IMSS-P: An intelligent approach to design & development of personalized meta search & page ranking system

Abstract

The proposed research work aims to discuss and explore various constraints of traditional web page search and ranking systems primarily in the present generation of big data. The primary objective is to facilitate a web user by presenting a most personalized web page ranking as a response to a user’s search query by considering tastes and browsing history of the user while previously searching on the web. This research intends to design and develop a machine learning based next generation of web page ranking algorithm, i.e., Advanced Cluster Vector Page Ranking algorithm (ACVPR). This ACVPR algorithm is implemented in the form of an Intelligent Mata Search System-Personalized tool to evaluate the performance of the algorithm. The ACVPR algorithm arm the user with a powerful meta-search tool to facilitate the user by providing a web page ranking order to quickly satisfy the personalized needs especially when the search query is erroneous or incomplete. An extensive mathematical and experimental evaluation of the developed logistic regression model by calculating and comparing various evaluation metrics such as specificity, sensitivity, precision, recall using R statistical tool shows the improved efficiency as compared to other popular search engines.