A service recommendation method based on user risk preference

Abstract

The invention is a service recommendation method based on user risk preference, which includes two parts: an attribute extraction method based on user risk preference and a service recommendation method based on user risk preference. The attribute extraction method is a reduction method for multi-attribute features of services. Considering that different users have different willingness to take risks for unknown services, users are divided according to different risk preferences, and corresponding rules are used for different types of users. Extracting from multiple non-functional attributes of services is more suitable for different types of users. Valid attributes of requirements, similarity calculations are performed, and recommendations are finally formed. This recommendation method solves the multi-attribute reduction problem of services, and provides a solution to the problem of uncertain intervals in service non-functional attribute scores in the service recommendation process, making the recommendation results more in line with the needs of different types of users, and providing service recommendations a powerful method and tool.

Description

technical field [0001] The present invention is a scheme for implementing classification and recommendation of services according to different needs of users under an open and heterogeneous complex network environment. Aiming at the characteristics of users' complexity and different expectations for unknown services in service recommendation, a service recommendation method based on user risk preference is proposed, which is suitable for service recommendation. Features, integrate the user's risk preference into the field of service attribute reduction and recommendation, so that the recommendation result is more in line with the user's expectations. It belongs to the field of information service and service computing. Background technique [0002] Service selection means that the user selects the required service from candidate services with the same functional attributes but different non-functional attributes in a certain way according to his own needs, so as to execute ...

AI Technical Summary

Problems solved by technology

Service recommendation is different from traditional recommendation methods. Users have clear application requirements, but it is difficult to judge the extent to which the service meets their own preferences when faced with a large number of services with the same function.

[0003] With the development of information technology and the expansion of the amount of information, most decision makers face too much information instead of too little when making decisions, so how to use existing technology to filter massive information to extract effective information and use effective information Making decisions has become an important research content in the field of decision-making. The reduction of a large number of service non-functional attributes in the field of service recommendation is a necessary process. The understanding of effective service attributes is different, and the impact of unknown evaluation on user needs is ignored. When users face a large number of unknown information choices, it means taking certain risks, and users have different expectations for unknown evaluation. If recommended Failure to take this into account by the system will create serious deviations from user expectations

Images