Study and Analysis of Apriori and K-Means Algorithms for Web Mining

Abstract

Emerging technologies such as cloud computing and data mining deal with issues such as scalability, security and efficiency. Web mining is broadly classified under data mining, refers to the resultant data combination obtained by assembling data available in the Web and information mining techniques. In general, mining can be defined as the process of extracting significant things from huge datasets. The applications of Web mining include the evaluation of the successful completion of a particular task, assessing the applicability of specific Web sites and comprehending client conduct. The proposed work is concerned with distributed networks and is aimed at improving the efficiency of data mining and cloud computing techniques. The main goal is to find a solution for generating various itemsets in each site. A well-known algorithm in data mining is the Apriori algorithm which discards infrequent items at the cost of useful data. A major limitation of this algorithm is its slowness, owing to increased transactions. The K-means segmentation algorithm is employed for increasing the efficiency by clustering the initial itemset.