Group recommendation method based on evidence theory

Abstract

The invention discloses a group recommendation method based on an evidence theory. The group recommendation method comprises the following steps of: 1, respectively representing scores of users on projects and affiliation relationships between the users and groups by using triples; 2, implementing a probability matrix decomposition method, and obtaining a prediction score of each member in the group for the project; 3, defining the weight and reliability of each member in the group; 4, for each group, fusing the prediction score information of the group members by using an evidence reasoning method to obtain a prediction score of the group for the project; and 5, sorting the projects from high to low according to the prediction scores of the group for the projects, and selecting the firstW projects to finally form a recommendation list of the group. According to the method, the weights of the group members can be fully considered when the group is recommended, certain reliability is given to evidences with different weights at the same time, a group recommendation result satisfying most of the group members as much as possible is synthesized by adopting an evidence theory method,and therefore the group recommendation effect is effectively improved.

Description

technical field [0001] The invention relates to the technical field of recommendation methods, in particular to a group recommendation method based on evidence theory. Background technique [0002] In recent years, with the development of online communities, a large number of users and information have emerged on major social media, which brings convenience to users and problems such as information overload. As an effective way to solve the problem of information overload, the recommendation system can help users quickly and effectively find the information they need in massive amounts of data, and meet the individual needs of different users. . [0003] Due to the social nature of human beings, people often like to participate in certain activities together in a certain group form, for example, having dinner with a group of colleagues in the same company after get off work, watching TV programs with family members, and planning travel locations with friends . In the past,...

AI Technical Summary

Problems solved by technology

The minimum regret strategy and the maximum happiness strategy only regard the opinions of the most dissatisfied or most satisfied members in the group as the opinions of the group, which obviously cannot effectively model the group preferences, while the average strategy combines the opinions of the group members when synthesizing group opinions The opinions of all members are treated equally, without taking into account the different contributions of group members in the actual situation. Members are unknown. In fact, different characteristics of different group members will make them play different roles in the process of generating group recommendation results. Existing research methods often ignore this point and reduce the accuracy of group recommendation methods.

Images