Explainable Recommendation for Event Sequences: A Visual Analytics Approach

Thumbnail Image


Publication or External Link






People use recommender systems to improve their decisions, for example, item recommender systems help them find films to watch or books to buy. Despite the ubiquity of item recommender systems, they can be improved by giving users greater transparency and control. This dissertation develops and assesses interactive strategies for transparency and control, as applied to event sequence recommender systems, which provide guidance in critical life choices such as medical treatments, careers decisions, and educational course selections. Event sequence recommender systems use archives of similar event sequences, such as patient histories or student academic records, to give users insight into the order and timing of choices, which are more likely to lead to their desired outcomes.

This dissertation's main contribution is the use of both record attributes and temporal event information as features to identify similar records and provide appropriate recommendations. While traditional item recommendations are generated based on choices by people with similar attributes, such as those who looked at this product or watched this movie, the event sequence recommendation approach allows users to select records that share similar attribute values and start with a similar event sequence, and then see how different choices of actions and the orders and times between them might lead to users' desired outcomes.

This dissertation applies a visual analytics approach to present and explain recommendations of event sequences. It presents a workflow for event sequence recommendation that is implemented in EventAction. Results from empirical studies show that these prototypes can assist users in making action plans and raise users' confidence in following their plans. It presents case studies in three domains to demonstrate the effectiveness and safety of generating event sequence recommendations based on personal histories. It also offers design guidelines for the construction of user interfaces for event sequence recommendation and discusses ethical issues in dealing with personal histories.

This dissertation contributes an analytical workflow, an interactive system, and design guidelines identified in empirical studies and case studies, opening new avenues of research in explainable event sequence recommendations based on personal histories. It enables people to make better decisions for critical life choices with higher confidence.