The Application Research Of Data Mining In Service Management Of Call Center

As rising heated industry in recent years, call Center plays a significant role on an enterprise's customer service. Based on the management data in the running of 10010 Call Center, Chongqing Unicom, the concepts, algorithm and realizations of decision tree are discussed in the dissertation. The data of staff's files and those of customers'satisfaction and evaluation with the Call Center are mined via the improved association rule algorithm. In this way, the key factors which influence the service quality are found, analyzed and testified. The research can be significant in both theory and practice. The most essential part of the dissertation is the study on classic ID3 algorithm and Apriori algorithm. The main works are listed as follows:①The detailed theoretical background of ID3 algorithm is introduced, and based on the treeing rules. The decision tree is obtained through the practical learning of this algorithm. In order to analyze the advantages and disadvantages of this algorithm, a comprehensive evaluation of ID3 is made, and two suggestions about the algorithm are proposed.1) ID3 algorithm is liable to select the attributes with more values, but the defect can be solved with the concept of information gain proportion.2) ID3 algorithm is restricted to deal with the discrete values, and a preprocessing technique is proposed to discretize the continuous-valued attributes, so that this disadvantage can be avoided.The improvement of ID3 algorithm is applied in the practice of Call Center Service, Chongqing Unicom. The relevant testing and evaluation are made.②The Apriori is an algorithm about the generation of association rule. Enlightened by the generation of candidate k-itemsets, k subsets of one item will be first find out, and then connected to get subsets of two items, and finally get the subsets of k-1 items. The algorithm can be used to find out the proper subset and to generate the protasis and apodosis of the association rule.③The next chapter of the dissertation is devoted to the realization of data mining system for the service quality evaluation. Two plans can be used to solve the problem. The second plan realizes the idea of applying the method that generates candidate k-itemsets from the Apriori arithmetic in the finding of proper subset, which helps to generate a comparatively detailed association rule, and can be a good reference for the Call Center service quality management. This dissertation provides a software framework on the data mining system for service quality evaluation. The relevant data are used in an experiment on the association rule algorithm and the result is well-analyzed.The fifth cycle customer service system of Chongqing Unicom has been successfully put into production in June, 2007, and the self-developed Call Center Human Resource Management System has also been updated in early 2007. As an effective extension of the fifth cycle of the customer service system, the data mining system for service quality evaluation correlates the customer service system and human resource management. Main function of the system is to mine data about customers'satisfaction and staff's files effectively via decision tree and association rule.