Yitong Wang

Title: Yitong Wang: Innovator in Item Recommendation Systems

Introduction

Yitong Wang is a notable inventor based in Hunan, China. He has made significant contributions to the field of item recommendation systems, holding a total of 3 patents. His work focuses on enhancing the accuracy of recommendations through innovative methods and systems.

Latest Patents

One of Yitong Wang's latest patents is titled "Item recommendation method based on importance of item in conversation session and system thereof." This patent describes a method that extracts the importance of each item in a conversation session. It combines this importance with a user's long-term preferences to provide accurate item recommendations. The system aims to improve the accuracy of recommendations by considering both current interests and long-term preferences.

Another significant patent is the "Session-based recommendation method and device." This invention utilizes a pre-trained recommendation model to perform item recommendations. It constructs a directed session graph based on the session to be predicted and inputs it into a gated graph neural network. This process determines a user's dynamic preferences and outputs recommended items based on prediction scores.

Career Highlights

Yitong Wang is affiliated with the National University of Defense Technology, where he continues to advance his research in recommendation systems. His innovative approaches have garnered attention in the academic and technological communities.

Collaborations

Yitong has collaborated with several colleagues, including Fei Cai and Wanyu Chen, who contribute to his research endeavors.

Conclusion

Yitong Wang's contributions to item recommendation systems demonstrate his innovative spirit and commitment to enhancing user experiences. His patents reflect a deep understanding of user preferences and the technology needed to improve recommendation accuracy.