Topic-Level Bayesian Surprise and Serendipity for Recommender Systems

TL;DR

This paper introduces a Bayesian surprise-based measure of serendipity for recommender systems, improving the recommendation of novel, surprising items and reducing filter bubble effects.

Contribution

It proposes a content-based, Bayesian surprise model for measuring serendipity and introduces a new dataset with manual annotations for evaluating surprise at the topic level.

Findings

Bayesian surprise correlates better with manual surprise annotations than heuristics.

Models using Bayesian surprise achieve higher serendipity in recommendations.

The new dataset enables effective evaluation of surprise and serendipity in recommender systems.

Abstract

A recommender system that optimizes its recommendations solely to fit a user's history of ratings for consumed items can create a filter bubble, wherein the user does not get to experience items from novel, unseen categories. One approach to mitigate this undesired behavior is to recommend items with high potential for serendipity, namely surprising items that are likely to be highly rated. In this paper, we propose a content-based formulation of serendipity that is rooted in Bayesian surprise and use it to measure the serendipity of items after they are consumed and rated by the user. When coupled with a collaborative-filtering component that identifies similar users, this enables recommending items with high potential for serendipity. To facilitate the evaluation of topic-level models for surprise and serendipity, we introduce a dataset of book reading histories extracted from Goodreads, containing over 26 thousand users and close to 1.3 million books, where we manually annotate 449 books read by 4 users in terms of their time-dependent, topic-level surprise. Experimental evaluations show that models that use Bayesian surprise correlate much better with the manual annotations of topic-level surprise than distance-based heuristics, and also obtain better serendipitous item recommendation performance.