Serialized neural collaborative filtering recommendation method based on self-attention mechanism

• Main Authors: WU YINCHEN, LI RUI, HEI LEI, JIANG ZHIPING, XU YUESHEN, QIU ZHIBO
• Format: Patent
• Language: Chinese; English
• Published: 23.05.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention discloses a serialized neural collaborative filtering recommendation method based on a self-attention mechanism, and relates to the technical field of recommendation systems.The method comprises the steps that commodity IDs having interactive records with a user are sorted into a user historical behavior record sequence; performing one-hot embedding on the user ID and the commodity ID to obtain a user feature vector and a commodity feature vector; putting the to-be-recommended commodity feature vector at the tail end of the user historical behavior record sequence, and inputting the to-be-recommended commodity feature vector into a self-attention mechanism to obtain a to-be-recommended commodity feature vector with rich sequence information; and inputting the commodity feature vector with rich sequence information and the user feature vector and the commodity feature vector which are obtained by original embedding into a neural collaborative filtering model for calculation, and finally performin