Service recommendation method and system based on social network analysis

Abstract

The invention discloses a service recommendation method and system based on social network analysis. The method includes the steps that according to the historical communication records of a first user, communication traffic information of each communication mode used when the first user interacts with at least another user is obtained; according to the communication traffic information, the interaction factor between the first user and another user is calculated; according to the historical service information of the first user, the use information of each service applied by the first user is determined and subjected to ex-right processing; according to the interaction factor between the first user and another user and the use information, subjected to ex-right processing, of each service applied by the first user, the recommended priority of each service applied by the first user is determined, and the services with the highest current priority are recommended to the first user. The service recommendation method and system based on the social network overcome the defect that a traditional method can only carry out analysis for a single service, a single client and a single communication mode, and can recommend the most needed service for the user.

Description

technical field

[0001] The invention relates to the technical field of data services, in particular to a service recommendation method and system based on social network analysis. Background technique

[0002] As the market competition among the three major operators intensifies, refined customer relationship modeling methods have been widely used, and the model based on customer behavior can recommend the most needed services for users. This type of method uses supervised or unsupervised data mining algorithms to predict the customer's business purchase propensity for a period of time in the future based on customer behavior, preferences, costs and other information, so as to find the most valuable customers, so as to achieve expansion. The purpose of customer size.

[0003] The traditional typical method of building a model through data mining for business recommendation includes the following steps: obtaining customer order and non-order data after a certain business rec...

Images