考虑时间效应的产品组合推荐方法和系统
By constructing an integrated classification tree and survival tree model, the probability of user clicks and conditional clicks is predicted, user responsiveness is calculated, and the strength of product combination recommendations is optimized. This solves the problems of low accuracy and single product recommendation in existing technologies, and achieves more accurate product combination recommendations.
CN115204963BActive Publication Date: 2026-07-17HEFEI UNIV OF TECH
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HEFEI UNIV OF TECH
- Filing Date
- 2022-05-10
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Existing recommendation methods have low accuracy and only recommend products for a single product, failing to effectively consider the time effect, resulting in poor product recommendation performance at the combination level.
Method used
We construct ensemble classification tree and ensemble survival tree models to predict user click probability and conditional click probability, calculate user responsiveness, and generate combination recommendation schemes through a product combination recommendation strength optimization model.
Benefits of technology
It improves the accuracy and effectiveness of product combination recommendations, provides more specific combination recommendation schemes, and enhances user responsiveness and recommendation results.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN115204963B_ABST
Abstract
本发明提供了一种考虑时间效应的产品组合推荐方法和系统，涉及产品推荐技术领域。本发明首先构建集成分类树和集成生存树分别预测用户点击概率以及用户在不同时间的点击概率；然后计算用户响应度并排序生成Top‑N产品推荐列表；最后基于产品组合构建推荐强度优化模型，获得组合中各个产品的推荐强度，提供了更为具体的推荐方案，有效提升了产品推荐的效果。
Need to check novelty before this filing date? Find Prior Art