Hybrid Association Rules to Classify and Discovering Item sets based on User Knowledge

Abstract

Increase large luminous data is posing research concept to predict and effective analysis of data in different types of business oriented applications. Intelligence of business maintenance is an aggressive concept to provide analysis customers, employees and suppliers for more effective decision making. It is basic representation to update and enhance their business in different formats like quality of service, and achieve profitability for business organizations. To provide analysis of data effectively, traditionally some of the data mining approaches like Classification, Clustering and Association were used to provide and increase the efficiency in real time business organizations. Based on in depth analysis of data relates to data delivery, reporting and predictive analysis, data mining techniques have failure to elaborate business in real works. So it aims to explore the advanced and statistical techniques apart from existing data mining or modeling approaches. In this paper, we propose and implement a novel HDM approach (which consist Association rules and Classification rules) for effective analysis of large data sets. Our approach is increases the services like quality of service, customer service, and report generations in business organizations. Furthermore, hybrid data mining approach is designed to assist to user throughout data analyzing task. By applying our new approach over large data to integrate domain expert knowledge in post processing to reduce set of rules in business intelligence service analysis. Our experimental results show effective maintenance of customer service to reduce quality of filtered rules by domain expert knowledge with interactive process in real world business organizations.