To improve the performance of multiple classifier system, a knowledge discovery based dynamic weighted voting （KD-DWV） is proposed based on knowledge discovery. In the method, all base classifiers may be allowed to ...To improve the performance of multiple classifier system, a knowledge discovery based dynamic weighted voting （KD-DWV） is proposed based on knowledge discovery. In the method, all base classifiers may be allowed to operate in different measurement/feature spaces to make the most of diverse classification information. The weights assigned to each output of a base classifier are estimated by the separability of training sample sets in relevant feature space. For this purpose, some decision tables （DTs） are established in terms of the diverse feature sets. And then the uncertainty measures of the separability are induced, in the form of mass functions in Dempster-Shafer theory （DST）, from each DTs based on generalized rough set model. From the mass functions, all the weights are calculated by a modified heuristic fusion function and assigned dynamically to each classifier varying with its output. The comparison experiment is performed on the hyperspectral remote sensing images. And the experimental results show that the performance of the classification can be improved by using the proposed method compared with the plurality voting （PV）.展开更多
The random forest model is universal and easy to understand, which is often used for classification and prediction. However, it uses non-selective integration and the majority rule to judge the final result, thus the ...The random forest model is universal and easy to understand, which is often used for classification and prediction. However, it uses non-selective integration and the majority rule to judge the final result, thus the difference between the decision trees in the model is ignored and the prediction accuracy of the model is reduced. Taking into consideration these defects, an improved random forest model based on confusion matrix (CM-RF)is proposed. The decision tree cluster is selectively constructed by the similarity measure in the process of constructing the model, and the result is output by using the dynamic weighted voting fusion method in the final voting session. Experiments show that the proposed CM-RF can reduce the impact of low-performance decision trees on the output result, thus improving the accuracy and generalization ability of random forest model.展开更多
Incredible progress has been made in human action recognition(HAR),significantly impacting computer vision applications in sports analytics.However,identifying dynamic and complex movements in sports like badminton re...Incredible progress has been made in human action recognition(HAR),significantly impacting computer vision applications in sports analytics.However,identifying dynamic and complex movements in sports like badminton remains challenging due to the need for precise recognition accuracy and better management of complex motion patterns.Deep learning techniques like convolutional neural networks(CNNs),long short-term memory(LSTM),and graph convolutional networks(GCNs)improve recognition in large datasets,while the traditional machine learning methods like SVM(support vector machines),RF(random forest),and LR(logistic regression),combined with handcrafted features and ensemble approaches,perform well but struggle with the complexity of fast-paced sports like badminton.We proposed an ensemble learning model combining support vector machines(SVM),logistic regression(LR),random forest(RF),and adaptive boosting(AdaBoost)for badminton action recognition.The data in this study consist of video recordings of badminton stroke techniques,which have been extracted into spatiotemporal data.The three-dimensional distance between each skeleton point and the right hip represents the spatial features.The temporal features are the results of Fast Dynamic Time Warping(FDTW)calculations applied to 15 frames of each video sequence.The weighted ensemble model employs soft voting classifiers from SVM,LR,RF,and AdaBoost to enhance the accuracy of badminton action recognition.The E2 ensemble model,which combines SVM,LR,and AdaBoost,achieves the highest accuracy of 95.38%.展开更多
Parkinson's disease(PD)is a neurodegenerative disorder characterized by motor and non-motor symptoms that significantly impact an individual's quality of life.Voice changes have shown promise as early indicato...Parkinson's disease(PD)is a neurodegenerative disorder characterized by motor and non-motor symptoms that significantly impact an individual's quality of life.Voice changes have shown promise as early indicators of PD,making voice analysis a valuable tool for early detection and intervention.This study aims to assess and detect the severity of PD through voice analysis using the mobile device voice recordings dataset.The dataset consisted of recordings from PD patients at different stages of the disease and healthy control subjects.A novel approach was employed,incorporating a voice activity detection algorithm for speech segmentation and the wavelet scattering transform for feature extraction.A Bayesian optimization technique is used to fine-tune the hyperparameters of seven commonly used classifiers and optimize the performance of machine learning classifiers for PD severity detection.AdaBoost and K-nearest neighbor consistently demonstrated superior performance across various evaluation metrics among the classifiers.Furthermore,a weighted majority voting(WMV)technique is implemented,leveraging the predictions of multiple models to achieve a near-perfect accuracy of 98.62%,improving classification accuracy.The results highlight the promising potential of voice analysis in PD diagnosis and monitoring.Integrating advanced signal processing techniques and machine learning models provides reliable and accessible tools for PD assessment,facilitating early intervention and improving patient outcomes.This study contributes to the field by demonstrating the effectiveness of the proposed methodology and the significant role of WMV in enhancing classification accuracy for PD severity detection.展开更多
Label fusion is a powerful image segmentation strategy that is becoming increasingly popular in medical imaging. However, satisfying the requirements of higher accuracy and less running time is always a great challeng...Label fusion is a powerful image segmentation strategy that is becoming increasingly popular in medical imaging. However, satisfying the requirements of higher accuracy and less running time is always a great challenge. In this paper we propose a novel patch-based segmentation method combining a local weighted voting strategy with Bayesian inference. Multiple atlases are registered to a target image by an advanced normalization tools(ANTs) algorithm. To obtain a segmentation of the target, labels of the atlas images are propagated to the target image. We first adopt intensity prior and label prior as two key metrics when implementing the local weighted voting scheme, and then compute the two priors at the patch level. Further, we analyze the label fusion procedure concerning the image background and take the image background as an isolated label when estimating the label prior. Finally, by taking the Dice score as a criterion to quantitatively assess the accuracy of segmentations, we compare the results with those of other methods, including joint fusion, majority voting, local weighted voting, majority voting based on patch, and the widely used Free Surfer whole-brain segmentation tool. It can be clearly seen that the proposed algorithm provides better results than the other methods. During the experiments, we make explorations about the influence of different parameters(including patch size, patch area, and the number of training subjects) on segmentation accuracy.展开更多
基金This project was supported by the National Basic Research Programof China (2001CB309403)
文摘To improve the performance of multiple classifier system, a knowledge discovery based dynamic weighted voting （KD-DWV） is proposed based on knowledge discovery. In the method, all base classifiers may be allowed to operate in different measurement/feature spaces to make the most of diverse classification information. The weights assigned to each output of a base classifier are estimated by the separability of training sample sets in relevant feature space. For this purpose, some decision tables （DTs） are established in terms of the diverse feature sets. And then the uncertainty measures of the separability are induced, in the form of mass functions in Dempster-Shafer theory （DST）, from each DTs based on generalized rough set model. From the mass functions, all the weights are calculated by a modified heuristic fusion function and assigned dynamically to each classifier varying with its output. The comparison experiment is performed on the hyperspectral remote sensing images. And the experimental results show that the performance of the classification can be improved by using the proposed method compared with the plurality voting （PV）.
基金Science Research Project of Gansu Provincial Transportation Department(No.2017-012)
文摘The random forest model is universal and easy to understand, which is often used for classification and prediction. However, it uses non-selective integration and the majority rule to judge the final result, thus the difference between the decision trees in the model is ignored and the prediction accuracy of the model is reduced. Taking into consideration these defects, an improved random forest model based on confusion matrix (CM-RF)is proposed. The decision tree cluster is selectively constructed by the similarity measure in the process of constructing the model, and the result is output by using the dynamic weighted voting fusion method in the final voting session. Experiments show that the proposed CM-RF can reduce the impact of low-performance decision trees on the output result, thus improving the accuracy and generalization ability of random forest model.
基金supported by the Center for Higher Education Funding(BPPT)and the Indonesia Endowment Fund for Education(LPDP),as acknowledged in decree number 02092/J5.2.3/BPI.06/9/2022。
文摘Incredible progress has been made in human action recognition(HAR),significantly impacting computer vision applications in sports analytics.However,identifying dynamic and complex movements in sports like badminton remains challenging due to the need for precise recognition accuracy and better management of complex motion patterns.Deep learning techniques like convolutional neural networks(CNNs),long short-term memory(LSTM),and graph convolutional networks(GCNs)improve recognition in large datasets,while the traditional machine learning methods like SVM(support vector machines),RF(random forest),and LR(logistic regression),combined with handcrafted features and ensemble approaches,perform well but struggle with the complexity of fast-paced sports like badminton.We proposed an ensemble learning model combining support vector machines(SVM),logistic regression(LR),random forest(RF),and adaptive boosting(AdaBoost)for badminton action recognition.The data in this study consist of video recordings of badminton stroke techniques,which have been extracted into spatiotemporal data.The three-dimensional distance between each skeleton point and the right hip represents the spatial features.The temporal features are the results of Fast Dynamic Time Warping(FDTW)calculations applied to 15 frames of each video sequence.The weighted ensemble model employs soft voting classifiers from SVM,LR,RF,and AdaBoost to enhance the accuracy of badminton action recognition.The E2 ensemble model,which combines SVM,LR,and AdaBoost,achieves the highest accuracy of 95.38%.
文摘Parkinson's disease(PD)is a neurodegenerative disorder characterized by motor and non-motor symptoms that significantly impact an individual's quality of life.Voice changes have shown promise as early indicators of PD,making voice analysis a valuable tool for early detection and intervention.This study aims to assess and detect the severity of PD through voice analysis using the mobile device voice recordings dataset.The dataset consisted of recordings from PD patients at different stages of the disease and healthy control subjects.A novel approach was employed,incorporating a voice activity detection algorithm for speech segmentation and the wavelet scattering transform for feature extraction.A Bayesian optimization technique is used to fine-tune the hyperparameters of seven commonly used classifiers and optimize the performance of machine learning classifiers for PD severity detection.AdaBoost and K-nearest neighbor consistently demonstrated superior performance across various evaluation metrics among the classifiers.Furthermore,a weighted majority voting(WMV)technique is implemented,leveraging the predictions of multiple models to achieve a near-perfect accuracy of 98.62%,improving classification accuracy.The results highlight the promising potential of voice analysis in PD diagnosis and monitoring.Integrating advanced signal processing techniques and machine learning models provides reliable and accessible tools for PD assessment,facilitating early intervention and improving patient outcomes.This study contributes to the field by demonstrating the effectiveness of the proposed methodology and the significant role of WMV in enhancing classification accuracy for PD severity detection.
基金Project supported by the National Natural Science Foundation of China(No.61203224)the Science and Technology Innovation Foundation of Shanghai Municipal Education Commission,China(No.13YZ101)
文摘Label fusion is a powerful image segmentation strategy that is becoming increasingly popular in medical imaging. However, satisfying the requirements of higher accuracy and less running time is always a great challenge. In this paper we propose a novel patch-based segmentation method combining a local weighted voting strategy with Bayesian inference. Multiple atlases are registered to a target image by an advanced normalization tools(ANTs) algorithm. To obtain a segmentation of the target, labels of the atlas images are propagated to the target image. We first adopt intensity prior and label prior as two key metrics when implementing the local weighted voting scheme, and then compute the two priors at the patch level. Further, we analyze the label fusion procedure concerning the image background and take the image background as an isolated label when estimating the label prior. Finally, by taking the Dice score as a criterion to quantitatively assess the accuracy of segmentations, we compare the results with those of other methods, including joint fusion, majority voting, local weighted voting, majority voting based on patch, and the widely used Free Surfer whole-brain segmentation tool. It can be clearly seen that the proposed algorithm provides better results than the other methods. During the experiments, we make explorations about the influence of different parameters(including patch size, patch area, and the number of training subjects) on segmentation accuracy.
