The importance of being dissimilar in Recommendation

TL;DR

This paper emphasizes the importance of incorporating dissimilarity alongside similarity in recommendation systems, demonstrating that this approach improves accuracy in memory-based nearest neighbors methods.

Contribution

It introduces a formal model for integrating dissimilarity with similarity measures in recommendation algorithms, enhancing their effectiveness.

Findings

Increased recommendation accuracy with dissimilarity integration

Formal modeling of dissimilarity in similarity measures

Effective improvement demonstrated through evaluation

Abstract

Similarity measures play a fundamental role in memory-based nearest neighbors approaches. They recommend items to a user based on the similarity of either items or users in a neighborhood. In this paper we argue that, although it keeps a leading importance in computing recommendations, similarity between users or items should be paired with a value of dissimilarity (computed not just as the complement of the similarity one). We formally modeled and injected this notion in some of the most used similarity measures and evaluated our approach showing its effectiveness in terms of accuracy results.