Efficient Data Mining Method for Finding Competitors from Large Unstructured E-Commerce Data
International Journal for Research in Applied Science and Engineering Technology
Data mining is the popular area of research that facilitates the improvement process of the business, such as the preference of the user of the mining. Mining information from the web used to obtain the opinion on the products or services. In the current competitive business scenario, it is necessary to identify the competitive characteristics and factors of an item that most affect its competitiveness. The competitiveness assessment always uses the opinions of customers in terms of rating
... erms of rating reviews and an abundant source of information from the web and other sources. The problem is to find the top competitors in other domain by considering the features of a particular domain. The system proposes a C 4.5 algorithm for better accuracy to find the top competitors. The unstructured data is structured by using the K-means algorithm. Then this structured data is clustered into an appropriate domain by using our proposed C 4.5 algorithm. For pattern matching previously used Apriori algorithm has some disadvantages so the system use FPgrowth algorithm. The rules provided by FPgrowth algorithm are more accurate than the Apriori algorithm. Finally the result shows that the proposed system is require less time to find top competitors and it more accurate than existing systems.