Shih-Feng Huang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Shih-Feng Huang has a total of 49 co-authors. Here are the top co-authors by number of shared publications:
Meihui Guo
Department of Applied Mathematics, National Sun Yat-sen University, Kaohsiung, Taiwan, ROC
5 shared publications- Meihui Guo
Department of Applied Mathematics, National Sun Yat-sen University, Kaohsiung, Taiwan
3 shared publications - Chi-Hsiang Chu
Department of Statistics, Tunghai University, Taichung 407224, Taiwan
2 shared publications Chih-Hao Chang
2 shared publications- Tzung-Pei Hong
Department of Computer Science and Information Engineering, National University of Kaohsiung, Kaohsiung 811726, Taiwan
2 shared publications - Yi-Ting Chen
Footprintku Inc., Kaohsiung, Taiwan (R.O.C.)
2 shared publications
Latest Publications
- 22 May 2026
- Published by Springer Nature in Journal of Classification
- p. 1-25
Temporal and Spatial Alignment-Based Approaches for Text Recognition
- 22 May 2026
- Published by Springer Nature in Journal of Classification
- p. 1-25
Temporal and Spatial Alignment-Based Approaches for Text Recognition
- 1 October 2025
- Published by Pleiades Publishing Ltd in Lobachevskii Journal of Mathematics
- Vol. 46 (10) , 5127-5142
Electronic Component Parameter Recognition by a Nonparametric Approach with an Assignment Algorithm
This study presents an automatic system to identify essential design parameters from observed features when drawing plans for electronic circuit board footprints. The system helps to reduce the error rate of manual recognition and enhances identification accuracy, thereby improving the efficiency of producing high-quality footprints and contributing to the production of superior consumer electronics. A nonparametric approach is proposed to define a score matrix for evaluating the relationships between the observed qualitative/quantitative features and the essential design parameters in a robust manner. The best matching result for the observed features and essential design parameters is obtained using an assignment algorithm with the proposed score matrix. A user-friendly sufficient condition is introduced to ensure the correct assignment produced by the proposed method. In addition, a consistent estimator of this sufficient condition is developed to facilitate practical implementation. Numerical results from the simulation study highlight the importance of this condition, demonstrating that distinct observed features can provide complementary information for accurate recognition. In the empirical study, we apply the proposed method to 15 electronic component packages and achieve satisfactory correct assignment percentages in identifying essential design parameters.
- 1 October 2025
- Published by Pleiades Publishing Ltd in Lobachevskii Journal of Mathematics
- Vol. 46 (10) , 5127-5142
Electronic Component Parameter Recognition by a Nonparametric Approach with an Assignment Algorithm
This study presents an automatic system to identify essential design parameters from observed features when drawing plans for electronic circuit board footprints. The system helps to reduce the error rate of manual recognition and enhances identification accuracy, thereby improving the efficiency of producing high-quality footprints and contributing to the production of superior consumer electronics. A nonparametric approach is proposed to define a score matrix for evaluating the relationships between the observed qualitative/quantitative features and the essential design parameters in a robust manner. The best matching result for the observed features and essential design parameters is obtained using an assignment algorithm with the proposed score matrix. A user-friendly sufficient condition is introduced to ensure the correct assignment produced by the proposed method. In addition, a consistent estimator of this sufficient condition is developed to facilitate practical implementation. Numerical results from the simulation study highlight the importance of this condition, demonstrating that distinct observed features can provide complementary information for accurate recognition. In the empirical study, we apply the proposed method to 15 electronic component packages and achieve satisfactory correct assignment percentages in identifying essential design parameters.
- 16 September 2025
- Published by Elsevier in Computational Statistics & Data Analysis
- Vol. 214
An algorithm for estimating threshold boundary regression models
- 16 September 2025
- Published by Elsevier in Computational Statistics & Data Analysis
- Vol. 214
An algorithm for estimating threshold boundary regression models
- 21 June 2024
- Published by Elsevier in International Journal of Forecasting
- Vol. 41 (1) , 345-360
A modified VAR-deGARCH model for asynchronous multivariate financial time series via variational Bayesian inference
- 21 June 2024
- Published by Elsevier in International Journal of Forecasting
- Vol. 41 (1) , 345-360
A modified VAR-deGARCH model for asynchronous multivariate financial time series via variational Bayesian inference
- 28 March 2024
- Published by Springer Nature
- p. 131-141
Automatic Identification of Table Contents in Electronic Component Specifications of EDA
- 28 March 2024
- Published by Springer Nature
- p. 131-141
Automatic Identification of Table Contents in Electronic Component Specifications of EDA
- 13 December 2023
- Published by Springer Nature in Statistics and Computing
- Vol. 34 (1) , 1-23
ECOPICA: empirical copula-based independent component analysis
- 13 December 2023
- Published by Springer Nature in Statistics and Computing
- Vol. 34 (1) , 1-23
ECOPICA: empirical copula-based independent component analysis
- 13 June 2023
- Published by Wiley in Asia-Pacific Journal of Financial Studies
- Vol. 52 (3) , 374-393
Less Volatile Value‐at‐Risk Estimation Under a Semi‐parametric Approach*
- 13 June 2023
- Published by Wiley in Asia-Pacific Journal of Financial Studies
- Vol. 52 (3) , 374-393
Less Volatile Value‐at‐Risk Estimation Under a Semi‐parametric Approach*
- Open Access
- 7 July 2022
- Published by Springer Nature in Scientific Reports
- Vol. 12 (1) , 1-9
Using the Kriging Correlation for unsupervised feature selection problems
This paper proposes a KC Score to measure feature importance in clustering analysis of high-dimensional data. The KC Score evaluates the contribution of features based on the correlation between the original features and the reconstructed features in the low dimensional latent space. A KC Score-based feature selection strategy is further developed for clustering analysis. We investigate the performance of the proposed strategy by conducting a study of four single-cell RNA sequencing (scRNA-seq) datasets. The results show that our strategy effectively selects important features for clustering. In particular, in three datasets, our proposed strategy selected less than 5% of the features and achieved the same or better clustering performance than when using all of the features.
- 7 July 2022
- Published by Springer Nature in Scientific Reports
- Vol. 12 (1) , 1-9
Open AccessUsing the Kriging Correlation for unsupervised feature selection problems
This paper proposes a KC Score to measure feature importance in clustering analysis of high-dimensional data. The KC Score evaluates the contribution of features based on the correlation between the original features and the reconstructed features in the low dimensional latent space. A KC Score-based feature selection strategy is further developed for clustering analysis. We investigate the performance of the proposed strategy by conducting a study of four single-cell RNA sequencing (scRNA-seq) datasets. The results show that our strategy effectively selects important features for clustering. In particular, in three datasets, our proposed strategy selected less than 5% of the features and achieved the same or better clustering performance than when using all of the features.
- Open Access
Classification of Plank Techniques Using Wearable Sensors
The plank is a common core-stability exercise. Developing a wearable inertial sensor system for distinguishing between acceptable and aberrant plank techniques and detecting specific deviations from acceptable plank techniques can enhance performance and prevent injury. The purpose of this study was to develop an inertial measurement unit (IMU)-based plank technique quantification system. Nineteen healthy volunteers (age: 20.5 ± 0.8 years, BMI: 22.9 ± 1.4 kg/m2) performed the standard plank technique and six deviations with five IMUs positioned on the occiput, cervical spine, thoracic spine, sacrum, and right radius to record movements. The random forest method was employed to perform the classification. The proposed binary tree classification model achieved an accuracy of more than 86%. The average sensitivities were higher than 90%, and the specificities were higher than 91%, except for one deviation (83%). These results suggest that the five IMU-based systems can classify the plank technique as acceptable or aberrant with good accuracy, high sensitivity, and acceptable specificity, which has significant implications in monitoring plank biomechanics and enabling coaching practice.
Open AccessClassification of Plank Techniques Using Wearable Sensors
The plank is a common core-stability exercise. Developing a wearable inertial sensor system for distinguishing between acceptable and aberrant plank techniques and detecting specific deviations from acceptable plank techniques can enhance performance and prevent injury. The purpose of this study was to develop an inertial measurement unit (IMU)-based plank technique quantification system. Nineteen healthy volunteers (age: 20.5 ± 0.8 years, BMI: 22.9 ± 1.4 kg/m2) performed the standard plank technique and six deviations with five IMUs positioned on the occiput, cervical spine, thoracic spine, sacrum, and right radius to record movements. The random forest method was employed to perform the classification. The proposed binary tree classification model achieved an accuracy of more than 86%. The average sensitivities were higher than 90%, and the specificities were higher than 91%, except for one deviation (83%). These results suggest that the five IMU-based systems can classify the plank technique as acceptable or aberrant with good accuracy, high sensitivity, and acceptable specificity, which has significant implications in monitoring plank biomechanics and enabling coaching practice.
- 13 January 2022
- Published by Elsevier in Applied Energy
- Vol. 309
Forecasting of high-resolution electricity consumption with stochastic climatic covariates via a functional time series approach
- 13 January 2022
- Published by Elsevier in Applied Energy
- Vol. 309
Forecasting of high-resolution electricity consumption with stochastic climatic covariates via a functional time series approach