Improving web search user query relevance using content based page rank

Abstract

World Wide Web is a huge data and information repository that contains the different formats and different aspects of data. User consumes the search engines to find the data according to their own aspects. But during search the search engines returns a significant amount of search results. Thus the results listing according to the user need is required. Therefore the search engines utilize the page rank algorithms to analyse and rank the search results according to the user query relevancy by estimating the web page importance over the web. That is a traditional kind of model which utilizes the mathematical model of web graph to estimate the importance of web page. In order to provide the user query relevance web ranks the proposed work investigates the web page ranking methods and recently developed improvements on web page ranking. In addition of that a new content based web page rank technique is also proposed for implementation. The proposed technique finds the importance of the web page by evaluating the user search query and the available contents on web pages. Thus this technique implements the K-mean clustering technique for data categorization and also utilizes the random walk theory for content ranking. The implementation of the proposed page rank estimation technique is given using JAVA based technology and the performance of the algorithms is evaluated in terms of their computational resource need and their search results relevancy. The results demonstrate the effectiveness of the proposed page rank technique and the efficiency. Thus the proposed page rank algorithm is much adoptable for query specific scenarios for page ranking.