Incorporating Deep Median Networks for Arabic Document Retrieval Using Word Embeddings-Based Query Expansion

The information retrieval (IR) process often encounters a challenge known as query-document vocabulary mismatch, where user queries do not align with document content, impacting search effectiveness. Automatic query expansion (AQE) techniques aim to mitigate this issue by augmenting user queries wit...

Full description

Saved in:
Bibliographic Details
Published in:Journal of information science theory and practice 2024-09, Vol.12 (3), p.36
Main Authors: Farhan, Yasir Hadi, Shakir, Mohanaad, Tareq, Mustafa Abd, Shannaq, Boumedyen
Format: Article
Language:English
Subjects:
arabic document retrieval
Arabic language
automatic query expansion
deep median networks
Information retrieval
natural language processing
Vocabularies & taxonomies
word embedding
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The information retrieval (IR) process often encounters a challenge known as query-document vocabulary mismatch, where user queries do not align with document content, impacting search effectiveness. Automatic query expansion (AQE) techniques aim to mitigate this issue by augmenting user queries with related terms or synonyms. Word embedding, particularly Word2Vec, has gained prominence for AQE due to its ability to represent words as real-number vectors. However, AQE methods typically expand individual query terms, potentially leading to query drift if not carefully selected. To address this, researchers propose utilizing median vectors derived from deep median networks to capture query similarity comprehensively. Integrating median vectors into candidate term generation and combining them with the BM25 probabilistic model and two IR strategies (EQE1 and V2Q) yields promising results, outperforming baseline methods in experimental settings.
ISSN:2287-9099
2287-4577
DOI:10.1633/JISTaP.2024.12.3.3