on this outcome, Naive Bayes Multi-label Classifier algorithm is applied to categorize tweets presenting student’s problems. This study presents a methodology and result that demonstrate how casual social media data can provide insight into student’s experiences.

Keywords: Data mining, Social networking, Web text analysis, Naïve bayes, Computer and education.