Exercise Recommendation Knowledge Graph for Chronic Disease
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing exercise recommendation systems fail to accurately tailor exercise programs to individuals with chronic diseases or sub-health conditions, as they do not adequately consider the user's physical condition and disease characteristics.
Innovation Solution
An exercise recommendation method that acquires basic user information, queries a pre-established exercise knowledge graph to obtain a list of target exercise modes and evaluation attributes, determines exercise programs based on these lists, ranks the programs, and recommends the most suitable ones.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If generic exercise recommendations are provided without considering individual conditions, then the system is simple and easy to implement, but the accuracy and suitability of exercise recommendations deteriorates
Solution Approach 1:
The patent segments the exercise recommendation process into multiple independent modules: user information acquisition module, knowledge graph query module, exercise program determination module, and ranking module. Each module handles a specific aspect of the recommendation process, allowing the system to achieve high accuracy through comprehensive evaluation while maintaining manageable complexity through modular design. The segmentation enables parallel processing of different evaluation attributes without creating a monolithic complex system.
Solution Approach 2:
The patent introduces an exercise knowledge graph as an intermediary data structure that stores pre-established relationships between exercise modes, diseases, and evaluation attributes. This knowledge graph acts as a mediator between user conditions and exercise recommendations, enabling accurate recommendations without requiring complex real-time analysis algorithms. The knowledge graph pre-processes and organizes medical and exercise data, simplifying the recommendation engine's task while maintaining high accuracy.
2Reliability
If multiple evaluation attributes are considered for exercise programs, then the suitability for chronic disease patients improves, but the processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary action by pre-establishing the exercise knowledge graph with all possible evaluation attributes, relationships, and exercise programs before actual recommendations are needed. During the recommendation process, the system only needs to query pre-computed data from the knowledge graph rather than performing complex real-time calculations. This preliminary preparation enables fast query-based recommendations while considering multiple evaluation attributes simultaneously.
Solution Approach 2:
The patent applies local quality by allowing different evaluation attributes to have different weights and importance levels based on specific user conditions and disease types. Rather than uniformly processing all attributes with equal computational resources, the system dynamically adjusts the depth and detail of evaluation for each attribute based on its relevance to the user's specific health condition, optimizing processing time while maintaining reliability.
3Adaptability or versatility
If exercise recommendations are personalized based on physical activity level and disease type, then the effectiveness increases, but the data requirements and system complexity increase
Solution Approach 1:
The patent creates a universal exercise knowledge graph that can serve multiple functions: it stores information for different disease types, accommodates various physical activity levels, and supports multiple evaluation attributes simultaneously. This single multi-functional knowledge graph structure enables personalized recommendations across diverse user populations without requiring separate data collection systems for each disease or user type, reducing overall data requirements while maintaining high adaptability.
Data Source
AI summary
An exercise recommendation method and apparatus are described. The method includes: acquiring basic information of a user; querying a pre-established exercise knowledge graph according to the basic information of the user to obtain a target exercise mode list comprising exercise modes, and an evaluation attribute list comprising evaluation attributes corresponding to the target exercise mode list; determining at least one exercise program according to the target exercise mode list and the evaluation attribute list; performing ranking processing on the at least one exercise program according to the evaluation attribute list to obtain a ranking result; recommending at least one target exercise program according to the ranking result.