Workout Effort Scoring Using Two-Stage Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing fitness tracking technologies do not provide a comprehensive and intuitive measure of user effort during workout sessions, lacking a standardized method to quantify and visualize the intensity and trends of user effort over time.
Innovation Solution
A method and system utilizing machine learning models, specifically XGBoost classifiers and regressors, to estimate user effort by analyzing workout session features such as heart rate, duration, and environmental factors, followed by a bias correction mechanism to refine the effort score based on user history.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If traditional fitness tracking methods are used to monitor workout metrics, then basic parameters like heart rate and duration can be tracked, but a comprehensive and intuitive measure of user effort cannot be provided
Solution Approach 1:
The user effort measurement is segmented into multiple components: a classification model that divides effort into discrete categories (easy, moderate, hard, very hard) and a regression model that provides continuous effort scores. This segmentation allows the system to capture different aspects of user effort separately and combine them for a comprehensive measurement.
Solution Approach 2:
The patent introduces machine learning models as intermediary components between raw workout data and user effort measurement. These models process multiple workout features (heart rate, speed, elevation, duration) and transform them into meaningful effort categories and scores, serving as mediators that bridge the gap between raw data and intuitive effort assessment.
2Measurement precision
If multiple workout features are analyzed to improve effort measurement accuracy, then user effort can be quantified more precisely, but the complexity of the measurement system increases
Solution Approach 1:
The analysis of multiple workout features is segmented across two separate machine learning models. The classification model handles categorical classification based on workout features, while the regression model handles continuous score prediction. This segmentation allows each model to specialize in one type of analysis, improving overall precision while managing complexity through modular architecture.
Solution Approach 2:
The machine learning models are designed to handle multiple workout features universally - including heart rate, speed, elevation, duration, and workout type - through a single integrated framework. This multi-functionality allows the system to process diverse input features without requiring separate processing pipelines for each feature type.
Data Source
AI summary
Disclosed are methods, systems, and computer-readable media to perform operations including: receiving, by a first machine learning model executed by one or more processors, one or more features related to the workout session; generating, by the first machine learning model and based on the one or more features, a first output including an estimated classification of the user effort for the workout session in a particular category of a plurality of known categories; receiving, by a second machine learning model executed by the one or more processors, the one or more features and the estimated classification output by the first machine learning model; and generating, by the second machine learning model and based on the one or more features and the estimated classification, a second output including an estimated score of the user effort for the workout session.