Correlation Based Feature Selection Algorithms for Varying Datasets of Different Dimensionality

Kowshalya, A. Meena; Madhumathi, R; Gopika, N.

doi:10.1007/s11277-019-06504-w

Cited by 28 publications

(11 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Different approaches have been introduced and implemented by researchers. The well-known approaches are based on the correlation between the attribute that selects the highly relevant features in the available dataset such as the correlation-based feature selection (CFS) [19], maximal discernibility pairs, Gaussian kernel, fuzzy rough sets [5], and measuring dependency through the Chromosomal fitness score like the rough set-based genetic algorithm and Rough sets-based incremental calculation dependency [20].…”

Section: Feature Reduction Methodsmentioning

confidence: 99%

Enhancing Big Data Feature Selection Using a Hybrid Correlation-Based Feature Selection

et al. 2021

View full text Add to dashboard Cite

This study proposes an alternate data extraction method that combines three well-known feature selection methods for handling large and problematic datasets: the correlation-based feature selection (CFS), best first search (BFS), and dominance-based rough set approach (DRSA) methods. This study aims to enhance the classifier’s performance in decision analysis by eliminating uncorrelated and inconsistent data values. The proposed method, named CFS-DRSA, comprises several phases executed in sequence, with the main phases incorporating two crucial feature extraction tasks. Data reduction is first, which implements a CFS method with a BFS algorithm. Secondly, a data selection process applies a DRSA to generate the optimized dataset. Therefore, this study aims to solve the computational time complexity and increase the classification accuracy. Several datasets with various characteristics and volumes were used in the experimental process to evaluate the proposed method’s credibility. The method’s performance was validated using standard evaluation measures and benchmarked with other established methods such as deep learning (DL). Overall, the proposed work proved that it could assist the classifier in returning a significant result, with an accuracy rate of 82.1% for the neural network (NN) classifier, compared to the support vector machine (SVM), which returned 66.5% and 49.96% for DL. The one-way analysis of variance (ANOVA) statistical result indicates that the proposed method is an alternative extraction tool for those with difficulties acquiring expensive big data analysis tools and those who are new to the data analysis field.

show abstract

Section: Feature Reduction Methodsmentioning

confidence: 99%

Enhancing Big Data Feature Selection Using a Hybrid Correlation-Based Feature Selection

et al. 2021

View full text Add to dashboard Cite

show abstract

“…The function evaluates attribute vector subsets that are correlated with the class label but not with each other. The CFS algorithm considers that irrelevant features have a low correlation with the class and should thus be discarded [31]. The CFS algorithm is applied to the NSL KDD dataset with 41 attributes, out of which 29 attributes have shown high correlation with the output variable and thus remaining attributes having low correlation are eliminated from the used dataset.…”

Section: Solution Methodologymentioning

confidence: 99%

Accuracy Based Fault Tolerant Two Phase - Intrusion Detection System (TP-IDS) Using Machine Learning and HDFS

Jadhav¹,

Pellakuri²

2021

RIA

View full text Add to dashboard Cite

“…A feature is regarded as excellent if it is always substantially linked with the class and is not redundant with other relevant characteristics. Correlation‐based feature selection entails picking relevant characteristics from a class, finding redundant features, and removing them from the original data 68 …”

Section: Feature Selectionmentioning

confidence: 99%

ONOS Flood Defender: An Intelligent Approach to Mitigate DDoS Attack in SDN

Aslam

Srivastava

Gore

2022

Trans Emerging Tel Tech

View full text Add to dashboard Cite

Software-Defined Networking (SDN) has made its place in the networks as new technology. SDN's programmable behavior enables it to change behavior on the fly, provides instructions for the task's automatic performance, dynamic scaling, and service integration. These advantages have made SDN necessary in networks. However, SDN suffers from the threat of DDoS attack. We have developed an approach to mitigate these threats by creating an ONOS Flood Defender Application (OFD App). This app effectively detects DDoS attack using supervised and ensemble machine learning techniques and mitigates them by tracebacking the attack traffic to its origin. Our results show that ensemble machine learning techniques perform better than single machine learning algorithm to detect DDoS attack. Random Forest classifier (RFC), an ensemble technique, performs best with the highest accuracy of 99.3% in detecting DDoS attack, followed by XGBoost classifier with an accuracy of 99%. The proposed framework of the OFD app is implemented in a Mininet emulator and ONOS SDN controller. The application performs effectively in terms of time, accuracy, and overhead of the system. Our app efficiently mitigates the attacks, thereby preventing a tremendous amount of damage to legitimate users.

show abstract

Correlation Based Feature Selection Algorithms for Varying Datasets of Different Dimensionality

Cited by 28 publications

References 14 publications

Enhancing Big Data Feature Selection Using a Hybrid Correlation-Based Feature Selection

Enhancing Big Data Feature Selection Using a Hybrid Correlation-Based Feature Selection

Accuracy Based Fault Tolerant Two Phase - Intrusion Detection System (TP-IDS) Using Machine Learning and HDFS

ONOS Flood Defender: An Intelligent Approach to Mitigate DDoS Attack in SDN

Contact Info

Product

Resources

About