Collaborative Filtering by Personality Diagnosis: A Hybrid Memory- and Model-Based Approach

TL;DR

This paper introduces personality diagnosis, a hybrid collaborative filtering method that combines memory-based and model-based approaches, offering probabilistic predictions and easy incremental updates, demonstrated on movie ratings and research paper data.

Contribution

The paper presents a new probabilistic collaborative filtering method called personality diagnosis that improves recommendation accuracy and interpretability.

Findings

Effective on movie rating data with high accuracy.

Applicable to user profile data from digital libraries.

Supports value of information computations for better insights.

Abstract

The growth of Internet commerce has stimulated the use of collaborative filtering (CF) algorithms as recommender systems. Such systems leverage knowledge about the known preferences of multiple users to recommend items of interest to other users. CF methods have been harnessed to make recommendations about such items as web pages, movies, books, and toys. Researchers have proposed and evaluated many approaches for generating recommendations. We describe and evaluate a new method called emph{personality diagnosis (PD)}. Given a user's preferences for some items, we compute the probability that he or she is of the same "personality type" as other users, and, in turn, the probability that he or she will like new items. PD retains some of the advantages of traditional similarity-weighting techniques in that all data is brought to bear on each prediction and new data can be added easily and incrementally. Additionally, PD has a meaningful probabilistic interpretation, which may be leveraged to justify, explain, and augment results. We report empirical results on the EachMovie database of movie ratings, and on user profile data collected from the CiteSeer digital library of Computer Science research papers. The probabilistic framework naturally supports a variety of descriptive measurements - in particular, we consider the applicability of a value of information (VOI) computation.