Business recommendation method, device and equipment

• Main Authors: YAN HAITAO, WANG ZIXIANG, GUO CI
• Format: Patent
• Language: Chinese; English
• Published: 13.11.2020
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The embodiment of the invention relates to the technical field of machine learning and discloses a service recommendation method, device and equipment. The method comprises steps of obtaining a service content matrix, a user score matrix and a user call relation adjacency matrix of a target user; generating a service hidden factor feature matrix, a user hidden factor feature matrix and a call relationship hidden factor feature matrix through a neural network according to the service content matrix, the user scoring matrix and the user call relationship adjacency matrix; and performing servicerecommendation according to the service implicit factor characteristic matrix, the user implicit factor characteristic matrix and the call relationship implicit factor characteristic matrix. The method is advantaged in that service recommendation efficiency is greatly improved, errors are reduced, and service recommendation and service prediction precision is improved. 本发明实施例涉及机器学习技术领域，公开了一种业务推荐方法、装置和设备，所述方法包括：获取目标用户的业务内容