Collaborative filtering method, apparatus, server, and storage medium for fusing time factors

摘要

A collaborative filtering method that fuses time factors, creating an exponential smoothing model and acquiring a time zone established for the exponential smoothing model, and the time zone contains multiple time periods And obtaining user preference degree values for a specified product within a plurality of user identifiers and user identifiers in a plurality of time periods, and iterative calculation for user preference degree values using an exponential smoothing model To obtain a smoothing result corresponding to the time period and generate a 疏 matrix using the smoothing result corresponding to the user identifier and the time period, and the 疏 matrix includes a plurality of users waiting for prediction. That the degree of preference is included, the collaborative filtering model is acquired, and the smoothing result corresponding to the time period is input to the collaborative filtering model; Provide training by filter ring model, and a calculating a predicted value of the preference degree of the plurality of predicted waiting user in sparse matrix. [Selection] Figure 2