Interest point recommendation method based on improved LSTM (Long Short Term Memory) and position jump

• Main Authors: ZHANG YIHAO, LAN PENGXIANG
• Format: Patent
• Language: Chinese; English
• Published: 12.08.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention particularly relates to an interest point recommendation method based on improved LSTM (Long Short Term Memory) and position jump, which comprises the following steps: acquiring a target trajectory sequence of a user and inputting the target trajectory sequence into a recommendation model, and outputting a corresponding interest point prediction result: firstly, dividing the target trajectory sequence into a historical trajectory sequence and a current trajectory sequence, and performing embedding processing; then capturing long-term interest behavior dependence of the user through an improved LSTM capable of learning LSTM irrelevant context representation; capturing short-term interest behavior dependence of the user through a position jumping algorithm capable of learning discontinuous non-adjacent interest points with different attention degrees; and finally, fusing the user long-term interest behavior dependence and the user short-term interest behavior dependence to obtain a corresponding i