User package recommendation method and system based on user preference decoupling

The invention discloses a user package recommendation method and system based on user preference decoupling, and the method comprises the steps: collecting package subscription data and satisfaction score data of a user, and building a user package subscription data set; constructing three embedded...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	TONG-WANG LIYAN, BAO BINGKUN, TAN ZHIYI
Format:	Patent
Sprache:	chi ; eng
Schlagworte:	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 COMMUNICATION TECHNIQUE ELECTRIC DIGITAL DATA PROCESSING ELECTRICITY PHYSICS SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR WIRELESS COMMUNICATIONS NETWORKS
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Beschreibung
Zusammenfassung:	The invention discloses a user package recommendation method and system based on user preference decoupling, and the method comprises the steps: collecting package subscription data and satisfaction score data of a user, and building a user package subscription data set; constructing three embedded representations to respectively represent the preferences of the user for the traffic, the voice and the price in the package, and extracting specific data from a user package subscription data set to train each embedded representation, so that each embedded representation only captures one of the preferences; self-adaptively fusing the three embedded representations by using an attention mechanism to obtain the overall preference representation of the user; and calculating the satisfaction degree of the user for the candidate packages based on the overall preference representation of the user, and generating a package recommendation list according to the satisfaction degree. According to the method provided by the