Mining Behavioral Patterns from Mobile Big Data

PhD Thesis Proposal Defence


Title: "Mining Behavioral Patterns from Mobile Big Data"

by

Mr. Tong LI


Abstract:

The widespread usage of Internet-connected mobile devices allows users' 
behavior to be recorded in large-scale and fine-grained digital traces in 
cyberspace. Such data, termed mobile big data, contain great social, 
economic, and academic values. Analyzing mobile big data has significant 
implications for all relevant stakeholders, ranging from smartphone 
manufacturers, network operators to app developers. In this thesis 
proposal, we propose three works towards an in-depth understanding and 
modeling users' behavior from mobile big data by leveraging large-scale 
and real-world datasets.

Firstly, we investigate daily activity patterns across people's mobile app 
usage based on a real-world dataset consisting of 653,092 users with 
971,818,946 mobile app usage records. We then discover five common 
patterns, including afternoon reading, nightly entertainment, pervasive 
socializing, commuting, and nightly socializing. We also show that people 
usually follow yesterday's activity patterns, but the patterns tend to 
deviate as the time-lapse increases. Secondly, we demonstrate that users' 
personality traits profoundly shape their digital behavior. We develop a 
multi-relational heterogeneous graph attention network (MRel-HGAN) to 
predict users' gender using spatiotemporal app usage behavior. By 
conducting extensive experiments on a real-world dataset covering 12,777 
users with gender labels, MRel-HGAN achieves a precision of 73.71%, 
outperforming the best baseline by 4.4%. Lastly, we reveal the long-term 
evolution process of mobile app usage by conducting a longitudinal study 
on 1,465 users from 2012 to 2017. Our findings indicate that users' app 
usage indeed changes over time. However, the evolution processes in 
app-category usage and individual app usage are different in terms of 
popularity distribution, usage diversity, and correlations.

In the end, we conclude this thesis proposal with future research 
directions and challenges related to mobile big data analysis.


Date:			Monday, 22 February 2021

Time:                  	4:00pm - 6:00pm

Zoom Meeing: 
https://hkust.zoom.us/j/92011528543?pwd=bTBDVjJFblRCSCtDWWl4dm1uRjJIZz09

Committee Members:	Dr. Pan Hui (Supervisor)
 			Prof. Raymond Wong (Chairperson)
 			Prof. James Kwok
 			Dr. Dimitris Chatzopoulos


**** ALL are Welcome ****