Knowledge graph enabled large language model shooting auxiliary training method and system

By constructing a knowledge graph and a large language model, combined with players' static physiological parameters and dynamic motion data, we have achieved accurate prediction of basketball shooting performance and analysis of the reasons for inaccuracy, providing personalized training guidance. This solves the problems of difficulty in quantifying training effects and poor interactivity in existing technologies, and enables real-time feedback and guidance.

CN122399327APending Publication Date: 2026-07-17BEIJING SPORT UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
BEIJING SPORT UNIV
Filing Date
2026-04-21
Publication Date
2026-07-17

Smart Images

  • Figure CN122399327A_ABST
    Figure CN122399327A_ABST
Patent Text Reader

Abstract

本申请公开了一种知识图谱赋能大语言模型的投篮辅助训练方法及系统，涉及智能体育技术领域，该方法包括：获取投篮数据，包括各位置球员在预设投篮场景下的生理参数与运动数据；预设场景由距离、位置和防守状态组合而成；将投篮数据输入融合预测模型，得投篮状态；根据投篮状态和对应数据，采用“实体‑关系‑属性”三层架构构建领域知识图谱，其中实体包括球员属性、运动数据、投篮状态及训练建议，关系和属性均来源于篮球训练领域预定义专业知识；将图谱中的结构化语料微调基础大语言模型；利用微调后模型进行投篮辅助训练。本申请可实现投篮状态精准预测、失准原因智能分析、个性化训练指导及实时交互反馈。
Need to check novelty before this filing date? Find Prior Art