A Graph-based Relevance Matching Model for Ad-hoc Retrieval
This work addresses the challenge of improving document retrieval accuracy for users in search and information retrieval systems, representing an incremental advancement by incorporating graph-based methods to enhance existing models.
The paper tackles the problem of ad-hoc retrieval by addressing the limitations of existing deep learning models that rely on local term-level interactions, proposing a graph neural network model that leverages document-level word relationships to improve relevance matching. The approach significantly outperforms strong baselines on two benchmarks and shows advantages over BERT on long documents.
To retrieve more relevant, appropriate and useful documents given a query, finding clues about that query through the text is crucial. Recent deep learning models regard the task as a term-level matching problem, which seeks exact or similar query patterns in the document. However, we argue that they are inherently based on local interactions and do not generalise to ubiquitous, non-consecutive contextual relationships. In this work, we propose a novel relevance matching model based on graph neural networks to leverage the document-level word relationships for ad-hoc retrieval. In addition to the local interactions, we explicitly incorporate all contexts of a term through the graph-of-word text format. Matching patterns can be revealed accordingly to provide a more accurate relevance score. Our approach significantly outperforms strong baselines on two ad-hoc benchmarks. We also experimentally compare our model with BERT and show our advantages on long documents.