Similar character recommendation method based on knowledge graph representation learning

• Main Authors: CHEN ZHOU, LI MINGCHEN, ZHU DAN, RUAN XIANGCHAO, WANG YANG, ZHANG KUN
• Format: Patent
• Language: Chinese; English
• Published: 26.02.2021
• Subjects: CALCULATING; 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 relates to a similar character recommendation method based on knowledge graph representation learning, which adopts a brand-new design strategy and comprises social user information processing, representation learning model training, clustering index of character point vectors in entity relationship vectors and similar character recommendation, a TransD representation learning algorithm is adopted to perform knowledge representation learning on all entities and relations in the knowledge graph, and semantic information implied in the graph is mined more deeply; for large-scale figure point vectors, an Annoy rapid calculation distance algorithm is adopted to cluster similar figures, and as semantics in a graph are considered in entity relation vectors, powerful information support is provided for clustering; for the incremental updating condition of the social knowledge graph, a representation learning recommendation algorithm is designed for periodic training, and the freshness of a recommendati