A Deep Relevance Matching Model for Ad-hoc Retrieval

TLDR

Deep neural networks have achieved breakthroughs in many domains, yet few successes have been reported for ad‑hoc retrieval because its unique relevance‑matching requirements are not well addressed by existing deep models. We argue that ad‑hoc retrieval is fundamentally a relevance‑matching task distinct from semantic‑matching NLP tasks, and aim to address this gap. DRMM is a joint deep architecture that maps query–document term matches into histograms, feeds them through a matching network, and applies a term‑gating network to capture exact matches, term importance, and diverse relevance signals. On two benchmark collections, DRMM significantly outperforms well‑known retrieval models and state‑of‑the‑art deep matching models.

Abstract

In recent years, deep neural networks have led to exciting breakthroughs in speech recognition, computer vision, and natural language processing (NLP) tasks. However, there have been few positive results of deep models on ad-hoc retrieval tasks. This is partially due to the fact that many important characteristics of the ad-hoc retrieval task have not been well addressed in deep models yet. Typically, the ad-hoc retrieval task is formalized as a matching problem between two pieces of text in existing work using deep models, and treated equivalent to many NLP tasks such as paraphrase identification, question answering and automatic conversation. However, we argue that the ad-hoc retrieval task is mainly about relevance matching while most NLP matching tasks concern semantic matching, and there are some fundamental differences between these two matching tasks. Successful relevance matching requires proper handling of the exact matching signals, query term importance, and diverse matching requirements. In this paper, we propose a novel deep relevance matching model (DRMM) for ad-hoc retrieval. Specifically, our model employs a joint deep architecture at the query term level for relevance matching. By using matching histogram mapping, a feed forward matching network, and a term gating network, we can effectively deal with the three relevance matching factors mentioned above. Experimental results on two representative benchmark collections show that our model can significantly outperform some well-known retrieval models as well as state-of-the-art deep matching models.