Journal of Contemporary Issues in Business and Government

Abstract

In present day, E-commerce is one of the fast-developing business models among many. The key aspect of E-commerce nowadays is shopping from any place at any time. In real time as vast amount of data are generated because of vast population of people and devices that are connected, have created challenges to handle the huge data stream that are arriving from every device. The data stream click is captured by another famous approach that is called as Web Data Mining. Every time a customer seeks for some details, or to browse some of the category of products or to do any transaction. All these functions leave trials of data as a resource for web data mining, which is required to portray the behaviour patterns of user. The organization in which the data are positioned will provide a means for analysis of data. The transactions data done by customer can be used for categorizing for suggesting systems and to get high profit. So this is done by using KNN (K-nearest neighbour), Random forest (RF) and SVM (Support Vector Machine) classifier in this paper. These approaches are useful to recommend the best strategy planning to achieve decision making and enhance sales of products. Among all of them random forest is best method which has 99.86% accuracy