Utilizing Embeddings for Ad-hoc Retrieval by Document-to-document Similarity

Abstract

Latent semantic representations of words or paragraphs, namely the embeddings, have been widely applied to information retrieval (IR). One of the common approaches of utilizing embeddings for IR is to estimate the document-to-query (D2Q) similarity in their embeddings. As words with similar syntactic usage are usually very close to each other in the embeddings space, although they are not semantically similar, the D2Q similarity approach may suffer from the problem of "multiple degrees of similarity". To this end, this paper proposes a novel approach that estimates a semantic relevance score (SEM) based on document-to-document (D2D) similarity of embeddings. As Word or Para2Vec generates embeddings by the context of words/paragraphs, the D2D similarity approach turns the task of document ranking into the estimation of similarity between content within different documents. Experimental results on standard TREC test collections show that our proposed approach outperforms strong baselines.

References in corpus (8)

• word2vec Explained: deriving Mikolov et al.'s negative-sampling word-embedding method
• Document Embedding with Paragraph Vectors
• A Dual Embedding Space Model for Document Ranking
• Using Word Embeddings for Automatic Query Expansion
• Neural Models for Information Retrieval
• Query Expansion with Locally-Trained Word Embeddings
• Semantic Regularities in Document Representations
• Uncertainty in Neural Network Word Embedding: Exploration of Threshold for Similarity

Cited by in corpus (1)

• Study of keyword extraction techniques for Electric Double Layer Capacitor domain using text similarity indexes: An experimental analysis