Improving Detection Accuracy for Imbalanced Network Intrusion Classification using Cluster-based Under-sampling with Random Forests

Miah, Md. Ochiuddin; Khan, Sakib Shahriar; Shatabda, Swakkhar; Farid, Dewan Md.

doi:10.1109/icasert.2019.8934495

Cited by 46 publications

(46 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…19 and 20 [13]. The lowest threshold is considered through a line, y = x in au-ROC curve where correctly classified data points represent 1 and misclassified instances reveal as 0 [38].…”

Section: Methodsmentioning

confidence: 99%

“…Random forest is an ensemble approach to classify a high volume of data with 315 superior accuracy [38]. Initially, it splits a set of instances, D into numerous subsets D 1 , D 2 , ..., D n depending on the number of features F .…”

Section: Random Forestmentioning

confidence: 99%

“…Initially, it splits a set of instances, D into numerous subsets D 1 , D 2 , ..., D n depending on the number of features F . Then, it constructs multiple decision trees based on the number of subsets and returns the most popular class among the trees as prediction [38,19]. The learning scheme of random forest is tabulated in Algorithm 1.…”

Section: Random Forestmentioning

confidence: 99%

“…There is an association 320 between the accuracy and number of engendered trees, the number of trees is directly proportional to accuracy. In machine learning, overfitting is one of the crucial problems and it may abate the classifier accuracy [38]. Random Forest classifier has overcome this difficulty.…”

Section: Random Forestmentioning

confidence: 99%

“…Random Forest classifier has overcome this difficulty. As it considers the vote of generated every decision trees, it will not overfit the model and gain greater accuracy than 325 a single classifier model [38,1].…”

Section: Random Forestmentioning

confidence: 99%

See 4 more Smart Citations

Prediction of Motor Imagery Tasks from Multi-Channel EEG Data for Brain-Computer Interface Applications

Miah

Rahman

Muhammod

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

The classification of motor imagery electroencephalogram (MI-EEG) is a pivotal part of the biosignal classification in the brain-computer interface (BCI) applications. Currently, this bio-engineering based technology is being employed by researchers in various fields to develop cutting edge applications. The classification of real-time MI-EEG signal is the core computing and challenging task in these applications. It is well-known that the existing classification methods are not so accurate due to the high dimensionality and dynamic behaviors of the real-time EEG data. To improve the classification performance of real-time BCI applications, this paper presents a clustering-based ensemble technique and a developed brain game that distinguishes different human thoughts. At first, we have gathered the brain signals, extracted and selected informative features from these signals to generate training and testing sets. After that, we have constructed several classifiers using Artificial Neural Network (ANN), Support Vector Machine (SVM), naïve Bayes, Decision Tree (DT), Random Forest, Bagging, AdaBoost and compared the performance of these existing approaches with suggested clustering-based ensemble technique. On average, the proposed ensemble technique improved the classification accuracy of roughly 5 to 15% compared to the existing methods. Finally, we have developed the targeted * Corresponding author Email address: dewanfarid@cse.uiu.ac.bd (Dewan Md. Farid) brain game employing our suggested ensemble technique. In this game, realtime EEG signal classification and prediction tabulation through animated ball are controlled via threads. By playing this game, users can control the movements of the balls via the brain signals of motor imagery movements without using any traditional input devices. All relevant codes are available via open repository at: https://github.com/mrzResearchArena/MI-EEG. Keywords: Brain Computer Interface (BCI); Brain Machine Interface (BMI); Brain Engineering; Motor Imagery Electroencephalogram (MI-EEG); Motor Cortex; Clustering; Ensemble Learning. 45 are not so accurate due to the high dimensionality and dynamic behaviors of the real-time EEG data [19, 20]. Sometimes, the signals are biased with artifacts and noise due to the low conductivity of the electrodes with the scalp [1]. The objective of this paper is to extend our prior works and ameliorates the classification performance by handling multiple electrodes/ neurons data at the same 50 time. The proposed clustering-based ensemble technique clustered the dataset 3 based on the position of the electrodes so that each cluster represents dissimilar information. It also selects the model dynamically based on the electrode locations to classify real-time EEG data. We have constructed several classifiers using Artificial Neural Network (ANN), Support Vector Machine (SVM), naïve 55Bayes, Decision Tree, Random Forest, Bagging, AdaBoost and compared the performance of these existing approaches with the suggested technique. We also developed a brai...

show abstract

“…19 and 20 [13]. The lowest threshold is considered through a line, y = x in au-ROC curve where correctly classified data points represent 1 and misclassified instances reveal as 0 [38].…”

Section: Methodsmentioning

confidence: 99%

Section: Random Forestmentioning

confidence: 99%

Section: Random Forestmentioning

confidence: 99%

Section: Random Forestmentioning

confidence: 99%

Section: Random Forestmentioning

confidence: 99%

See 3 more Smart Citations