Multiresolution hierarchical support vector machine for classification of large datasets

Alwajidi, Safaa; Li, Yang

doi:10.1007/s10115-022-01755-9

Cited by 3 publications

(4 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This study utilizes the hierarchical SVM construction method, categorizing the music performance movements recorded by wearable sensors into two initial motion subcategories. 6 It further refines these motion subcategories until a singular motion type classification is reached. Initially, based on the intensity of the music performance data captured by the wearable sensors, the movements of music performers are categorized into either gentle or intense motions.…”

Section: Classification Of Performance Skills Of Music Motionsmentioning

confidence: 99%

“…Since the eigenvector is formed by the linear combination of the elements of the projection space, the existence coefficient 𝛾 ij satisfies the Equation (6).…”

Section: Dimensionality Reductionmentioning

confidence: 99%

“…Additionally, the classification of various motion types collected by wearable sensors can also be achieved through indirect methods. This study utilizes the hierarchical SVM construction method, categorizing the music performance movements recorded by wearable sensors into two initial motion subcategories 6 . It further refines these motion subcategories until a singular motion type classification is reached.…”

Section: Dimensionality Reduction Of Performance Skills Of Music Base...mentioning

confidence: 99%

See 2 more Smart Citations

Wearing sensor data integration for promoting the performance skills of music in IoT

Li,

Shi,

Pan

2024

Internet Technology Letters

View full text Add to dashboard Cite

This study integrates multi‐node wearable sensor data to improve music performance skills. A window‐adding method is used during time‐frequency feature extraction. By incorporating kernel functions, we present a generalized discriminant analysis (GDA) method to reduce the high‐dimensional sensor features while retaining performance traits. Experiments demonstrate that the proposed GDA approach achieves higher accuracy (92.71%), precision (90.54%), and recall (88.68%) compared to linear discriminant analysis (82.39% accuracy) and principal component analysis (88.56% accuracy) in classifying motions performed by music performers. The integrated analysis of wearable sensor data facilitates comprehensive feedback to strengthen proficiency across various music performance skills.

show abstract

Section: Classification Of Performance Skills Of Music Motionsmentioning

confidence: 99%

“…Since the eigenvector is formed by the linear combination of the elements of the projection space, the existence coefficient 𝛾 ij satisfies the Equation (6).…”

Section: Dimensionality Reductionmentioning

confidence: 99%

Section: Dimensionality Reduction Of Performance Skills Of Music Base...mentioning

confidence: 99%

See 1 more Smart Citation

Wearing sensor data integration for promoting the performance skills of music in IoT

Li,

Shi,

Pan

2024

Internet Technology Letters

View full text Add to dashboard Cite

show abstract

“…The data's diversity also could extract a subset of attributes from the original attributes as a training set, generally accepted methods are random feature bagging [7] and empowerment feature generation bagging [8]. In this paper, the member classifiers are designed based on the Support vector machine [9], on the basis of the selfservice sampling strategy, the features are randomly selected based on the importance of the features as the probability of the features being selected. The selection of samples and features through the dual random process not only distinguishes the data, but also preserves as much sample information as possible.…”

Section: Introductionmentioning

confidence: 99%

Evaluate the performance of the support vector machines ensemble

Liu

Qiu

2023

Third International Conference on Artificial Intelligence and Computer Engineering (ICAICE 2022)

View full text Add to dashboard Cite

Diversity among the members of classifiers is deemed to be a key point in classifier ensemble. However, there doesn't exist a widely accepted diversity measure and construct. In this paper, we propose a sample and feature double random construction of training sample variability. A support vector machine is used as the base classifier to construct the difference by distinguishing the regularization term C and the kernel function. Based on the negative correlation theory, the base classifier generalization error and disparity judgment functions are proposed, and the base classifier is integrated by ranking according to the judgment functions, which could achieve a higher accuracy rate by the support vector machine ensemble.

show abstract