Research on Sensor Motion State Recognition Model Based on Functional Kernel Logistic Ensemble Regression

View at publisher Summarise with VIRAT
Authors: Jun Liu
Cited by: 0
References: 30

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Motion state recognition, as a non-contact biometric identification technology, holds broad application prospects in fields such as intelligent security and health monitoring. Based on this, this paper proposes a functional kernel logistic ensemble regression model grounded in regenerative kernel theory. The proposed method utilizes regenerative kernel theory to model the nonlinear relationship between functional covariates and motion states. Furthermore, considering that varying numbers of basis functions in the functional principal component basis expansion may impact model performance during functional data approximation, this study employs ensemble learning principles to determine the optimal number of basis functions instead of relying on model selection. Moreover, ensemble learning algorithms balance variance and bias in predictive models while addressing parameter selection issues, thereby enhancing model robustness. Empirical analysis demonstrates the proposed method's competitive advantage over classical benchmark models.

Fields of Study & Subjects

Machine Learning and ELM Advanced Statistical Methods and Models Biometric Identification and Security

Publication details

Journal / source
IEEE International Conference on Power, Electronics and Green Energy (ICPEGE)
Pages
682–686
Publication date
28 January 2026
Language
EN