Structural Learning of Diverse Ranking

TL;DR

This paper introduces a unified structural learning framework that simultaneously optimizes relevance and diversity in search results by directly maximizing diversity-correlated evaluation measures, leading to improved performance.

Contribution

The paper presents a novel bi-criteria learning framework that directly optimizes diversity measures without relying on explicit subtopic information, enhancing adaptability and flexibility.

Findings

Significantly outperforms existing diversification methods on TREC datasets

Effectively balances relevance and diversity in search results

Flexible feature-based approach improves diversity optimization

Abstract

Relevance and diversity are both crucial criteria for an effective search system. In this paper, we propose a unified learning framework for simultaneously optimizing both relevance and diversity. Specifically, the problem is formalized as a structural learning framework optimizing Diversity-Correlated Evaluation Measures (DCEM), such as ERR-IA, a-NDCG and NRBP. Within this framework, the discriminant function is defined to be a bi-criteria objective maximizing the sum of the relevance scores and dissimilarities (or diversity) among the documents. Relevance and diversity features are utilized to define the relevance scores and dissimilarities, respectively. Compared with traditional methods, the advantages of our approach lie in that: (1) Directly optimizing DCEM as the loss function is more fundamental for the task; (2) Our framework does not rely on explicit diversity information such as subtopics, thus is more adaptive to real application; (3) The representation of diversity as the feature-based scoring function is more flexible to incorporate rich diversity-based features into the learning framework. Extensive experiments on the public TREC datasets show that our approach significantly outperforms state-of-the-art diversification approaches, which validate the above advantages.