Action Evaluation Model for Wearable Device Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current wearable devices can only provide action classification and lack the capability for action scoring, limiting their application in evaluating user actions effectively.
Innovation Solution
An action evaluation model is built using supervised learning based on machine learning algorithms, which generates standard action feature vectors and classification vectors to enable both action classification and scoring by selecting representation action feature vectors with relative component maximal values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If current identification technologies are used, then action classification can be provided, but action scoring capability is lacking
Solution Approach 1:
The model is segmented into two distinct stages: a feature extraction stage that processes raw sensor data into feature vectors, and a classification stage that evaluates actions based on these features. This segmentation allows the system to provide both action classification and scoring capabilities while managing complexity through modular architecture.
Solution Approach 2:
The action evaluation model is designed to perform multiple functions simultaneously: it classifies actions into categories and provides scoring evaluations for each action. This multi-functionality is achieved through the unified model structure that outputs both classification labels and evaluation scores, enhancing the versatility of wearable devices.
2Adaptability or versatility
If a unified model provides both action classification and scoring, then functionality is enhanced, but computational complexity increases
Solution Approach 1:
The feature extraction stage performs preliminary processing of raw sensor data before the classification stage. By pre-computing meaningful features from raw data, the system reduces the computational burden on the classification stage, enabling dual functionality while managing overall computational complexity.
Solution Approach 2:
Feature vectors serve as an intermediary representation between raw sensor data and final action evaluation. This intermediate representation captures essential characteristics of actions in a compressed form, facilitating both classification and scoring while reducing computational requirements compared to processing raw data directly.
Data Source
AI summary
An action evaluation model building apparatus and an action evaluation model building method thereof are provided. The action evaluation model building apparatus stores a plurality of raw data sets and a plurality of standard action labels corresponding thereto. Based on machine learning algorithms, the action evaluation model building apparatus computes the raw data sets and performs a supervised learning to build a feature vector creation model and a classifier model. The action evaluation model building apparatus determines a representation action feature vector of each standard action label by randomly generating a plurality of action feature vectors and inputting them into the classifier model. The action evaluation model building apparatus builds an action evaluation model based on the feature vector creation model, the classifier model and the representation action feature vectors.