A Hybrid Page Rank Algorithm for Web Pages

In the past, the World Wide Web has taken a rapid growth and search engines are the most popular ways of finding any kind of information on it. There are many cases, where the users are having so many web pages where they need to find there solutions in response to the searched query. Generally there are not many users that go beyond the first few pages. Hence the search engines face a challenge on how to provide the best meaningful web pages in response for a user query. In spite of having so many page rank algorithms used by the search engines, the pages that are needed by the users may actually get lost in the tremendous amount of information returned by the search engines. The users that use the search engines are not expert in the domain of search engines, hence the information need to be grouped that is provided by the web pages so that they can navigate the information quickly. Hence the analysis of the pages importance would be done by various matrices and enable the representation of interactions between pages in the form of a network. And finally the analysis of what is the efficiency of rank values of the pages using the proposed system and existing system is carried out. For this purpose, algorithms from the discipline of graph theory are applied, including the Dijkstra’s algorithm for impact analysis, Page Rank and Weighted Page Rank algorithm for calculating the rank values of each web page are used. The weighted page rank algorithm is used to find the page rank of the web pages and with the help of Dijkstra’s algorithm an optimal path is decided on the basis of in-links and out-links. The Dijkstra’s algorithm gives the direction to the Weighted Page Rank algorithm to calculate the page rank values of the pages. This research proposed a Hybrid Page Rank System (HPRS). The HPRS system is composed of three modules User Interface, Operation module and Knowledge Base. The tools used are visual basic server and SQL server database and the framework on which the system is implemented is dot net. The proposed system is implemented in asp.net language. The result of the research compares the rank values of the existing and the proposed system. The performance is calculated for each page simultaneously, is done between the existing Page Rank Algorithm and the Hybrid Page Rank Algorithm. The proposed system gives more efficient rank values of the pages than the existing algorithm.