Personalized session recommendation method for long-term and short-term interest extraction

• Main Authors: ZHANG XIONGTAO, QIN RONGZHEN, LIU ZHUOJIE, LOU JUNGANG
• Format: Patent
• Language: Chinese; English
• Published: 12.09.2023
• 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 personalized session recommendation method for long-term and short-term interest extraction. The method comprises the steps of firstly obtaining personalized information of a user and a session sequence of the user, then performing feature extraction and feature association on the personalized information of the user and the session sequence of the user based on a deep convolutional neural network model to obtain a fusion feature vector, and then determining a recommended article based on the fusion feature vector. In this way, the limitation of a traditional recommendation algorithm in facing user interest changes can be overcome to a certain extent. 公开了一种长短期兴趣提取的个性化会话推荐方法。其首先获取用户的个性化信息以及所述用户的会话序列，接着，基于深度卷积神经网络模型对所述用户的个性化信息以及所述用户的会话序列进行特征提取与特征关联以得到融合特征向量，然后，基于所述融合特征向量，确定推荐物品。这样，可以在一定程度上克服传统推荐算法在面对用户兴趣变化时的局限性。