An efficient document information retrieval using hybrid global search optimization algorithm with density based clustering technique

Bhushan Inje

doi:10.1007/s10586-023-03976-1

Return

An efficient document information retrieval using hybrid global search optimization algorithm with density based clustering technique

2023-02-08

PRE

Bhushan Inje *

Kapil Kumar Nagwanshi

Radha Krishna Rambola

DOI：10.1007/s10586-023-03976-1

Original

Original request for help

Save

Abstract

En 中文

Due to the increased size of data, there is a need for retrieving the right document for the user efficiently, which finds various applications in the research community. In this work, we propose Hybrid Global Search Optimization with Density based clustering (HGSODC) that extend the current state of the art, which is mostly based on searching a document from closed frequent terms to bring efficient result by alleviating convergence problem. Firstly, the documents are preprocessed by removing stop words, stemming, and then grouped using hierarchical density-based spatial clustering of applications with noise (HDBSCAN) clustering, and then closed frequent patterns mining is performed at each document. Secondly, the search is done using the HGSOA algorithm, and the documents are retrieved. We determine the effectiveness of the HGSODC approach through a set of experiments under the NPL, LISA, and CACM corpus. Compared to some existing related work, a wide range of evaluations are provided to show the strength of the proposed method in terms of precision, recall, MAP, F-score, accuracy, and convergence rate by running multiple experiments to compare our approaches with different baselines. The results indicate that the proposed HGSODC approach outperforms the traditional document information retrieval methods based on returned document quality and running time.

Keywords:

Optimization

Clustering

Preprocessing

Pattern mining

Document retrieval