A method and system for quantifying strategy knowledge point personalized recommendation based on transaction performance attribution

CN122412962APending Publication Date: 2026-07-17GUANGZHOU XUSHUO NETWORK TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610805308.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-06-05
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing quantitative trading learning platforms cannot effectively match users' trading results with their knowledge gaps, resulting in a disconnect between teaching and practice and failing to provide personalized and dynamic learning path guidance.

Method used

We construct a quantitative strategy knowledge graph and use a Bayesian network, combined with machine learning attribution models and cognitive diagnostic models, to automatically convert users' trading performance and strategy codes into knowledge point mastery deficiency information and dynamically update the learning path.

Benefits of technology

It enables automatic diagnosis of knowledge gaps based on users' trading performance, providing personalized and dynamic learning paths, thereby improving learning efficiency and practical effectiveness.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122412962A_ABST
    Figure CN122412962A_ABST
Patent Text Reader

Abstract

本发明公开了一种基于交易绩效归因的量化策略知识点个性化推荐方法及系统,属于计算机辅助教育及量化金融技术领域。该方法包括:构建量化策略知识图谱并建立贝叶斯网络;采集学习行为、模拟交易绩效及策略源代码;通过认知诊断模型获得初步掌握概率;分别从策略代码和绩效数据中提取基因型特征和时序特征,输入多任务深度网络归因模型,得到每个知识点的缺陷概率;将初步掌握概率作为先验、缺陷概率作为证据,在贝叶斯网络中执行全局信念传播,得到后验掌握概率并构成动态画像;最后结合学习目标规划避开已掌握节点的最优路径并推荐资源,形成学习‑实践‑诊断的闭环。本发明能够将交易结果和代码特征自动映射为具体知识缺陷,实现精准、动态的个性化学习推荐。
Need to check novelty before this filing date? Find Prior Art