mRMR-PSO: A Hybrid Feature Selection Technique with a Multiobjective Approach for Sign Language Recognition

Bansal, Sandhya; Wadhawan, Savita; Goel, Rajeev K.

doi:10.1007/s13369-021-06456-z

Cited by 28 publications

(8 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To further demonstrate the superiority of the proposed feature selection strategy in this study, selection is performed separately using Relief F [14], Max-Relevance and Min-Redundancy (mRMR) [24], OFES [15], and the feature selection strategy proposed in this paper. Subsequently, different classification models are employed for fault diagnosis.…”

Section: Experimental Results Of Cwru Bearing Datasetmentioning

confidence: 99%

A bearing fault diagnosis method with improved symplectic geometry mode decomposition and feature selection

Chen,

Zheng

2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

A rolling bearing fault diagnosis method based on improved symplectic geometry mode decomposition and feature selection was proposed to solve the problem of low fault identification due to the influence of noise on early bearing fault features. First, the symplectic geometry mode decomposition is improved to enhance its robustness in decomposing signals with noise, then the time domain, frequency domain, and time-frequency features of each symplectic geometric component are extracted as feature vectors. Second, a comprehensive feature selection strategy is proposed to select the optimal subset of features that are conducive to fault classification. Finally, considering the problem of low classification accuracy of a single machine learning model, the AdaBoost-WSO-SVM model is constructed for fault classification using the AdaBoost algorithm of integrated learning. Experimental decomposition of complex signals with noise indicates that the improved symplectic geometry mode decomposition is more effective compared to traditional symplectic geometry mode decomposition. Subsequently, multiple experiments were conducted using the bearing datasets from Case Western Reserve University (CWRU) and Jiangnan University (JNU). The experimental results reveal that, after comprehensive feature selection and ensemble learning pattern recognition experiments on the CWRU dataset, the average accuracy of fault diagnosis can reach 99.67%. On the JNU dataset, the proposed fault diagnosis method achieves an average accuracy of 95.03%. This suggests that, compared to other feature selection methods and classification models, the proposed approach in this paper exhibits higher accuracy and generalization capabilities.

show abstract

Section: Experimental Results Of Cwru Bearing Datasetmentioning

confidence: 99%

A bearing fault diagnosis method with improved symplectic geometry mode decomposition and feature selection

Chen,

Zheng

2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…Bansal et al [26] designed a Minimum Redundancy and Maximum Relevance based Particle Swarm Optimization (mRMR-PSO) approach for sign language recognition. Here, publically available datasets were taken to identify three different sign languages and Multi-Class SVM model improved the classification accuracy while minimizing the number of features.…”

Section: Transformer and Optimization-based Sign Language Recognitionmentioning

confidence: 99%

Dynamic Doubled-handed sign language Recognition for deaf and dumb people using Vision Transformers

Vaidhya,

Anand

2024

Preprint

View full text Add to dashboard Cite

Sign language is an effective communication tool to convey information to each other, that is a bridge to reduce the communication gap between deaf and dumb people. The word level sign language recognition is a challenging task due to the wide range of body gestures, unidentified signals and hand configuration. To overcome this issue, a novel Inverted Residual Network Convolutional Vision Transformer based Mutation Boosted Tuna Swarm Optimization (IRNCViT-MBTSO) algorithm is proposed for recognizing double-handed sign language. The proposed dataset is designed to identify different dynamic words and the predicted images are preprocessed to enhance the generalization ability of the model and improve image quality. The local features are extracted after performing feature graining and the global features are captured from the preprocessed images by implementing the ViT transformer model. These extracted features are concatenated to generate a feature map and are classified into different dynamic words using the Inverted Residual Feed Forward Network (IRFFN). The TSO algorithm tunes the parameters of the IRNCViT model that is responsible for handling high-dimensional problems and convergence issues. The Mutation operator is introduced in this optimization phase to escape local optimum issues while updating the position of tuna. The performance valuation of this proposed model is performed in terms of recognition accuracy, convergence and visual output of the dataset that showed the highest performance than other state-of-the-art methods.

show abstract

“…To overcome the above-mentioned issue, feature selection is one of the most emerging solutions. The MRMR is a well-known feature selection technique for selecting features with minimum redundancy and maximum relevance [36,37]. Equation 11represents the microarray dataset with dimension m × n, where m stands for several samples and n stands for several features.…”

Section: Minimum Redundancy Maximum Relevance (Mrmr)mentioning

confidence: 99%

En-MinWhale: An Ensemble Approach Based on MRMR and Whale Optimization for Cancer Diagnosis

Panigrahi,

Pati,

Sahu

et al. 2023

IEEE Access

View full text Add to dashboard Cite

According to the WHO, Cancer is a prominent cause of mortality worldwide, accounting for ∼ 10 million fatalities at the end of 2020. The most common types of cancers include Lung, Breast, CNS, Leukemia, Colon, and Cervical Cancer. Early detection of cancer can decrease the death toll. According to the study, if the cancer is identified at its early stage, the death rate can be reduced to ∼85%. In order to reduce the death toll, machine learning (ML) emerges as a significant solution. When it comes to cancer research with ML, biopsy and microarray data come into the front. The biopsy data is less useful as it excludes a patient's genetic information. However, due to the genetic information, the microarray data emerges as a solution to detecting cancer disease. Dealing with microarray data also has some consequences, and high dimensionality is one of them. This article reports an ML-based ensemble model to tackle microarray data issues and provide an effective model for cancer detection. The reported model uses Minimum Redundancy Maximum Relevance (MRMR) as the feature selection algorithm. The Whale Optimization Algorithm (WOA) is implemented for the featured dataset to select the optimistic number of features without affecting microarray data relevance. Then, four ML-based classification models, including Support Vector Machine, Decision Tree, Multi-Layer Perceptron, and Random Forest, are applied as the base learners to make initial predictions. Finally, the voting ensemble technique is applied to the initial prediction to develop an ensemble model for cancer prediction. The proposed En-MinWhale model is evaluated over six different types of cancer microarray datasets, including Lung, Leukemia, Ovarian, CNS, Breast, and Colon Cancer. Finally, the performance of the proposed model is evaluated using 11 various evaluative parameters, including accuracy, precision, specificity, sensitivity, F-β score, etc. The En-MinWhale model shows 94.09%, 95.83%, 94.86%, 95.00%, 94.85%, and 96.77% accuracy for Lung, Leukemia, Ovarian, CNS, Breast, and Colon Cancer datasets, respectively, that outperforms other considered hybrid models and can help out the physicians in Cancer diagnosis.

show abstract

mRMR-PSO: A Hybrid Feature Selection Technique with a Multiobjective Approach for Sign Language Recognition

Cited by 28 publications

References 53 publications

A bearing fault diagnosis method with improved symplectic geometry mode decomposition and feature selection

A bearing fault diagnosis method with improved symplectic geometry mode decomposition and feature selection

Dynamic Doubled-handed sign language Recognition for deaf and dumb people using Vision Transformers

En-MinWhale: An Ensemble Approach Based on MRMR and Whale Optimization for Cancer Diagnosis

Contact Info

Product

Resources

About