Improved memetic algorithm enabled intelligent multi agent (IMAEIMA) system for web mining

Abstract

Wider web space, the searching of a relevant data is the most curious problem for the common people accessing the web. For retreving the relevant information the user request is given to search engine. The relevant pages combined with irrelevant pages are returned to the user. The proposed work emphasizes an Improved Memetic Algorithm Enabled Intelligent Multi Agent (IMAEIMA) for searching the most appropriate pages when submitting complex queries. Improved Memetic algorithm is the traditional genetic algorithm combined with local search and random selection. In this proposed system Improved Memetic algorithm additionally enhanced with logarithmic weight function for more accuracy. Intelligent Agents are introduced in this IMAEIMA to improve its performance and accuracy by reacting intelligently based on feedback and previous experience. This system helps to retrieve relevant pages from web databases with high precision and recall. The derived architecture reveals greater precision and recall overriding the conventional search algorithms.