Analysis And Design Of Telephone Voice System Based On Data Mining

The telephone voice System is also called Call Center System,which is the most convenient communication bridge between guests and the company.Interactive Voice Response System(IVRS),which is the core part of the telephone Voice system.In the development of business complexity and customer diversification,self-service broadcasting and static self-service inquiry have become increasingly found to be impractical.Level,for example,too cumbersome and lead to the guest could have self-help implementation can't achieve a simple function,need to contact the customer service agent,it may lead more further complaints,or cause to damage the image of our company.This violates the original shunt,self-help purpose.In this case,the user experience is poor,and in all likelihood,the menu is not properly set.As an approach to solve this problem,IT is concluded that the self-service voice system needs to perceive the user's behavior and transfer from the static voice system to the intelligent broadcast system.Therefore,the research object of this master's thesis is mainly two aspects,one is the combination of self-service system and data analysis technology,and the other is the data warehouse technology of self-service system.This master's degree paper used in the process of data mining is a choice of guests in self-service IVR voice input habit(Road Map)and attribute data,using cluster analysis and correlation analysis method in data mining implementation customer segmentation.Provide personalized menu of functions for different types of customers to improve the efficiency of IVR guests.The prerequisite of data mining is data warehouse technology.This thesis utilizes data mining technology to implement customer behavior mining to adjust the IVR self-service menu,and make an optimized speech scheduling method.First,the customer collects the customer's track information in the self-service broadcasting menu;Then use the data warehouse for processing;Then we use the correlation analysis algorithm to find the rules between each functional menu.Use this as a basis to perceive customer's usage habits.This master's degree thesis of IVR is also made in this paper,the design of system structure frame,and introduces the SPSS modeler software to help in data mining,making the whole data analysis more easily accepted by the management.In the end,the thesis also implements real-time feedback through the self-service speech system.In the original static speech system,the dynamic elements are added to enhance the customer's operating experience in the voice system.And achieve the goal of reducing theaverage user stay in IVR.It is more efficient to tune the voice menu dynamically after the correlation analysis,thus saving the company a lot of human resources.