Improving Identification Accuracy and Feature Analysis in Individual Identification Using Center of Gravity Sway

Q: What does this research mean for the field?

Machine learning models, particularly Support Vector Machines and Random Forests, can achieve over 98% accuracy in individual identification by analyzing dynamic center of pressure data during specific balance board movement phases. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

Yuta Kojima; Noboru Katayama

doi:10.1541/ieejeiss.146.624

May 31, 2026IEEJ Transactions on Electronics Information and Systems

Improving Identification Accuracy and Feature Analysis in Individual Identification Using Center of Gravity Sway

View at Publisher

Ask AI

Bookmark

Share

Key Points

The aim is to enhance personal identification accuracy through the analysis of center of gravity sway data during balance exercises.
Performed experiments on 10 subjects analyzing COP data using a Wii Balance Board.
Data was classified into segments based on acceleration, focusing on stabilization, standing, and pre-fall phases.
Utilized SVM and random forest classification methods with 36 features derived from COP data.
SVM achieved an accuracy of over 99%, while random forest reached accuracy above 98%.
XGBoost classification accuracy was approximately 95%, indicating SVM and random forest's superiority.
SHAP analysis highlighted the significance of lateral dynamic characteristics in certain phases contributing to high classification accuracy.

Implication

Study improves personal identification accuracy using dynamic center of pressure data, suggesting effective classification methods.

Abstract

We have improved personal identification accuracy using dynamic center of pressure (COP) data during stepping on and stepping off the balance board. Wii Balance Board data was resampled at 50Hz, log-transformed, smoothed (Savitzky-Golay), and classified into five segments (S1-S5) based on X-axis acceleration. To reduce variability, only stabilization (S2), standing (S3), and pre-fall (S4) were analyzed. In 10-subject experiments using 36 features (including mean, standard deviation, and skewness of position and velocity), SVM (>99%) and random forest (>98%) outperformed XGBoost (∼95%). SHAP analysis confirmed lateral dynamic characteristics in S2 and S4 contribute to classification, demonstrating high accuracy and interpretability.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Kojima et al. (2026) studied this question.

synapsesocial.com/papers/6a1fc3a6dee9eb8c0dce524b https://doi.org/10.1541/ieejeiss.146.624

View at Publisher

Ask AI

Bookmark

Share

Improving Identification Accuracy and Feature Analysis in Individual Identification Using Center of Gravity Sway

Key Points

Implication

Abstract

Discussion

Cite This Study

Improving Identification Accuracy and Feature Analysis in Individual Identification Using Center of Gravity Sway

Key Points

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading