Simultaneous Relevance and Diversity: A New Recommendation Inference Approach

TL;DR

This paper introduces a novel heterogeneous inference method for recommender systems that simultaneously enhances relevance and diversity by enabling them to support each other within a unified model, outperforming existing approaches.

Contribution

The paper presents a new collaborative filtering inference technique, negative-to-positive, that inherently balances relevance and diversity as collaborative objectives.

Findings

Outperforms existing methods on relevance and diversity metrics

Applicable to various recommendation scenarios and levels of system complexity

Validated on public datasets and real-world production data

Abstract

Relevance and diversity are both important to the success of recommender systems, as they help users to discover from a large pool of items a compact set of candidates that are not only interesting but exploratory as well. The challenge is that relevance and diversity usually act as two competing objectives in conventional recommender systems, which necessities the classic trade-off between exploitation and exploration. Traditionally, higher diversity often means sacrifice on relevance and vice versa. We propose a new approach, heterogeneous inference, which extends the general collaborative filtering (CF) by introducing a new way of CF inference, negative-to-positive. Heterogeneous inference achieves divergent relevance, where relevance and diversity support each other as two collaborating objectives in one recommendation model, and where recommendation diversity is an inherent outcome of the relevance inference process. Benefiting from its succinctness and flexibility, our approach is applicable to a wide range of recommendation scenarios/use-cases at various sophistication levels. Our analysis and experiments on public datasets and real-world production data show that our approach outperforms existing methods on relevance and diversity simultaneously.