Analysis On The Relation Between 3G Users Internet Satisfaction And Network Index

With the competition between the three major telecom operators intensified and the development of mobile Internet, customer satisfaction has attracted more and more their attention. Improving customer satisfaction become one of the important measures to telecom operators attract new customers and retain the stock of the users. In recent years, telecom operators regularly survey customers to track customer satisfaction trends. Then, the customer satisfaction is early impacted by upgrading network performance to avoid churn. In view of this, the urgent need is to establish a relationship between the user subjective satisfaction and network index that can targeted improve customer satisfaction and user experience.Based on the subjective evaluation data accessed by 3G users internet satisfaction surveys and corresponding network index data, this paper analyses the correlation between the two, establishes a correlation model between the two, and in-depth analyses time window of network metrics effect on subjective evaluation of customer satisfaction. On this basis, the paper establishes regression model between 3G users internet satisfaction and network index. In the modeling process, this paper detailed compares of the SVR, Gaussian process regression and BP neural network, and obtains optimal time window of the regression model by grid search method. With this regression model, telecom operators can access for the user satisfaction targeted tune network index.