Open AccessComparing the Performance of SOM with Traditional Methods for Document Clustering Using Wordnet Ontologies
Author(s) -
Abhishek Sawalkar,
Mohit Mandlecha,
Dnyanesh Kulkarni,
Ratnamala S. Paswan
Publication year - 2022
Publication title -
international journal for research in applied science and engineering technology
Language(s) - English
Resource type - Journals
ISSN - 2321-9653
DOI - 10.22214/ijraset.2022.41554
Subject(s) - wordnet , cluster analysis , document clustering , brown clustering , computer science , information retrieval , hierarchical clustering , correlation clustering , relevance (law) , fuzzy clustering , clustering high dimensional data , data mining , artificial intelligence , cure data clustering algorithm , political science , law
Retrieving useful information has become challenging due to the rapid expansion of web material. To improve the retrieval outcomes, efficient clustering methods are required. Document clustering is the process of identifying similarities and differences among given objects and grouping them into clusters with comparable features. We used WordNet lexical as an addition to compare several document clustering techniques in this article. The suggested method employs WordNet to determine the relevance of the concepts in the text, and then clusters the content using several document clustering algorithms (K-means, Agglomerative Clustering, and self-organizing maps). We wish to compare alternative ways for making document clustering algorithms more successful. Keywords: Document clustering, Clustering technique, Self-organizing maps, WordNet, K-means, Hierarchical Clustering.