Linear Discriminant Classifier Research Articles

Clustered event related spectral perturbation (ERSP) feature in right hand motor imagery classification.

A technology that allows humans to interact with machines more directly and efficiently would be desirable. Research on brain-computer interfaces (BCIs) provides the possibility for computers to understand human thoughts in a straightforward manner thereby facilitating communication. As a branch of BCI research, motor imagery (MI) techniques analyze the brain signals and help people in many aspects such as rehabilitation, clinical applications, entertainment, and system controlling. In this study, an imagery experiment consisting of four kinds of right-hand movements (gripping, opening, pronation, and supination) was designed. Then a novel feature, namely, clustered feature was proposed based on the event-related spectral perturbation (ERSP) calculated from the EEG signal. Based on the selected features, two classical classifiers (support vector machine and linear discriminant classifier) were trained, achieving an acceptable accurate result (80%, on average).

Choosing an Optimal Machine Learning Classifier for Facial Recognition

Facial recognition is vital in today's technological world. There is a need to develop facial recognition systems which have high accuracy and at the same time high input to output response. To overcome the problem of low accuracy of facial recognition system we have attempted to combine different algorithm’s and develop a high accuracy facial recognition system. On combining different algorithms, we have observed that the Random Forest classifier achieved 96% accuracy in 2 secs. On the other hand, the Linear Discriminant Analysis classifier achieved 97% in 0 secs. Here we will compare the 6 classifiers to choose a optimal classifier.

Lead Detection in the Arctic Ocean from Sentinel-3 Satellite Data: A Comprehensive Assessment of Thresholding and Machine Learning Classification Methods

In the Arctic Ocean, obtaining water levels from satellite altimetry is hampered by the presence of sea ice. Hence, water level retrieval requires accurate detection of fractures in the sea ice (leads). This paper describes a thorough assessment of various surface type classification methods, including a thresholding method, nine supervised-, and two unsupervised machine learning methods, applied to Sentinel-3 Synthetic Aperture Radar Altimeter data. For the first time, the simultaneously sensed images from the Ocean and Land Color Instrument, onboard Sentinel-3, were used for training and validation of the classifiers. This product allows to identify leads that are at least 300 meters wide. Applied to data from winter months, the supervised Adaptive Boosting, Artificial Neural Network, Naïve-Bayes, and Linear Discriminant classifiers showed robust results with overall accuracies of up to 92%. The unsupervised Kmedoids classifier produced excellent results with accuracies up to 92.74% and is an attractive classifier when ground truth data is limited. All classifiers perform poorly on summer data, rendering surface classifications that are solely based on altimetry data from summer months unsuitable. Finally, the Adaptive Boosting, Artificial Neural Network, and Bootstrap Aggregation classifiers obtain the highest accuracies when the altimetry observations include measurements from the open ocean.

Cross-Platform Implementation of an SSVEP-Based BCI for the Control of a 6-DOF Robotic Arm.

Robotics has been successfully applied in the design of collaborative robots for assistance to people with motor disabilities. However, man-machine interaction is difficult for those who suffer severe motor disabilities. The aim of this study was to test the feasibility of a low-cost robotic arm control system with an EEG-based brain-computer interface (BCI). The BCI system relays on the Steady State Visually Evoked Potentials (SSVEP) paradigm. A cross-platform application was obtained in C++. This C++ platform, together with the open-source software Openvibe was used to control a Stäubli robot arm model TX60. Communication between Openvibe and the robot was carried out through the Virtual Reality Peripheral Network (VRPN) protocol. EEG signals were acquired with the 8-channel Enobio amplifier from Neuroelectrics. For the processing of the EEG signals, Common Spatial Pattern (CSP) filters and a Linear Discriminant Analysis classifier (LDA) were used. Five healthy subjects tried the BCI. This work allowed the communication and integration of a well-known BCI development platform such as Openvibe with the specific control software of a robot arm such as Stäubli TX60 using the VRPN protocol. It can be concluded from this study that it is possible to control the robotic arm with an SSVEP-based BCI with a reduced number of dry electrodes to facilitate the use of the system.

Real-Time Locomotion Recognition Algorithm for an Active Pelvis Orthosis to Assist Lower-Limb Amputees

Powered hip exoskeletons, in combination with passive prostheses, have been recently proposed to improve the economy and pattern of walking of lower-limb amputees within clinical scenarios. However, for everyday life support, a real-time control strategy that can accurately recognize different locomotion modes and transitions is required. In this letter, we proposed a novel locomotion recognition algorithm for an Active Pelvis Orthosis designed to assist people with lower-limb amputation, in quasi-static (sit-to-stand/stand-to-sit) and dynamic locomotion modes (walking and stairs negotiation). Two finite-state machines were combined to recognize in real-time the participants’ locomotion, one was a rule-based algorithm and one was based on four linear discriminant analysis classifiers. Four transfemoral amputees took part in the experiments and performed a circuit of tasks in two conditions, namely in <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">transparent mode</i> (the exoskeleton was controlled to provide null output impedance), and in <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">assistive mode</i> (the exoskeleton was controlled to output an assistive torque consistently with the locomotion mode recognized by the algorithm), to test the algorithm in real-time conditions. The median (25 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">th</sup> , 75 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">th</sup> percentile) between-subjects recognition accuracy was 94.8% (93.4%, 96.5%) with user-dependent models. Offline analysis on user-independent models with leave-one-subject-out validation resulted in between-subjects recognition accuracy equal to 95.9% (94.0%, 97.8%). The results of this study pave the way for future experimentations of the technology in ecological scenarios.

Detection of liver abnormalities—A new paradigm in medical image processing and classification techniques

AbstractThe liver is the body's most essential organ, and all human activities are interrelated with normal liver function. Any malfunction of the liver may lead to fatal diseases; therefore, early detection of liver abnormalities is essential. Modern medical imaging techniques combined with engineering procedures are reducing human suffering caused by liver disease. This study uses multiple classifiers to detect liver cirrhosis in ultrasonic images. The ultrasound images were obtained from The Cancer Imaging Archive database. A gray‐level co‐occurrence matrix (GLCM) and statistical approaches are used to extract features from normal and liver‐cirrhosis images. The extracted GLCM features are normalized and classified using nonlinear regression, linear regression, logistic regression, Bayesian Linear Discriminant Classifiers (BLDC), Gaussian Mixture Model (GMM), Firefly, Cuckoo search, Particle Swarm Optimization (PSO), Elephant search, Dragon Fly, Firefly GMM, Cuckoo search GMM, PSO GMM, Elephant search GMM, and Dragon Fly GMM classifiers. Benchmark metrics, such as sensitivity, specificity, accuracy, precision, negative predictive value, false‐negative rate, balanced accuracy, F1 score, Mathew correlation coefficient, F measure, error rate, Jaccard metric, and classifier success index, are assessed to identify the best‐performing classifier. The GMM classifier outperformed other classifiers for statistical features, and it achieved the highest accuracy (98.39%) and lowest error rate (1.61%). Moreover, the Dragon Fly GMM classifier achieved 90.69% for the GLCM feature used to classify liver cirrhosis.

Multiclass recognition of AD neurological diseases using a bag of deep reduced features coupled with gradient descent optimized twin support vector machine classifier for early diagnosis

SummaryAlzheimer's disease (AD) is an advanced neurodegenerative disease of the brain that affects the nerve system of brain. Previously, several feature extraction and classification methods were discussed, but that methods provide high over fitting problem, which leads to minimization of detection accuracy. To overcome these issues, the multi class classification of AD diseases using bag of deep feature reduction technique and twin support vector machine classifier (TSVM) optimized with gradient decent optimizer is proposed in this manuscript for classifying the AD disease as severe AD, mild cognitive impairment, healthy control. At first, the input EEG signals are pre‐processed. To decrease the execution time and processing time with feature size, a bag of deep features reduction technique is used. The reduced feature signals are classified by optimized TSVM. The simulation process is implemented in MATLAB environment. The proposed model achieves higher accuracy 33.84%, 28.93%, 33.03%, 27.93%, higher precision 22.87%, 16.97%, 16.97%, and 36.97%, compared with the existing methods, such as piecewise aggregate approximation support vector machine (MCC‐EEG‐PAA‐SVM), convolutional neural network (MCC‐EEG‐CNN), conformal kernel‐based fuzzy support vector machine (MCC‐EEG‐CKF‐SVM), Pearson correlation coefficient‐based feature selection strategy with linear discriminant analysis classifier (MCC‐EEG‐ PCC‐LDA).

Comprehensive Analysis of DNA Methylation and Transcriptome to Identify PD-1-Negative Prognostic Methylated Signature in Endometrial Carcinoma.

Background Epigenetic mechanism plays an important role in endometrial carcinoma (EC). This study was designed to analyze the epigenetic mechanism between DNA methylation-driven genes (DEDGs) and drugs targeting DEDGs and to develop a DEDG score model for predicting the prognosis of EC. Methods Expression profile and methylation profile data of PD-1-negative EC samples were obtained from TCGA. To obtain intersected DEDGs, differentially expressed genes (DEGs) and differentially methylated genes from tumor tissues and normal tissues were analyzed by limma. A linear discriminant classification model was constructed using the gene expression profile of DMDGs, methylation profile of TSS1500, TSS200, and gene body regions. Principal component analysis (PCA) and ROC analysis were conducted. The protein-drug interactions analysis of DMDGs was performed using Network Analyst 3.0 tool. Lasso Cox regression analysis was used to screen prognostic DNA methylation driving gene and to build a risk score model. The ROC curve and Kaplan-Meier survival curve were plotted to evaluate the model prediction capability. Results A total of 96 DMDGs were screened from the three regions, distributed on 22 chromosomes, with consistent methylation patterns in different gene regions. Both the expression profile and methylation profile of the three regions can neatly distinguish tumor samples from normal ones, with a high classifying performance. A gene signature, which consisted of ELFN1-AS1 and ZNF132, could classify EC patients into a high-risk group and low-risk group. Prognosis of the high-risk group was significantly worse than that of the low-risk group. The risk model showed a high performance in predicting the prognosis of EC. Conclusion We successfully established a risk score system with two DMDGs, which showed a high prediction accuracy of EC prognosis.

Computer-aided diagnosis of COVID-19 from chest X-ray images using histogram-oriented gradient features and Random Forest classifier.

The decision-making process is very crucial in healthcare, which includes quick diagnostic methods to monitor and prevent the COVID-19 pandemic disease from spreading. Computed tomography (CT) is a diagnostic tool used by radiologists to treat COVID patients. COVID x-ray images have inherent texture variations and similarity to other diseases like pneumonia. Manually diagnosing COVID X-ray images is a tedious and challenging process. Extracting the discriminant features and fine-tuning the classifiers using low-resolution images with a limited COVID x-ray dataset is a major challenge in computer aided diagnosis. The present work addresses this issue by proposing and implementing Histogram Oriented Gradient (HOG) features trained with an optimized Random Forest (RF) classifier. The proposed HOG feature extraction method is evaluated with Gray-Level Co-Occurrence Matrix (GLCM) and Hu moments. Results confirm that HOG is found to reflect the local description of edges effectively and provide excellent structural features to discriminate COVID and non-COVID when compared to the other feature extraction techniques. The performance of the RF is compared with other classifiers such as Linear Regression (LR), Linear Discriminant Analysis (LDA), K-nearest neighbor (kNN), Classification and Regression Trees (CART), Random Forest (RF), Support Vector Machine (SVM), and Multi-layer perceptron neural network (MLP). Experimental results show that the highest classification accuracy (99. 73%) is achieved using HOG trained by using the Random Forest (RF) classifier. The proposed work has provided promising results to assist radiologists/physicians in automatic COVID diagnosis using X-ray images.

Artificial Intelligence: A Universal Virtual Tool to Augment Tutoring in Higher Education.

Artificial intelligence is an emerging technology that revolutionizes human lives. Despite the fact that this technology is used in higher education, many professors are unaware of it. In this current scenario, there is a huge need to arise, implement information bridge technology, and enhance communication in the classroom. Through this paper, the authors try to predict the future of higher education with the help of artificial intelligence. This research article throws light on the current education system the problems faced by the subject faculties, students, changing government rules, and regulations in the educational sector. Various arguments and challenges on the implementation of artificial intelligence are prevailing in the educational sector. In this concern, we have built a use case model by using a student assessment data of our students and then built a synthesized using generative adversarial network (GAN). The dataset analyzed, visualized, and fed to different machine learning algorithms such as logistic Regression (LR), linear discriminant analysis (LDA), K-nearest neighbors (KNN), classification and regression trees (CART), naive Bayes (NB), support vector machines (SVM), and finally random forest (RF) algorithm and achieved a maximum accuracy of 58%. This article aims to bridge the gap between human lecturers and the machine. We are also concerned about the psychological emotions of the faculty and the students when artificial intelligence takes control.

MGA-IDS: Optimal feature subset selection for anomaly detection framework on in-vehicle networks-CAN bus based on genetic algorithm and intrusion detection approach

Controller area network (CAN) bus which provides efficient, reliable and robust communication between electronic control units (ECUs) is the most frequently used protocol for in-vehicle networks. However, the lack of security mechanisms in the CAN protocol makes it vulnerable to inside and outside cyber-attacks. Therefore, an intrusion detection system (IDS), which is a widely used method to detect malicious activities, is preferred to improve the security of the CAN buses. In spite of the fact that various supervised and unsupervised machine learning algorithms are employed to increase the performance of IDSs, obtaining high classification performance is still a challenge for them. First, a lot of irrelevant and redundant features in the datasets result in long computational times with low detection performances. Second, different classification performances are acquired based on classifiers and a combination of the features. Third, many of the models suffer from unknown and different types of attacks. For these reasons, a new intrusion detection framework is proposed in this paper based on feature selection and classifier. Initially, we propose a meta-heuristic algorithm called modified genetic algorithm (MGA) m-feature selection for dimension reduction by selecting optimal feature subset based on k-fold cross validation. Then, we utilize five different linear and nonlinear classifiers: support vector classifier (SVC), logistic regression classifier (LRC), decision tree classifier (DTC), k-nearest neighbors classifier (KNC), and linear discriminant analysis classifier (LDAC) as candidate classifiers to develop an efficient IDS. Finally, we select the best classifier from the candidates and build an IDS. The experimental results reveal that the proposed MGA-DTC presents a better performance in terms of several metrics based on not only the HCRL-car hacking dataset but also UNSW-NB15, and CIC-IDS2017 datasets.

A face recognition system using convolutional feature extraction with linear collaborative discriminant regression classification

Face recognition is one of the important biometric authentication research areas for security purposes in many fields such as pattern recognition and image processing. However, the human face recognitions have the major problem in machine learning and deep learning techniques, since input images vary with poses of people, different lighting conditions, various expressions, ages as well as illumination conditions and it makes the face recognition process poor in accuracy. In the present research, the resolution of the image patches is reduced by the max pooling layer in convolutional neural network (CNN) and also used to make the model robust than other traditional feature extraction technique called local multiple pattern (LMP). The extracted features are fed into the linear collaborative discriminant regression classification (LCDRC) for final face recognition. Due to optimization using CNN in LCDRC, the distance ratio between the classes has maximized and the distance of the features inside the class reduces. The results stated that the CNN-LCDRC achieved 93.10% and 87.60% of mean recognition accuracy, where traditional LCDRC achieved 83.35% and 77.70% of mean recognition accuracy on ORL and YALE databases respectively for the training number 8 (i.e. 80% of training and 20% of testing data).

Intrusion detection technique in wireless sensor network using grid search random forest with Boruta feature selection algorithm

Attacks in wireless sensor networks (WSNs) aim to prevent or eradicate the network's ability to perform its anticipated functions. Intrusion detection is a defense used in wireless sensor networks that can detect unknown attacks. Due to the incredible development in computer-related applications and massive Internet usage, it is indispensable to provide host and network security. The development of hacking technology tries to compromise computer security through intrusion. Intrusion detection system (IDS) was employed with the help of machine learning (ML) Algorithms to detect intrusions in the network. Classic ML algorithms like support vector machine (SVM), K-nearest neighbour (KNN), and filter-based feature selection often led to poor accuracy and misclassification of intrusions. This article proposes a novel framework for IDS that can be enabled by Boruta feature selection with grid search random forest (BFS-GSRF) algorithm to overcome these issues. The performance of BFS-GSRF is compared with ML algorithms like linear discriminant analysis (LDA) and classification and regression tree (CART) etc. The proposed work was implemented and tested on network security laboratory — knowledge on discovery dataset (NSL-KDD). The experimental results show that the proposed model BFS-GSRF yields higher accuracy (i.e., 99%) in detecting attacks, and it is superior to LDA, CART, and other existing algorithms.

Data Mining Techniques for Iraqi Biochemical Dataset Analysis

This research aims to analyze and simulate biochemical real test data for uncovering the relationships among the tests, and how each of them impacts others. The data were acquired from Iraqi private biochemical laboratory. However, these data have many dimensions with a high rate of null values, and big patient numbers. Then, several experiments have been applied on these data beginning with unsupervised techniques such as hierarchical clustering, and k-means, but the results were not clear. Then the preprocessing step performed, to make the dataset analyzable by supervised techniques such as Linear Discriminant Analysis (LDA), Classification And Regression Tree (CART), Logistic Regression (LR), K-Nearest Neighbor (K-NN), Naïve Bays (NB), and Support Vector Machine (SVM) techniques. CART gives clear results with high accuracy between the six supervised algorithms. It is worth noting that the preprocessing steps take remarkable efforts to handle this type of data, since its pure data set has so many null values of a ratio 94.8%, then it becomes 0% after achieving the preprocessing steps. Then, in order to apply CART algorithm, several determined tests were assumed as classes. The decision to select the tests which had been assumed as classes were depending on their acquired accuracy. Consequently, enabling the physicians to trace and connect the tests result with each other, which extends its impact on patients’ health.

Explainable Biomarkers for Automated Glomerular and Patient-Level Disease Classification.

Pathologists use multiple microscopy modalities to assess renal biopsy specimens. Besides usual diagnostic features, some changes are too subtle to be properly defined. Computational approaches have the potential to systematically quantitate subvisual clues, provide pathogenetic insight, and link to clinical outcomes. To this end, a proof-of-principle study is presented demonstrating that explainable biomarkers through machine learning can distinguish between glomerular disorders at the light-microscopy level. The proposed system used image analysis techniques and extracted 233 explainable biomarkers related to color, morphology, and microstructural texture. Traditional machine learning was then used to classify minimal change disease (MCD), membranous nephropathy (MN), and thin basement membrane nephropathy (TBMN) diseases on a glomerular and patient-level basis. The final model combined the Gini feature importance set and linear discriminant analysis classifier. Six morphologic (nuclei-to-glomerular tuft area, nuclei-to-glomerular area, glomerular tuft thickness greater than ten, glomerular tuft thickness greater than three, total glomerular tuft thickness, and glomerular circularity) and four microstructural texture features (luminal contrast using wavelets, nuclei energy using wavelets, nuclei variance using color vector LBP, and glomerular correlation using GLCM) were, together, the best performing biomarkers. Accuracies of 77% and 87% were obtained for glomerular and patient-level classification, respectively. Computational methods, using explainable glomerular biomarkers, have diagnostic value and are compatible with our existing knowledge of disease pathogenesis. Furthermore, this algorithm can be applied to clinical datasets for novel prognostic and mechanistic biomarker discovery.

A fisher linear discriminant analysis classifier fused with naïve Bayes for simultaneous detection in an asynchronous brain-computer interface

BackgroundAn asynchronous event-related potential-based brain computer interface (ERP-BCI) permits the subjects to output intentions at their own pace, which provides a more free and practical communication pathways without the need for muscle activity. The core of constructing this type of system is to discriminate both the intentions and brain states. New methodsThis study proposes a fisher linear discriminant analysis classification algorithm fused with naïve Bayes (B-FLDA) for the ERP-BCI to simultaneous recognize the subjects’ intentions, working and idle states. This method uses the spectral characteristics of visual-evoked potential and the time-domain characteristics of ERP to simultaneously detect brain states and target stimulus, and obtain the final discrimination result through probability fusion. ResultsThe accuracy and the information transfer rate increase to 98.61% and 62.80 bits/min under 10 repetitions and 1 repetition, respectively. The three parameters of receiver operator characteristic curve have achieved better performance. Comparison with existing methodsTen subjects participate in this study with the proposed algorithms and two other control methods. The accuracy and information transfer rate of this algorithm are better than the other methods. ConclusionsIt indicates that the naïve Bayes-FLDA algorithm is able to improve the performance of an asynchronous BCI system by detecting the intentions and states simultaneously.

Automated ECG multi-class classification system based on combining deep learning features with HRV and ECG measures

Electrocardiogram (ECG) serves as the gold standard for noninvasive diagnosis of several types of heart disorders. In this study, a novel hybrid approach of deep neural network combined with linear and nonlinear features extracted from ECG and heart rate variability (HRV) is proposed for ECG multi-class classification. The proposed system enhances the ECG diagnosis performance by combining optimized deep learning features with an effective aggregation of ECG features and HRV measures using chaos theory and fragmentation analysis. The constant-Q non-stationary Gabor transform technique is employed to convert the 1-D ECG signal into 2-D image which is sent to a pre-trained convolutional neural network structure, called AlexNet. The pair-wise feature proximity algorithm is employed to select the optimal features from the AlexNet output feature vector to be concatenated with the ECG and HRV measures. The concatenated features are sent to different types of classifiers to distinguish three distinct subjects, namely congestive heart failure, arrhythmia, and normal sinus rhythm (NSR). The results reveal that the linear discriminant analysis classifier has the highest accuracy compared to the other classifiers. The proposed system is investigated with real ECG data taken from well-known databases, and the experimental results show that the proposed diagnosis system outperforms other recent state-of-the-art systems in terms of accuracy 98.75%, specificity 99.00%, sensitivity of 98.18%, and computational time 0.15 s. This demonstrates that the proposed system can be used to assist cardiologists in enhancing the accuracy of ECG diagnosis in real-time clinical setting.

Effective face recognition using dual linear collaborative discriminant regression classification algorithm

Effective face recognition using dual linear collaborative discriminant regression classification algorithm

Using machine learning algorithms to multidimensional analysis of subjective thermal comfort in a library

The thermal comfort in public buildings is affected by multiple psychological and physical factors. A deep understanding of these factors is necessary for intelligent ventilation control and architectural design. In this study, it was quantified the physiological and psychological parameters of the occupants in a library situated in Changsha using a questionnaire. Then, with the use of Principal Component Analysis (PCA), the dimensionality of the database was reduced. We found that the Zero-mean and unit variance projection is better than the Zero-mean projection, and 43-dimensional data can replace more than 90% of the original 61-dimensional data. Machine learning algorithms were used to analyze the results after PCA, which showed that the effect of thermal comfort on emotions is greater than that of emotions on thermal comfort. In addition, the performance of six machine learning algorithms (Linear Regression (LR), Linear Discriminant Analysis (LDA), K-Nearest Neighbors (KNN), Classification and Regression Trees (CART), Gaussian Naive Bayes (NB), and Support Vector Machine (SVM)) were compared. It is found that the predicted results of LDA were more accurate, and other algorithms showed different performances in different cases. These findings can contribute to the study of the subjective and objective feelings of indoor thermal comfort in public buildings, thereby guiding architectural design, intelligent control of ventilation systems, and realizing human-building interaction interfaces.

Multivariate Specifications in the Mineral Processing Field: An introduction

The concept of multivariate specifications on incoming raw material was introduced in the mid-1990s in the chemical process industry as a tool to determine if a lot should be accepted from a supplier prior to its purchase. The objective of this paper is to adapt this concept to the mineral processing field using a simulation case study to assess the profitability of processing ore with different characteristics. To keep it simple, only the ore properties are considered to influence the final quality attributes: the concentrate flow rate and grade. Defining multivariate specification regions for raw ore properties is illustrated using simulated data from a grinding-flotation process where the feed ore average mineral grain size, grade and hardness are modified. This involves defining process performance classes based on an economic criterion, building a projection to latent structure PLS model, and adjusting statistical limits in the latent space of the model (i.e. the specification). The resulting specification region in the latent space based on a linear discriminant classifier allows to correctly classify 91% of the lots of ore in terms of their profitability.