Construction method of interpretable recommendation model based on knowledge graph
The invention discloses a construction method of an interpretable recommendation model based on a knowledge graph, which comprises the following steps: S1, converting entity and relationship category data in the knowledge graph into embedded vectors through Word2Vec, and splicing the embedded vector...
Gespeichert in:
Hauptverfasser: | , , , |
---|---|
Format: | Patent |
Sprache: | chi ; eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | The invention discloses a construction method of an interpretable recommendation model based on a knowledge graph, which comprises the following steps: S1, converting entity and relationship category data in the knowledge graph into embedded vectors through Word2Vec, and splicing the embedded vectors to form a sentence matrix; s2, processing the sentence matrix through a convolution kernel to obtain different feature mappings; s3, enabling different feature mappings to pass through a pooling layer, and enabling different path sequences to be pooled and then subjected to fixed-length feature representation; s4, inputting the features output by the pooling layer into the RNN network, and outputting an implicit vector as a path representation; s5, predicting the interaction probability of the user and the candidate item based on a weighted pooling function, calculating score contributions of different paths, and selecting the path with the maximum score as recommendation; s6, evaluating the recommended path by a |
---|