基于马斯洛需求层次理论的推荐方法及装置

骆宗伟; 姜珊; 石欣晨; 李斌; 杨宇; 黄志云

骆宗伟

南方科技大学 ; 深圳市傲天科技股份有限公司

南方科技大学

518055 广东省深圳市南山区西丽学苑大道1088号

南方科技大学 ; 深圳市傲天科技股份有限公司

518055 广东省深圳市南山区西丽学苑大道1088号 (广东,深圳,南山区)

南方科技大学

518055 广东省深圳市南山区西丽学苑大道1088号 (广东,深圳,南山区)

CN201710159897.6

2017-03-17

CN107590672B

2021-01-22

2021-01-22

授权

2021-01-22

授权发明

第一

本发明适用于信息技术领域，提供了一种基于马斯洛需求层次理论的推荐方法及装置，包括：从运营商的管道数据中获取若干个应用程序；遍历所述若干个应用程序中的每一个应用程序，根据所述应用程序的功能和用户群将其划分至对应的需求层次中，以构建应用程序的马斯洛理论体系；计算待推荐用户在所述马斯洛理论体系中的坐标信息；根据所述坐标信息对所述待推荐用户进行聚类分析，以将具有相似需求的待推荐用户聚为一类；根据用户的兴趣爱好计算同一聚类中不同的待推荐用户之间的关联性，根据所述关联性对所述待推荐用户进行推荐操作。本发明实现了以人为中心、从人的基本需求出发的用户聚类，有利于挖掘出用户的深层需求，提升个性化推荐的效果。

The invention is applicable to the technical field of information, and provides a recommendation method and a device based on the Maslow's hierarchy of needs. The method comprises steps: a plurality of applications are acquired from pipeline data of an operator; each of the plurality of applications is traversed and is divided to a corresponding hierarchy of needs according to the functions and the user group of the application, and a Maslow's theory system for the application is built; the coordinate information of a to-be-recommended user in the Maslow's theory system is calculated; according to the coordinate information, clustering analysis is carried out on the to-be-recommended user, and the to-be-recommended users with the similar needs are clustered to one group; and according to hobbies and interests of the users, the correlation between the different to-be-recommended users in the same cluster is calculated, and according to the correlation, recommendation operation is carried out on the to-be-recommended user. User clustering with a person as the center and starting from the basic needs of the person is realized, the deep needs of the user can be excavated, and the effect of enhancing personalized recommendation is realized.

G06Q30/02 ; H04L29/08

（+PATENT GRANT）[2021-01-22][CN]

1

1

阳开亮

深圳中一专利商标事务所

PatSnap

骆宗伟,姜珊,石欣晨,等. 基于马斯洛需求层次理论的推荐方法及装置[P]. 2021-01-22.