Exercise Indicator Modeling for Cross-Period Running Form Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing devices for evaluating a subject's running form lack the ability to accurately compare exercise information across different periods, especially when conditions such as fatigue, weather, or terrain change, leading to inconsistent and misleading assessments.
Innovation Solution
An information processing device that generates period-specific models based on exercise data to derive estimation data, allowing for the presentation of exercise indicators in a mutually comparable manner, using a first model for one period and a second model for another period, with the first indicator assumed to be a certain value.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single evaluation model is used across different periods, then the device complexity is reduced, but the measurement precision of exercise information comparison deteriorates due to changing conditions
Solution Approach 1:
The patent divides the evaluation process into period-specific segments by generating separate estimation models for different time periods. Each model is trained on data from its specific period, allowing the system to adapt to changing conditions while maintaining manageable complexity through modular, time-bound model instances.
Solution Approach 2:
The patent implements dynamic adaptation by generating new estimation models for each period based on data from that specific period. This allows the evaluation system to dynamically adjust to changing conditions (fatigue, weather, terrain) rather than using a static single model, improving measurement precision across varying contexts.
2Measurement precision
If period-specific models are generated for each time period, then the measurement precision of exercise information comparison is improved, but the device complexity increases
Solution Approach 1:
The patent segments the modeling process into distinct period-specific models, where each model handles a specific time period. This segmentation improves measurement precision by capturing period-specific patterns while organizing complexity into manageable, isolated model units that can be independently managed and processed.
Solution Approach 2:
The system employs automatic model generation and selection mechanisms that reduce manual intervention. The processor automatically generates appropriate estimation models for each period and selects the correct model based on the input data's time period, allowing the system to self-manage the complexity of multiple models without proportionally increasing operational burden.
3Loss of time
If exercise data from different periods is compared directly without period-specific modeling, then the processing time is reduced, but the reliability of the evaluation deteriorates due to inconsistent conditions
Solution Approach 1:
The patent implements dynamic model generation that adapts to each period's specific conditions. By automatically creating period-specific estimation models, the system maintains evaluation consistency and reliability across different time periods while minimizing manual processing time, as the model selection and generation is automated based on the input data's temporal characteristics.
Data Source
AI summary
In an information processing device, at least one processor is configured to derive, based on a first model generated with as an input a first indicator expressing first exercise information of a subject acquired in a first period, and as an output a second indicator expressing second exercise information, first estimation data expressing the second indicator relative to first indicator specified in the first period, the second exercise information being different than the first exercise information of the subject, the second indicator having correlation with the first indicator; derive, based on a second model generated with as an input the first indicator acquired in the second period, and as an output the second indicator, second estimation data expressing the second indicator relative to the first indicator specified in the second period; and present the first estimation data and the second estimation data in a mutually comparable manner on a presenter.