Location-Based Social Networks (LBSNs), with the most popular being Yelp and Foursquare, have become a vital part of our society due to their assistance on users needs. For example, foreign tourists in San Fransisco can easily find highly-reputed restaurants even if it is their first time in the city. As LBSNs have become widely used by users, understanding user engagement and predicting user churn are essential to the maintainability of the services. Many researchers largely focus on churn prediction problem to predict users who are about to leave the services. Although previous studies achieve high performance by adopting various ML techniques and DL approaches, we are still far from explaining why those users churn from the services. In other words, the interpretability of the models is limited in prior works.
Can we develop a new method or algorithm in order to enable Interpretable Churn? Interpretable churn would allow us to provide human-readable explanations to the service providers, maintainers, or governmental decision-makers for their various services of interest.
Available Datasets: (1) Yelp, (2) Foursquare
Reference papers (can be a good starting point)
 2018. WWW. Ill Be Back: On the Multiple Lives of Users of a Mobile Activity Tracking Application. Z Lin et al.
 2019. IMWUT. GeoLifecycle: User Engagement of Geographical Exploration and Churn Prediction in LBSNs. YD Kwon et al.
1. You will need to be self-motivated and proactive, which means you should be able to set your own milestones and finish them on time.
2. Strong interest in analyzing data and finding common or new patterns from it
3. Data, network analytics tools (SNAP, Gephi, NetworkX) is a plus
1. Opportunity to gain hands-on experience in Social Network Analysis and Machine Learning research
2. Opportunity to tackle a challenging and novel problem that no one has solved before
3. Opportunity for having a top-tier academic publication, depending on the quality and the amount of novel contribution of the project