Year : 2018, Volume : 9, Issue : 1
First page : ( 21) Last page : ( 38)
Print ISSN : 2249-3212. Online ISSN : 0975-8089. Published online : 2018  1.
Article DOI : 10.5958/0975-8089.2018.00003.9

Ontology-Based Semantic Learning of Genetic-Algorithm-Optimised Back Propagation Artificial Neural Network for Personalised Web Search

Chawla Suruchi^1,*

¹Assistant Professor, Department of Computer Science, Shaheed Rajguru College of Applied Science for Women, University of Delhi, Delhi, India

*Email id: sur_chawla@rediffmail.com

Abstract

In this paper, a novel approach is proposed for training of genetic algorithm (GA), back propagation (BP) and artificial neural network (ANN) based on semantic clusters of web query sessions and used as classifier for web page recommendation. The document concept vector is generated on the basis of one Word Net ontology using word sense ambiguation, and web query session concept vectors are formed for clustering. These semantic clusters of web query sessions are used for GA-based weight optimisation of ANN and training of GA-based weight optimised BP–ANN for classification of web queries to clusters. During web search, users’ queries are expanded with related concepts based on Word Net ontology and are classified to clusters using trained GA BP–ANN for recommendation of web pages and customisation of user information search. The experimental results were verified statistically and showed the improvement in precision of recommended search results. Hence, it is concluded that semantic-based learning of GA–BP–ANN improves classification accuracy of web queries to clusters and generates the recommendations of semantically similar web pages for web search personalisation.

Top

Keywords

Information retrieval, Information scent, Search engines, Genetic algorithm, Artificial neural network, Ontology, Personalised web search.

Top