Research On Web Page Rank Based On Improved Page Rank Algorithm

With the rapid development of the Internet information technology today, it seems that overnight, big data(Big Data) becomes one of the most popular vocabulary. Although users can obtain all kinds of information by using search engines conveniently, they are also faced with the problem of how to remove those redundancy data who impact the information retrieval efficiency and accuracy from the mass storage of information. In general, users only concern the previous pages of the target pages, so how to improve search quality and users' satisfaction become particularly important by the search results page ranking. For web pages, in addition to the text message, the link structure of pages is also an important path for users to acquire the helpful message.The classicial web page rank algorithm---Page Rank just uses the link structure of web pages to iterativly calculate the weights of each page, which can greatly improve the accuracy of the page weight calculation to a large extent. However, the Page Rank algorithm still has many problems to be studied and to be broken, so it has important application value to carry out related research.This paper firstly describes the research background and significance of the Page Rank algorithm based on Map Reduce, and then analyzes and summarizes the research status at home and abroad. On this basis, the key factors that affect the performance and accuracy of the algorithm are analyzed, that is, the number of iterations and the “ theme drift ”, and then propose an improved algorithm: sub graph estimation Page Rank page ranking algorithm and personalized intelligent recommendation weight distribution method.Then the improved algorithm has been theoretically analyzed, including the number of iterations of the algorithm, the time complexity and accuracy. Finally, the Page Rank algorithm and its improved algorithm are implemented on the Map Reduce programming model, and the rationality and validity of the algorithm are proved by the comparative analysis of the experimental data. Compared with the traditional algorithm, the time complexity of the improved algorithm is low, the number of iterations is small, and the accuracy is high.