Patents by Inventor Junjun NING

Junjun NING has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Method and apparatus for predicting cancellations of reservation services based on fusion of multi-scale time series features
    Patent number: 12688513
    Abstract: Provided are a method and an apparatus for predicting cancellations of reservation services based on fusion of multi-scale time series features, relating to the field of predicting cancellations of reservation services. The method includes: obtaining reservation feature data of a reservation service order to be predicted, where the reservation feature data includes service provider feature information, user reservation feature information, statistical time series data, and statistical non-time series data; performing multi-scale time series feature fusion on various time series in the statistical time series data, and inputting fused data, together with other reservation feature data, into a multi-task learning prediction network for cancellation prediction. The multi-task learning prediction network includes a main cancellation prediction model and an auxiliary rejection prediction model. This application can accurately predict whether a reservation service order to be predicted is canceled during service.
    Type: Grant
    Filed: October 21, 2024
    Date of Patent: July 21, 2026
    Assignee: Zhejiang University of Science and Technology
    Inventors: Wenjian Xu, Lei Zhang, Junjun Ning, Yuyang Ji, Jian Wan
  • METHOD AND APPARATUS FOR PREDICTING CANCELLATIONS OF RESERVATION SERVICES BASED ON FUSION OF MULTI-SCALE TIME SERIES FEATURES
    Publication number: 20260050938
    Abstract: Provided are a method and an apparatus for predicting cancellations of reservation services based on fusion of multi-scale time series features, relating to the field of predicting cancellations of reservation services. The method includes: obtaining reservation feature data of a reservation service order to be predicted, where the reservation feature data includes service provider feature information, user reservation feature information, statistical time series data, and statistical non-time series data; performing multi-scale time series feature fusion on various time series in the statistical time series data, and inputting fused data, together with other reservation feature data, into a multi-task learning prediction network for cancellation prediction. The multi-task learning prediction network includes a main cancellation prediction model and an auxiliary rejection prediction model. This application can accurately predict whether a reservation service order to be predicted is canceled during service.
    Type: Application
    Filed: October 21, 2024
    Publication date: February 19, 2026
    Inventors: Wenjian XU, Lei ZHANG, Junjun NING, Yuyang JI, Jian WAN