MOOC Data Analytics: Social Network Analysis of Discussion Forum Data

The Hong Kong University of Science and Technology
Department of Computer Science and Engineering

Final Year Thesis Oral Presentation

Title: "MOOC Data Analytics: Social Network Analysis of Discussion Forum Data"

By

Lanxiao XU

Abstract

As massive open online course (MOOC) has drawn great attention by offering 
simulated real world learning experience, there is a surge of research on 
MOOC experiences. Among the diverse services provided on MOOC websites, 
forums have great research potential due to their rich content and the 
complex social networks hidden behind forum posts. In this project, machine 
learning techniques are used to reveal patterns of user forum behavior from 
MOOC forum social network, where users are grouped into different clusters 
by different forum participation styles. During the course period, grouping 
of users is not static and changes as more users participate in forum 
posting. These dynamic forum user patterns acquired can serve as an 
indicator to explain and even predict other types of MOOC user activities 
such as dropout behavior. As the number of active users often tends to 
decrease greatly during the whole course period for many MOOC courses, it 
is crucial to discover future dropouts so that actions can be take to keep 
users active. This project utilize the user patterns discovered in MOOC 
forum data and combine them with some other features describing user 
engagements to achieve the goal of dropouts prediction.

Date:                   Tuesday, 28 April 2015

Time:                   10:30 - 11:10am

Venue:                  Room 5560
                        Lifts 27/28

Committee Members:      Prof. Dit-Yan Yeung (Supervisor)
                        Dr. Raymond Wong (Reader)