A Survey on Feature Selection and Extraction Techniques for High-Dimensional Microarray Datasets

Manikandan, G.; Abirami, S.

doi:10.1007/978-981-10-8258-0_14

Cited by 18 publications

(10 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, feature selection could result in the improvement of performance during classification as well as make IDSs lightweight. One study [115] offered the following benefits of feature selection: (1) Reduced dimensionality of feature space, to limit storage requirements and increase algorithm speed.…”

Section: Feature Reduction (Feature Selection) Techniques For Ids On Iotmentioning

confidence: 99%

“…Further objectives of feature selection as informed by [115,118] include (i) increasing the speed of classifiers and reducing the capacity of storage; (ii) conserving resources for upcoming data collection processes; (iii) obtaining additional information and comprehensive knowledge of the process of data generation. In the area of IDS, feature selection has been utilized in improving processes of classification.…”

Section: Feature Selectionmentioning

confidence: 99%

“…Filters methods can be further categorized into two groups, namely the feature weighting procedure and subset search procedure. The feature weighting procedure assigns weights to features individually and ranks them based on their relevance to the target concept [115,120]. In filter methods, the feature selection process is optimized for the classification procedure.…”

Section: Feature Selectionmentioning

confidence: 99%

“…Furthermore, dimensionality reduction, referred to also as the extraction of features, entails that there would be a decrease in the feature space size, without any loss of information from the initial feature space. One of the disadvantages of extracting features is the lack of interpretability and details of the lost features [115]. Some vital features that could be possessed by IDS as informed by the literature are as follows [115,118]: (1) It is tolerant to fault, and continually runs with little or no need for human supervision.…”

Section: The Advantage Of Feature Extraction and Selectionmentioning

confidence: 99%

“…One of the disadvantages of extracting features is the lack of interpretability and details of the lost features [115]. Some vital features that could be possessed by IDS as informed by the literature are as follows [115,118]: (1) It is tolerant to fault, and continually runs with little or no need for human supervision. Furthermore, the IDS has the ability to recover from accidental or malicious activity causing system crashes.…”

Section: The Advantage Of Feature Extraction and Selectionmentioning

confidence: 99%

See 4 more Smart Citations

Classifier Performance Evaluation for Lightweight IDS Using Fog Computing in IoT Security

et al. 2021

View full text Add to dashboard Cite

In this article, a Host-Based Intrusion Detection System (HIDS) using a Modified Vector Space Representation (MVSR) N-gram and Multilayer Perceptron (MLP) model for securing the Internet of Things (IoT), based on lightweight techniques and using Fog Computing devices, is proposed. The Australian Defence Force Academy Linux Dataset (ADFA-LD), which contains exploits and attacks on various applications, is employed for the analysis. The proposed method is divided into the feature extraction stage, the feature selection stage, and classification modeling. To maintain the lightweight criteria, the feature extraction stage considers a combination of 1-gram and 2-gram for the system call encoding. In addition, a Sparse Matrix is used to reduce the space by keeping only the weight of the features that appear in the trace, thus ignoring the zero weights. Subsequently, Linear Correlation Coefficient (LCC) is utilized to compensate for any missing N-gram in the test data. In the feature selection stage, the Mutual Information (MI) method and Principle Component Analysis (PCA) are utilized and then compared to reduce the number of input features. Following the feature selection stage, the modeling and performance evaluation of various Machine Learning classifiers are conducted using a Raspberry Pi IoT device. Further analysis of the effect of MLP parameters, such as the number of nodes, number of features, activation, solver, and regularization parameters, is also conducted. From the simulation, it can be seen that different parameters affect the accuracy and lightweight evaluation. By using a single hidden layer and four nodes, the proposed method with MI can achieve 96% accuracy, 97% recall, 96% F1-Measure, 5% False Positive Rate (FPR), highest curve of Receiver Operating Characteristic (ROC), and 96% Area Under the Curve (AUC). It also achieved low CPU time usage of 4.404 (ms) milliseconds and low energy consumption of 8.809 (mj) millijoules.

show abstract

Section: Feature Reduction (Feature Selection) Techniques For Ids On Iotmentioning

confidence: 99%

Section: Feature Selectionmentioning

confidence: 99%

Section: Feature Selectionmentioning

confidence: 99%

Section: The Advantage Of Feature Extraction and Selectionmentioning

confidence: 99%

Section: The Advantage Of Feature Extraction and Selectionmentioning

confidence: 99%

See 3 more Smart Citations

Classifier Performance Evaluation for Lightweight IDS Using Fog Computing in IoT Security

et al. 2021

View full text Add to dashboard Cite

show abstract

A Review of Feature Reduction Methods for QSAR-Based Toxicity Prediction

Idakwo

Luttrell

Chen

et al. 2019

Challenges and Advances in Computational Chemistry and Physics

View full text Add to dashboard Cite

Feature Mining and Classification of Microarray Data Using Modified ResNet-SVM Architecture

Bej¹,

Roy²,

Suchinta³

et al. 2021

Computational Intelligence in Pattern Recognition

View full text Add to dashboard Cite

Microarray data have been one of the torch-bearing approaches when ushered for prediction and diagnosis of deadly diseases. But, much of its potential has been overshadowed by its structural orientation of low sample and high dimension which have provided the dilemma for researchers working in this field of study. In recent times, many approaches have been adopted to tackle this dilemma through feature extraction and thereafter classification that have stapled very concrete results. But, the methodologies lacked direct implementation and required the transformation of the data into a relevant intermediate matrix, before being mobilized for feature extraction and classification. In this paper, an effective hybrid alternative has been proposed which utilizes a modified single-dimensional (1D) Residual Network (ResNet) for lucid feature extraction directly from the gathered datasets and a Support Vector Machine (SVM) classifier for diseases classification. The proposed methodology is tested on three datasets (namely Colon cancer, Leukemia, and Lung cancer) and derived a staggering classification accuracy.

show abstract

A Survey on Feature Selection and Extraction Techniques for High-Dimensional Microarray Datasets

Cited by 18 publications

References 52 publications

Classifier Performance Evaluation for Lightweight IDS Using Fog Computing in IoT Security

Classifier Performance Evaluation for Lightweight IDS Using Fog Computing in IoT Security

A Review of Feature Reduction Methods for QSAR-Based Toxicity Prediction

Feature Mining and Classification of Microarray Data Using Modified ResNet-SVM Architecture

Contact Info

Product

Resources

About