Interpreting Students Learning Behavior And Issues Using Social Media Data Mining

Abstract

Students' casual discussions on on-line networking (e.g., Twitter, Face book) shed lightweight into their
academic experiences. Their opinions, emotions, and considerations regarding the training method. Info from such
instrumented environmental will provide valuable data to tell students learning. Anal zing down such information, on the
opposite hand, is difficult. The complexness of students' experiences mirrored from social media substance needs human
translation. Then again, the growing size of information demands automatic data analysis strategies. In this paper, we
tend to developed work method to integrate each chemical analysis and large-scale data processing. We tend to targeting
engineering students' Twitter post to grasp drawback and issue in their academic experiences. We tend to at first
conducted a chemical analysis on samples taken from around 25,000 tweets associated with engineering students’ life.
We tend to discovered that engineering students expertise problems, like overwhelming study burden, absence of social
engagement, and lack of sleep. In lightweight of those results, we tend to enforce a multi-label classification rule to
classify tweets reflective students’ problems. We then used the rule to coach a detector of scholars problems from
around thirty five,000 tweets streamed at the geo-area of Purdue University. This work, astonishingly initial time,
introduces associate degree approach and results that show however informal social media information will give insights
into of students’ experiences.