A Comparative Analysis Of Web Page Personalization Model Using Various Optimization Techniques

Abstract

This study synthesizes ideas from three independent investigations, each bringing unique techniques to promote web page personalization in the dynamic world of information technology. The overall theme in these works is the use of clustering techniques, notably the Weighted Clustering (WC) algorithm, in conjunction with optimization methods to enhance the personalization process. The first study uses WC to cluster web pages based on domains, with a user learning module that adapts to individual preferences. Important parts of this method include using the Word Net ontology for query formulation and profiling. When compared to existing methods, the Oppositional based Fire Fly Optimization (OFFO) approach improves clustering results by displaying greater precision (89.16%), recall (78.09%), and f-measure (83.26%). The second technique refines clustering by introducing an inventive dimension using the Improved Whale Optimization Algorithm (WOA) and a tumbling effect. This study proves the WOA model's efficacy in handling an increasing volume of web documents and efficiently delivering appropriate search results. The third method, known as NM-AFO, proposes the Weighted Clustering with Nelder Mead (NM) based Artificial Flora Optimization (AFO) Algorithm, which combines clustering techniques with similarity measurements. NM-AFO refines WC results using the Nelder Mead optimization algorithm, ensuring an efficient personalization process. The experimental analyses in these research together highlight NM-AFO's superior performance, emphasizing its effectiveness in resolving user queries and giving customized web page recommendations. This comparative synthesis emphasizes the advantages of each strategy, providing a thorough view of the growing landscape of web page personalization research.