Md. Ochiuddin Miah

Bio: Md. Ochiuddin Miah is an academic researcher from United International University. The author has contributed to research in topics: Random forest & Naive Bayes classifier. The author has an hindex of 3, co-authored 5 publications receiving 17 citations.

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

Abstract: Network intrusion classification i n t he imbalanced big data environment becomes a significant and important issue in information and communications technology (ICT) in this digital era. Presently, intrusion detection systems (IDSs) are commonly using tool to detect and prevent internal and external network attacks/intrusions. IDSs are majorly bifurcated into host-based and network-based systems, and use pattern-matching techniques to detect intrusions that known as misuse-based intrusion detection system. Machine learning (ML) and data mining (DM) algorithms are widely using for classifying intrusions in IDS over the last few decades. One of the major challenges for building IDS employing machine learning and data mining algorithms is to improve the intrusion classification accuracy and also reducing the false-positive rate. In this paper, we have introduced a new method for improving detection rate to classify minority-class network attacks/ intrusions using cluster-based under-sampling with Random Forest classifier. The proposed method is a multi-layer classification approach, which can process the highly imbalanced big data to correctly identify the minority/ rare class-intrusions. Initially, the proposed method classify a data point/ incoming data is attack/ intrusion or not (like normal behaviour), if it’s an attack then the proposed method try to classify attack type and later sub-attack type. We have used cluster-based under-sampling technique to deal with class-imbalanced problem and popular ensemble classifier Random Forest for addressing overfitting problem. We have used KDD99 intrusion detection benchmark dataset for experimental analysis and tested the performance of proposed method with existing machine learning algorithms like: Artificial N eural Network (ANN), naive Bayes (NB) classifier, Random Forest, and Bagging techniques.

Real-Time EEG Classification of Voluntary Hand Movement Directions using Brain Machine Interface

Abstract: Intelligent systems for bio-signals processing and modelling are a method for creating signals to measure the brain activities to perform task by an external device. Brain Machine Interface (BMI) is a part of bioengineering that connects brain with machine directly in order to command and control the machines. Recently, bioengineering researchers are employing BMI techniques to explore advance knowledge for discovering biological fundamental problems. In this paper, we have proposed an ensemble method to improve the prediction accuracy of real time electroencephalogram signals classification and developed a system that can distinguish different human thoughts. Initially, we have collected the brain signals, and then extracted and selected informative features from these signals to engender training and testing data. We have built several classifiers using Artificial Neural Network (ANN), Decision Tree (DT), naive Bayes (NB) classifier, Bagging, Random Forest and compare the performance of these existing learning methods with proposed ensemble classifier. The proposed method achieved 99% and 79% accuracy on average for binary-class and ternary-class classification in compare with existing classifiers. Finally, we have applied the proposed ensemble classifier for developing a brain game that can control the ball using brain signal of voluntary movements without any need of conventional input devices.

Brain–Machine Interface for Developing Virtual-Ball Movement Controlling Game

Abstract: Intelligent Systems for bio-signals processing and modeling are a method for creating signals to measure the brain activities to perform a task by an external device. Brain–Machine Interface (BMI) that is also known as Brain–Computer Interface (BCI), Neural Control Interface (NCI), Mind–Machine Interface (MMI), and Direct Neural Interface (DNI) is a direct communication pathway between brain and machine. Recently, computational modeling researchers are applying BMI techniques to explore advanced knowledge for discovering biological fundamental problems. In this paper, we have explored BMI techniques and developed a system that can distinguish human thoughts. Initially, we have obtained the brain signals and extracted features from these signals to build training and test data. We have designed binary-class and three-class classifiers by employing OneR, naive Bayes (NB) classifier, decision tree (DT) induction, Random Forest, and Bagging classifiers. Random Forest achieved 93.16 and 62.84% accuracy for binary-class and three-class classification. On the contrary, decision tree (C4.5) classifier achieved 90.89 and 65.66% accuracy for binary-class and three-class classification. Then we have considered overall performance and applied decision tree classifier for developing an interactive game that can operate through brain–machine interface without physical interaction with the computer.

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

Abstract: 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), naive 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 brain game employing our suggested ensemble technique. In this game, real-time 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.

Brain Machine Interface

Abstract: How does the brain a lump of 'pinkish gray meat' produce the richness of conscious experience, or any subjective experience at all? Scientists and philosophers have historically likened the brain to contemporary information technology, from the ancient Greeks comparing memory to a 'seal ring in wax,' to the 19th century brain as a 'telegraph switching circuit, ' to Freud's subconscious desires 'boiling over like a steam engine,' to a hologram, and finally, the computer. [8]

EEG based stress level identification

Abstract: This paper investigates detection of patterns in brain waves while induced with mental stress. Electroencephalogram (EEG) is the most commonly used brain signal acquisition method as it is simple, economical and portable. An automatic EEG based stress recognition system is designed and implemented in this study with two effective stressors to induce different levels of mental stress. The Stroop colour-word test and mental arithmetic test are used as stressors to induce low level and high level of stress respectively, and their relevant C# applications are developed in Microsoft Visual Studio to interface with Emotiv Epoc device. Power band features from EEG signals are analyzed and using the relative difference of beta and alpha power as feature along with Support Vector Machine as classifier, three-levels of stress can be recognized with an accuracy of 75%. For two-level stress analysis, accuracy of 88% and 96% are achieved for Stroop colour-word test and mental arithmetic test respectively.

LEDPatNet19: Automated Emotion Recognition Model based on Nonlinear LED Pattern Feature Extraction Function using EEG Signals

Abstract: Electroencephalography (EEG) signals collected from human brains have generally been used to diagnose diseases. Moreover, EEG signals can be used in several areas such as emotion recognition, driving fatigue detection. This work presents a new emotion recognition model by using EEG signals. The primary aim of this model is to present a highly accurate emotion recognition framework by using both a hand-crafted feature generation and a deep classifier. The presented framework uses a multilevel fused feature generation network. This network has three primary phases, which are tunable Q-factor wavelet transform (TQWT), statistical feature generation, and nonlinear textural feature generation phases. TQWT is applied to the EEG data for decomposing signals into different sub-bands and create a multilevel feature generation network. In the nonlinear feature generation, an S-box of the LED block cipher is utilized to create a pattern, which is named as Led-Pattern. Moreover, statistical feature extraction is processed using the widely used statistical moments. The proposed LED pattern and statistical feature extraction functions are applied to 18 TQWT sub-bands and an original EEG signal. Therefore, the proposed hand-crafted learning model is named LEDPatNet19. To select the most informative features, ReliefF and iterative Chi2 (RFIChi2) feature selector is deployed. The proposed model has been developed on the two EEG emotion datasets, which are GAMEEMO and DREAMER datasets. Our proposed hand-crafted learning network achieved 94.58%, 92.86%, and 94.44% classification accuracies for arousal, dominance, and valance cases of the DREAMER dataset. Furthermore, the best classification accuracy of the proposed model for the GAMEEMO dataset is equal to 99.29%. These results clearly illustrate the success of the proposed LEDPatNet19.