一种基于树结构的菜谱优化方法、系统、存储介质及设备
By converting recipes into a tree structure and utilizing latent space encoding to optimize recipes with similar logic, combined with deep learning and human evaluation, the problem of low efficiency and high cost in traditional recipe optimization is solved, generating recipes that match tastes.
CN117236287BActive Publication Date: 2026-07-17UNIV OF JINAN
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- UNIV OF JINAN
- Filing Date
- 2023-09-26
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Traditional recipe optimization methods rely on manual evaluation, which is inefficient and focuses only on the quantity of ingredients while ignoring the cooking logic, resulting in high optimization costs and long recipe generation time.
Method used
The recipe text is converted into a tree structure, and a new recipe is generated using the constraint rule set of the tree structure. Recipes with similar logic are selected for optimization through latent space encoding, and recipes that match tastes are generated by combining human evaluation and deep learning models.
Benefits of technology
It improves recipe optimization efficiency, reduces time costs, generates recipes with clear logic, caters to popular tastes, and requires no professional chef experience.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN117236287B_ABST
Abstract
本发明涉及文本数据处理领域，具体为一种基于树结构的菜谱优化方法、系统、存储介质及设备，包括以下步骤：获取菜谱中的文本信息转换为菜谱树并构建数据集，基于菜谱树构建约束规则集合，根据得到的约束规则集合基于菜谱生成模型得到新的菜谱；在生成的新菜谱中选取待优化的目标菜谱进行潜空间编码，根据数据集中所有菜谱的潜空间编码与目标菜谱潜空间编码之间距离，筛选出距离最小的多个菜谱，筛选出的多个菜谱和目标菜谱共同作为待优化的菜谱；获取待优化菜谱的评价结果作为初始解，得到推荐的潜空间向量并生成规则序列的菜谱树，继续获取对应菜谱的评价结果并经若干次循环后取评价结果最高分对应的菜谱树作为优化后的菜谱。
Need to check novelty before this filing date? Find Prior Art