Multi-class System Research Articles

Kernel rootkit prevention model using multiclass

Malicious individuals can access a computer network or application thanks to a series of programmes known as rootkit malware. These kernel rootkits use covert methods to conceal the kernel components, various control frameworks, and system activities, making identifying or prohibiting their presence in the target machine challenging. The bulk of rootkit detection and prevention techniques used today are particular to a system and dependent on recognized sources, making them ineffective for growing, evolving, concealed, or unnamed rootkits. This study proposes using the kernel rootkit prevention model using multiclass (KRPMM) system to identify hash values and detect/prevent such rootkits. The file downloaded by the client, who is availing of the service, is not permitted into the node used by the client in the cloud. But, it is redirected to the node wherein the file that has been downloaded and has entered the node anew is examined by a program which is specially coded to test the presence of rootkit in the file by some mechanisms and then comes to a conclusion of either the file being malicious or the file being clean and is free of rootkits. KRPMM tested only 64 rootkits.

Convolutional neural network-based real-time mosquito genus identification using wingbeat frequency: A binary and multiclass classification approach

Global rises in dengue hemorrhagic fever, especially in Asia and Latin America, underscore the necessity for enhanced public health interventions. Aedes spp. mosquitoes are the primary vectors; however, species such as Culex quinquefasciatus pose significant health risks by transmitting diseases such as filariasis, impacting millions of people worldwide. This study introduces a real-time convolutional neural network-based mosquito classification system using wingbeat frequency for identifying various mosquito species, with emphasis on Aedes sp. We proposed and assessed two models: a binary classification and a multiclass system. The binary system exhibited an outstanding accuracy of 91.76% in distinguishing between Aedes aegypti and Culex quinquefasciatus. The multiclass system accurately identified female and male Aedes aegypti and Culex quinquefasciatus with a precision of 87.16%. This innovative approach serves as a potential tool for dengue infection control and a versatile instrument for combating various mosquito-borne illnesses, enhancing vector surveillance for comprehensive disease management.

Production Systems with Parallel Heterogeneous Servers of Limited Capacity: Accurate Modeling and Performance Analysis

Heterogeneous systems of limited capacity have general applications in manufacturing, but also in logistic or service systems due to the differences in server or workstation performance or work assignment; this is in close relationship with system flexibility, where saturation and blocking are ordinary situations of systems with high demand and limited capacity, and thus, accurate loss quantification is essential for performance evaluation. Multi-class systems of limited capacity have been studied much less than parallel homogeneous systems (Erlang models). In this context, accurate models for parallel heterogeneous ordered-entry systems were developed: without any prior queue, i.e., M/Mi/c/c, and with a k-capacity queue, i.e., M/Mi/c/c + k. These new matrix models gave an exact state formulation, and their accuracy was verified using discrete event simulation and comparison with literature results. Also, the effect of the queue capacity was studied in relationship to the pattern of service rates. Next, the heterogeneous recirculating system model was also developed with good approximation results. Finally, the proposed models were applied to evaluate systems with non-exponential service times using a new hybrid methodology by combining the Markovian model and the Monte Carlo method (MCM) for normal or lognormal service times, which also yielded useful good approximations to the simulated system.

Convolutional Tsetlin Machine-based Training and Inference Accelerator for 2-D Pattern Classification

The Tsetlin Machine (TM) is a machine learning algorithm based on an ensemble of Tsetlin Automata (TAs) that learns propositional logic expressions from Boolean input features. In this paper, the design and implementation of a Field Programmable Gate Array (FPGA) accelerator based on the Convolutional Tsetlin Machine (CTM) is presented. The accelerator performs classification of two pattern classes in 4 × 4 Boolean images with a 2 × 2 convolution window. Specifically, there are two separate TMs, one per class. Each TM comprises 40 propositional logic formulas, denoted as clauses, which are conjunctions of literals. Include/exclude actions from the TAs determine which literals are included in each clause. The accelerator supports full training, including random patch selection during convolution based on parallel reservoir sampling across all clauses. The design is implemented on a Xilinx Zynq XC7Z020 FPGA platform. With an operating clock speed of 40 MHz, the accelerator achieves a classification rate of 4.4 million images per second with an energy per classification of 0.6 μJ. The mean test accuracy is 99.9% when trained on the 2-dimensional Noisy XOR dataset with 40% noise in the training labels. To achieve this performance, which is on par with the original software implementation, Linear Feedback Shift Register (LFSR) random number generators of minimum 16 bits are required. The solution demonstrates the core principles of a CTM and can be scaled to operate on multi-class systems for larger images.

A Geometric Backoff Time Distribution of IEEE 802.11 DCF: An Analytical Study

Today’s networks, including WLANs, transportdifferent classes of service. A service differentiation is thenessential to provide QoS. However, IEEE 802.11 for WLANs wasprimarily designed for best effort traffic and did not provide QoSspecifications. IEEE 802.11e MAC has been then described tosupport QoS in WLAN. In this paper, we propose a new schemefor service differentiation which is based on the 802.11 standardand requires minor modifications. In fact, we act on DCF whichuses the backoff procedure to solve contention in WLANs.For this scheme, we use, instead of the uniform distribution,a geometric distribution for random backoff time selection.Using a multi-class system, we propose three parametrizationsof the geometric distribution which imply different dynamicdifferentiation modes and we provide an analytical study, usinga Markov chain model, to compare our differentiation modes.We discuss our numerical results which give the performancesevaluation of the proposed mechanism in term of throughputand delay.

Autonomous diagnosis of pediatric cutaneous vascular anomalies using a convolutional neural network

ObjectivesDesign and validate a novel handheld device for the autonomous diagnosis of pediatric vascular anomalies using a convolutional neural network (CNN). Study designRetrospective, cross-sectional study of medical images. Computer aided design and 3D printed manufacturing. MethodsWe obtained a series of head and neck vascular anomaly images in pediatric patients from the database maintained in a large multidisciplinary vascular anomalies clinic. The database was supplemented with additional images from the internet. Four diagnostic classes were recognized in the dataset – infantile hemangioma, capillary malformation, venous malformation, and arterio-venous malformation. Our group designed and implemented a convolutional neural network to recognize the four classes of vascular anomalies as well as a fifth class consisting of none of the vascular anomalies. The system was based on the Inception-Resnet neural network using transfer learning. For deployment, we designed and built a compact, handheld device including a central processing unit, display subsystems, and control electronics. The device focuses upon and autonomously classifies pediatric vascular lesions. ResultsThe multiclass system distinguished the diagnostic categories with an overall accuracy of 84%. The inclusion of lesion metadata improved overall accuracy to 94%. Sensitivity ranged from 88% (venous malformation) to 100% (arterio-venous malformation and capillary malformation). ConclusionsAn easily deployed handheld device to autonomously diagnose pediatric skin lesions is feasible. Large training datasets and novel neural network architectures will be required for successful implementation.

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.

Load-balancing for multi-skilled servers with Bernoulli routing

We study the optimal Bernoulli routing in a multiclass queueing system with a dedicated server for each class as well as a common (or multi-skilled) server that can serve jobs of all classes. Jobs of each class arrive according to a Poisson process. Each server has a holding cost per customer and use the processor sharing discipline for service. The objective is to minimize the weighted mean holding cost. First, we provide conditions under which classes send their traffic only to their dedicated server, only to the common server, or to both. A fixed point algorithm is given for the computation of the optimal solution. We then specialize to two classes and give explicit expressions for the optimal loads. Finally, we compare the cost of multi-skilled server with that of only dedicated or all common servers. The theoretical results are complemented by numerical examples that illustrate the various structural results as well as the convergence of the fixed point algorithm.

Bounding the Inefficiency of the Multiclass, Multicriteria C-Logit Stochastic User Equilibrium in a Transportation Network

We derive the exact inefficiency upper bounds of the multiclass C-Logit stochastic user equilibrium (CL-SUE) in a transportation network. All travelers are classified on the basis of different values of time (VOT) into M classes. The multiclass CL-SUE model gives a more realistic path choice probability in comparison with the logit-based stochastic user equilibrium model by considering the overlapping effects between paths. To find efficiency loss upper bounds of the multiclass CL-SUE, two equivalent variational inequalities for the multiclass CL-SUE model, i.e., time-based variational inequality (VI) and monetary-based VI, are formulated. We give four different methods to define the inefficiency of the multiclass CL-SUE, i.e., to compare multiclass CL-SUE with multiclass system optimum, or to compare multiclass CL-SUE with multiclass C-Logit stochastic system optimum (CL-SSO), under the time-based criterion and the monetary-based criterion, respectively. We further investigate the effects of various parameters which include the degree of path overlapping (the commonality factor), the network complexity, degree of traffic congestion, the VOT of user classes, the network familiarity, and the total demand on the inefficiency bounds.

Mind your own customers and ignore the others: Asymptotic optimality of a local policy in multi-class queueing systems with customer feedback

This work contributes to the investigation of optimal routing and scheduling policies in multi-class multi-server queueing systems with customer feedback. We propose a new policy, dubbed local policy that requires access to only local queue information. Our new local policy specifies how an idle server chooses the next customer by using the queue length information of not all queues, but only those this server is eligible to serve. To gain useful insights and mathematical tractability, we consider a simple W model with customer feedback, and we establish limit theorems to show that our local policy is asymptotically optimal among all policies that may use the global system information, with the objective of minimizing the cumulative queueing costs measured by convex functions of the queue lengths. Numerical experiments provide convincing engineering confirmations of the effectiveness of our local policy for both W model and a more general non-W model.

Fluctuation limits for mean-field interacting nonlinear Hawkes processes

We investigate the asymptotic behavior of networks of interacting non-linear Hawkes processes modeling a homogeneous population of neurons in the large population limit. In particular, we prove a functional central limit theorem for the mean spike-activity thereby characterizing the asymptotic fluctuations in terms of a stochastic Volterra integral equation. Our approach differs from previous approaches in making use of the associated resolvent in order to represent the fluctuations as Skorokhod continuous mappings of weakly converging martingales. Since the Lipschitz properties of the resolvent are explicit, our analysis in principle also allows to derive approximation errors in terms of driving martingales. We also discuss extensions of our results to multi-class systems.

A Hybrid Multi-Class MAC Protocol for IoT-Enabled WBAN Systems

This study proposes a hybrid MAC protocol that can efficiently and effectively optimize the communication channel access of a WBAN multi-class system. The proposed protocol consists of two major processes that include the contention phase (CP) and the transmission phase (TP). In the CP, only the biomedical devices that have health packets to transmit randomly contend with equal probabilities using a slotted ALOHA scheme for transmission opportunities and the successful biomedical devices are allocated a transmission time-slot by employing a reservation-based time division multiple access (TDMA) scheme in the transmission phase. A multi-objective optimization problem was formulated to maximize the system sum-throughput, packet success-access-ratio, as well as the reservation ratio, and solved by the controller (i.e., access point) to determine the optimal length of the CP and the number of biomedical devices that can transmit in the TP. Monte Carlo simulation was performed and the optimization solution improved the proposed protocol’s performances. For validation purposes, the simulated results in MATLAB revealed that the proposed protocol performs better than the contemporary system in the context of the system sum-throughput, reservation ratio, and the average health packet delay with performance gains of about 9.2%, 9.5%, and 9.6% respectively.

Approximations for dynamic multi-class manufacturing systems with priorities and finite buffers

Capacity planning models for tactical to operational decisions in manufacturing systems require a performance evaluation component that relates demand processes with production resources and system state. Steady-state queueing models are widely used for such performance evaluations. However, these models typically assume stationary demand processes. With shorter new product development and life cycles, increasing customization, and constantly evolving customer preferences the assumption of stationary demand processes is not always reasonable. Nonstationary demand processes can capture the dynamic nature of modern manufacturing systems. The ability to analyze dynamic manufacturing systems with multiple products and finite buffers is essential for estimating throughput rates, throughput times, and work-in-process levels as well as evaluating the impact of proposed capacity plans and resource allocations. In this article, we present computationally efficient numerical approximations for the performance evaluation of dynamic multi-product manufacturing systems with priorities and finite buffers. The approximation breaks time into short periods, estimates throughput and arrival rates based on system status and current arrival processes, and then pieces the periods together through flow balance equations. The dynamic nature of product demands is modeled through non-homogeneous Poisson processes. The performance of these approximations is presented for practically sized flowshops and jobshops.

Machine Learning Models Reveal the Importance of Clinical Biomarkers for the Diagnosis of Alzheimer's Disease.

Alzheimer's Disease (AD) is a neurodegenerative disease that causes complications with thinking capability, memory and behavior. AD is a major public health problem among the elderly in developed and developing countries. With the growth of AD around the world, there is a need to further expand our understanding of the roles different clinical measurements can have in the diagnosis of AD. In this work, we propose a machine learning-based technique to distinguish control subjects with no cognitive impairments, AD subjects, and subjects with mild cognitive impairment (MCI), often seen as precursors of AD. We utilized several machine learning (ML) techniques and found that Gradient Boosting Decision Trees achieved the highest performance above 84% classification accuracy. Also, we determined the importance of the features (clinical biomarkers) contributing to the proposed multi-class classification system. Further investigation on the biomarkers will pave the way to introduce better treatment plan for AD patients.

Ensemble-learning based neural networks for novelty detection in multi-class systems

In most real-world systems or processes, determining the complete set of classes during the training phase is generally impossible. There is a high chance that novelties or abnormal data can appear in future phases which might severely affect the performance of the machine learning system. Novelty detection is of great importance in many critical systems and domains, such as business intelligence, process monitoring, information security, clinical decision support etc. Most of the available methods for novelty detection use a one-class classification (OCC) criterion, i.e. treating multiple known classes as a single ”Normal” class, whose aim is to distinguish data samples between “Normal” and “Not Normal” classes. In this paper, the problem of novelty detection in multi-class systems is addressed through ensemble based learning of neural networks (EBNN), capable of both detecting novelties and classifying the known normal samples in future datasets. Moreover, the model is analogous to the semi-supervised learning system as it is trained using only the available normal classes. Evaluation of the proposed model (EBNN) on UCI machine learning datasets showed that the model not only outperforms other models in detecting novelties but also has a better multi-class classification accuracy for known normal classes. The proposed model implements a novel activation function in its framework and differs from the commonly available novelty detection models in three aspects. First, the model is much simpler to implement and does not need any initial assumptions about the model. Second, the model does not require any novel or abnormal data during training phase (semi-supervised learning). Third, it can be used as a two in one system to detect novelties and at the same time to classify data based on known classes.

Decoding covert visual attention based on phase transfer entropy

Covert attention to spatial and color features in the visual field is a relatively new control signal for brain-computer interfaces (BCI). To guide the processing resources to the related visual scene aspects, covert attention should be decoded from human brain. Here, a novel expert system is designed to decode covert visual attention based on the EEG signal provided from 15 subjects during a new task based on a change in lumination to two blue and orange color on the right and the left side of the screen, which is evaluated in two cases of binary and multi-class systems. For the first time, Phase transfer entropy (PTE) has been used in these systems, and after selecting the optimal decoding feature, the frequency band (8-13 Hz) Alpha and Beta1 (13-20 Hz) have the best performance compared to other frequency bands. Two-class classification accuracies of the designed system in two frequency bands (Alpha and Beta1) are 91.87% and 89.53%, respectively. Also, the accuracies are 65.11% and 63.38% for multi-class classification in specified frequency bands. In these frequency bands, the parietal and frontal lobes showed the most significant difference in comparison to the other lobes. Also, the obtained results declared that the expert system's performance in the Alpha band by the extracted features from the Posterior region is better than all frequency bands in other different brain regions. The performance of the proposed expert system by PTE is significantly better than the previous phase synchronization based features. Results have shown that the PTE feature performs better than the common methods for decoding covert visual attention.

Optical Biopsy: Automated Classification of Airway Endoscopic Findings Using a Convolutional Neural Network.

Create an autonomous computational system to classify endoscopy findings. Computational analysis of vocal fold images at an academic, tertiary-care laryngology practice. A series of normal and abnormal vocal fold images were obtained from the image database of an academic tertiary care laryngology practice. The benign images included normals, nodules, papilloma, polyps, and webs. A separate set of carcinoma and leukoplakia images comprised a single malignant-premalignant class. All images were classified with their existing labels. Images were randomly withheld from each class for testing. The remaining images were used to train and validate a neural network for classifying vocal fold lesions. Two classifiers were developed. A multiclass system classified the five categories of benign lesions. A separate analysis was performed using a binary classifier trained to distinguish malignant-premalignant from benign lesions. Precision ranged from 71.7% (polyps) to 89.7% (papilloma), and recall ranged from 70.0% (papilloma) to 88.0% (nodules) for the benign classifier. Overall accuracy for the benign classifier was 80.8%. The binary classifier correctly identified 92.0% of the malignant-premalignant lesions with an overall accuracy of 93.0%. Autonomous classification of endoscopic images with artificial intelligence technology is possible. Better network implementations and larger datasets will continue to improve classifier accuracy. A clinically useful optical cancer screening system may require a multimodality approach that incorporates nonvisual spectra. NA Laryngoscope, 132:S1-S8, 2022.

Scheduling Queues with Customer Transfers

We consider a multiclass single-server system wherein customers of each class form their own queue and can either abandon the system or transfer to other queues while waiting. A system manager makes dynamic scheduling decisions to minimize system-wide waiting cost. We solve the stochastic control problem under both infinite-horizon discounted and average cost criteria. By studying a Brownian control problem that arises in the conventional heavy traffic regime, we identify (under each cost criterion) an asymptotically optimal policy that prioritizes across different classes according to a workload-dependent dynamic index rule. The proposed policies acknowledge the impact of customer transfers on the optimal allocation of service resources and generate rich managerial insights. One of the technical difficulties in the proof of the asymptotic optimality result is to identify a natural stability condition and formally show that the sequence of diffusion-scaled processes is uniformly stable under that condition.

Performance analysis and stability of multiclass orbit queue with constant retrial rates and balking

In this paper, we consider a single-server retrial model with multiple classes of customers. Arrival of customers follow independent Poisson rule. A new customer, facing a busy server upon his arrival, may join the corresponding (class-dependent) orbit queue with a class-dependent probability, or leaves the system forever (balks). The orbit queues follow constant retrial rate discipline, that is, only one (oldest) orbital customer of each orbit queue makes attempts to occupy the server, in a gap of class-dependent exponential times. Within each class, service times are assumed to be independent and identically distributed (iid). We show that this setting generalizes the so-called two-way communication systems.This multiclass system with general service time distributions is analysed using regenerative approach. Necessary and sufficient stability conditions, as well as some explicit expressions for the basic steady-state probabilities, are obtained. A restricted, two-way communication model with exponential service time distributions, is analysed by matrix-analytic method. Moreover, we combine both methods to efficiently derive explicit solutions for the restricted model.An extensive simulation analysis is performed to gain deep insight into the model stability and performance. It is shown that both the simulated and exact results agree on some important measures for which analytical expressions are available, and hence establish the validity of our theoretical treatment. We numerically study the sophisticated dependence structure of the model to uncover the orbits interaction. We give further details and intuitive explanation for the system performance which complements the derived explicit expressions.

Multiclass AQM on a TCP/IP router: A control theory approach

SummaryActive queue management (AQM) is a well‐known technique to improve routing performance under congested traffic conditions. It is often deployed to regulate queue sizes, thus aiming for constant transmission delay. This work addresses AQM using an approach based on control theory ideas. Compared with previous results in the literature, the novelty is the consideration of heterogeneous traffic, ie, multiclass traffic. Thus, each traffic class may have different discarding policies, queue sizes, and bandwidth share. This feature brings the proposal nearer to real network management demands than previous approaches in the literature. The proposed technique assumes that each class already has a simple controller, designed a priori, and focuses on designing a static state‐feedback controller for the multiclass system, where the design is based on using LMIs for the calculations. For this, optimization problems with LMI constraints are proposed to compute the state‐feedback gains that ensure stability for a large set of admissible initial conditions. These conditions ensure not only closed‐loop stability but also some level of performance. As far as we know, this is the first control theory based approach for the AQM problem on TCP/IP routers that allows a multiclass AQM while also considering time‐varying delays and input saturation. This is an important step to frame AQM in a more formal, yet realistic context, enabling it to address important service level agreement (SLA) directives. The proposal is tested on a simulated system at the end of this paper, showing the feasibility and performance of the approach in the presence of multiclass traffic.