Résultats de recherche pour « Query expansion techniques »

Abstract In Information Retrieval, query expansion refers to the procedure ofreformulating a user's search query by adding further search terms to theoriginal query to improve retrieval quality. Short queries, specifically, oftensuffer from ambiguity, i.e., they refer to multiple possible senses. Theobjective of query expansion, then, is to augment with terms that best match theuser's information need. In this work, we propose and analyse a system forsemantic query expansion that leverages contextual information extracted fromuser profiles to generate queries capable of disambiguating multi-sense searchqueries. We generate synthetic profiles from a knowledge graph of Wikipediaarticles, on which we train individual word2vec embeddings. Using a k-nearestneighbour search, we determine semantically close keywords to append to a givenquery. We evaluate this user-centric approach across corpora of varying sizes,assessing the extent to which sentinel articles are retrieved. Our findingssuggest that tailoring the expansion to the user's context results in animprovement of median rank by 2 levels, and an increase in category retrievalamong the top three results of 6% on average.

The self-supervised training method can fully leverage the structural characteristics of unlabeled datasets, enhancing the technical prowess of natural language processing. We explore the application of self-supervision’s advantages to query expansion. To achieve this, we amassed a substantial query corpus and trained it using a generative model. Simultaneously, we introduced the SG-BERT/GPT2 extension method, leveraging two generators, BERT and GPT2. We also conducted an analysis of the impact on code search before and after generative model training, as well as the influence of the original query length on the expanded model. Experimental results consistently demonstrate the superiority of the trained model over the untrained one. Additionally, in deep learning-based code search methods, query expansion proves most effective for datasets with relatively short original query lengths, while longer expansions yield the opposite effect. Finally, when compared to various established query expansion methods, the SG-BERT/GPT2 expansion method, rooted in the self-supervised generative model, consistently outperforms them across multiple language test datasets. With ongoing self-supervised generation and continuous optimization of the query corpus, further enhancements in effectiveness can be achieved.

<p>The self-supervised training method can fully leverage the structural characteristics of unlabeled datasets, enhancing the technical prowess of natural language processing. We explore the application of self-supervision’s advantages to query expansion. To achieve this, we amassed a substantial query corpus and trained it using a generative model. Simultaneously, we introduced the SG-BERT/GPT2 extension method, leveraging two generators, BERT and GPT2. We also conducted an analysis of the impact on code search before and after generative model training, as well as the influence of the original query length on the expanded model. Experimental results consistently demonstrate the superiority of the trained model over the untrained one. Additionally, in deep learning-based code search methods, query expansion proves most effective for datasets with relatively short original query lengths, while longer expansions yield the opposite effect. Finally, when compared to various established query expansion methods, the SG-BERT/GPT2 expansion method, rooted in the self-supervised generative model, consistently outperforms them across multiple language test datasets. With ongoing self-supervised generation and continuous optimization of the query corpus, further enhancements in effectiveness can be achieved.</p>