UIPC-MF: User-Item Prototype Connection Matrix Factorization for Explainable Collaborative Filtering

TL;DR

UIPC-MF is a prototype-based matrix factorization method that improves the explainability of collaborative filtering recommendations by associating users and items with prototypes and learning their connection weights.

Contribution

It introduces a novel prototype connection matrix factorization approach that enhances explainability without sacrificing recommendation accuracy.

Findings

Outperforms baseline methods on Hit Ratio and NDCG

Provides better transparency in recommendations

Effective on multiple datasets

Abstract

Recommending items to potentially interested users has been an important commercial task that faces two main challenges: accuracy and explainability. While most collaborative filtering models rely on statistical computations on a large scale of interaction data between users and items and can achieve high performance, they often lack clear explanatory power. We propose UIPC-MF, a prototype-based matrix factorization method for explainable collaborative filtering recommendations. In UIPC-MF, both users and items are associated with sets of prototypes, capturing general collaborative attributes. To enhance explainability, UIPC-MF learns connection weights that reflect the associative relations between user and item prototypes for recommendations. UIPC-MF outperforms other prototype-based baseline methods in terms of Hit Ratio and Normalized Discounted Cumulative Gain on three datasets, while also providing better transparency.