Query Expansion with Locally-Trained Word Embeddings
arXiv:1605.07891
Abstract
Continuous space word embeddings have received a great deal of attention in the natural language processing and machine learning communities for their ability to model term similarity and other relationships. We study the use of term relatedness in the context of query expansion for ad hoc information retrieval. We demonstrate that word embeddings such as word2vec and GloVe, when trained globally, underperform corpus and query specific embeddings for retrieval tasks. These results suggest that other tasks benefiting from global embeddings may also benefit from local embeddings.
References in corpus (1)
Cited by in corpus (20)
- Semantic Models for the First-stage Retrieval: A Comprehensive Review
- Neural Models for Information Retrieval
- Neural Information Retrieval: A Literature Review
- Firsthand Opiates Abuse on Social Media: Monitoring Geospatial Patterns of Interest Through a Digital Cohort
- Match-Tensor: a Deep Relevance Model for Search
- Semantic Product Search
- Learning to Match Using Local and Distributed Representations of Text for Web Search
- Expert Finding in Heterogeneous Bibliographic Networks with Locally-trained Embeddings
- A Fast Deep Learning Model for Textual Relevance in Biomedical Information Retrieval
- Word Embeddings for the Construction Domain
- NPRF: A Neural Pseudo Relevance Feedback Framework for Ad-hoc Information Retrieval
- A Query-Driven Topic Model
- Toward Incorporation of Relevant Documents in word2vec
- Query Expansion for Cross-Language Question Re-Ranking
- Getting Started with Neural Models for Semantic Matching in Web Search
- Utilizing Embeddings for Ad-hoc Retrieval by Document-to-document Similarity
- RelEmb: A relevance-based application embedding for Mobile App retrieval and categorization
- QA4PRF: A Question Answering based Framework for Pseudo Relevance Feedback
- Event-Driven Query Expansion
- Retrieving Multi-Entity Associations: An Evaluation of Combination Modes for Word Embeddings