Hybrid Approach for Improvement of Web page Response Time Kushwant Kaur # , Prof. Kanwalvir Singh Dhindsa * # Research scholar, Dept. Of CSE, BBSBEC(Fatehgarh Sahib), Punjab(India) * Associate Professor, Dept Of CSE BBSBEC (Fatehgarh Sahib), Punjab (India) Abstract— The objective of this paper is to propose an integrated web mining technique for improving the response time of web pages and reducing user perceived latency. This hybrid technique is based on integration of two Web mining techniques: Web caching and Web pre-fetching. By integration, these techniques may compliment each other since Web Caching exploits temporal locality and Web Pre-fetching utilizes spatial locality. Pre-fetching the predicted web objects into the proxy server cache can increase cache hit-ratio. A comparison between web pre-fetching techniques is done to present the benefits of hybrid approach. This paper also proposes that if Prefetching techniques like Domain Top Prefetching and Dynamic Web Prefetching are combined then it can further improve the Hybrid approach. Keywords— Web mining; user perceived latency; proxy server; Web caching; Web pre-fetching; temporal locality; spatial locality; cache hit-ratio; Domain Top prefetching; dynamic prefetching. I. INTRODUCTION This Web has become today a virtual society as being used for communication channels and search purposes. Network congestion has become the biggest problem due to the massive use of Internet and World Wide Web. Server is overloaded with user’s frequent requests for web page access. This dramatic increase is due to popularization of new applications and services like e-commerce, e-learning, e-business, multimedia contents etc. and has given rise to problems like user perceived latency, global traffic, damaging the quality of service and backbone link congestion. User perceived latency has been considered as the most serious problem as it results in impatience which is the most common reason users terminate their visit at web sites. Potential sources of latency are the overloaded web servers, network congestion, low bandwidth, bandwidth underutilization and propagation delay. Web mining can be used to improve latency. Web mining is a type of data mining used to automatically discover and extract information from Web documents and other web services. It consists of specific techniques, algorithms and methodologies to mine the web, mainly because the web has a great amount of unstructured data and the changes are frequent and rapid. Web mining techniques can be implemented on the web logs maintained byte servers as to discover user access and traversal paths. Web caching has been used as one of the effective techniques to reduce network traffic, access latencies and bandwidth underutilization. [4] Cache storage space is limited and some pages need to remove when cache is full and the new pages are to be brought into the cache. This cache replacement may lead to inefficiency as the deleted page may be requested again. A lot of studies have been done o improve Web caching performance. For example, [6] applied a technique of predicting the future web access using Web Log Mining for improvement of caching performance. An integrated techniques has been suggested in [8] as an approach of clustering and classification techniques for the building the model of cache replacement policy. The knowledge and comprehension of the behavior of a web user are important factors in a wider range of fields related to web architecture, design, and engineering. The information that can be extracted from web user’s behavior permits to infer and predict future accesses. This information can be used for improving Web usability as proposed by [3] developing on-line marketing techniques suggested by [9] or reducing user-perceived latency, which is the main goal of prefetching techniques. These techniques use access predictors to process a user request before the user actually makes it. The goal of the paper is to increase the hit ratio by proxy pre-fetching and lessen the burden on the proxy server and the network. In our scheme, we propose an integrated approach for improvement of web performance and to reduce the web page response/retrieval time. In Domain Top technique of pre-fetching; the proxy finds the popular domains using access profiles and searches the popular documents in each domain [11]. Based on these Top-Domain and Top-Documents, proxy makes the rank list for prefetching, the client requests a file in a certain domain and proxy forwards to them their most popular documents in the rank list. Heavy computation is not required to find the popular domains and documents, but only needs a very small amount of rank list that stores them at the proxy. The paper is organized as follows. A brief description of Web log mining process is reviewed with the help of sample data taken from IIS and apache server logs in Section 2. Section 3 describes the web mining techniques for performance improvement along with the integration of Kushwant Ka et al, / (IJCSIT) International Journal of Computer Science and Information Technologies, Vol. 5 (5) , 2014, 6755-6759 www.ijcsit.com 6755