Machine Learning Frameworks for Intelligent Information Retrieval: From Query Suggestions to Scientific Literature Curation

Main Article Content

Md Jahid Alam Riad

Abstract

This paper explores the integration of Machine Learning (ML) frameworks in Intelligent Information Retrieval (IR) systems, focusing on query suggestions, document retrieval, and scientific literature curation. We propose a hybrid approach combining ontology-based systems and deep learning models such as BERT and Word2Vec to enhance semantic understanding in IR tasks. Experimental results show that our ML-based models outperform traditional systems like BM25, providing improved precision, recall, and NDCG scores in document retrieval. Additionally, MLdriven clustering and recommendation systems demonstrate better performance in literature curation, offering more accurate classifications and user-centric suggestions. Our findings highlight the potential of ML-enhanced IR systems in addressing the limitations of traditional keyword-based approaches, enabling more effective retrieval and curation of scientific literature.

Article Details

Section
Articles