Efficient Document Retrieval System using Locality Sensitive Hashing Nearest Neighbor Algorithm and Weighted Jaccard Distance for Retrieving Closest Personalities

Abstract

The process of retrieving significant documents based on the search key from a corpus has been a vital research problem in the information retrieval field. This paper proposes an efficient way to retrieve documents related to different personalities extracted from Wikipedia. The proposed method utilizes the Locality Sensitive Hashing Nearest Neighbor algorithm combined with Weighted Jaccard Distance to measure document similarity with enhanced precision. This document retrieval system demonstrates competitive performance compared to existing methods in the Personality Identification domain. The introduction of a document centroid normalization technique significantly improves the effectiveness of information retrieval by enabling better discrimination between documents. The personality document search results were compared for different distance measures using performance metrics like Normalized Discounted Cumulative Gain and Mean Average Precision. The results presented in this paper show that the TF-IDF scheme with Locality Sensitive Hashing Nearest Neighbor Algorithm using the Weighted Jaccard Distance can yield superior retrieval efficiency when contrasted with alternative approaches found in the existing literature.