Improvement In Performance Metrics Research Articles

Improving the prediction of school dropout with the support of the semi-supervised learning approach

School dropout is a phenomenon characterized by being influenced by several variables. This research used Machine Learning techniques, especially in the context of the semi-supervised learning strategy, to predict the risk of dropout in undergraduate courses at a Brazilian higher education institution. Two phases of experiments were conducted, the first using Feature Selection techniques and the second applying a semi-supervised learning strategy to improve performance metrics collected from the increase in the number of instances of students labeled as Graduated. As a main result, we obtained a model capable of classifying dropout with 90% accuracy and 86% Macro-F1.

Investigating the Increased CO2 Capture Performance of Amino Acid Functionalized Nanoporous Materials from First-Principles and Grand Canonical Monte Carlo Simulations

Nanoporous materials such as metal-organic frameworks (MOFs) and covalent-organic frameworks (COFs) have been identified as key candidates for environmental remediation through catalytic reduction and sequestration of pollutants. Given the prevalence of CO2 as a target molecule for capture, MOFs and COFs have seen a long history of application in the field. More recently, functionalized nanoporous materials have been demonstrated to improve performance metrics associated with the capture of CO2. We employ a multiscale computational approach including ab initio density functional theory (DFT) calculations and classical grand canonical Monte Carlo (GCMC) simulations, to investigate the impact of amino acid (AA) functionalization in three such nanoporous materials. Our results demonstrate a nearly universal improvement of CO2 uptake metrics such as adsorption capacity, accessible surface area, and CO2/N2 selectivity for six AAs. In this work, we elucidate the key geometric and electronic properties associated with improving the CO2 capture performance of functionalized nanoporous materials.

When performance is not enough-A multidisciplinary view on clinical decision support.

Scientific publications about the application of machine learning models in healthcare often focus on improving performance metrics. However, beyond often short-lived improvements, many additional aspects need to be taken into consideration to make sustainable progress. What does it take to implement a clinical decision support system, what makes it usable for the domain experts, and what brings it eventually into practical usage? So far, there has been little research to answer these questions. This work presents a multidisciplinary view of machine learning in medical decision support systems and covers information technology, medical, as well as ethical aspects. The target audience is computer scientists, who plan to do research in a clinical context. The paper starts from a relatively straightforward risk prediction system in the subspecialty nephrology that was evaluated on historic patient data both intrinsically and based on a reader study with medical doctors. Although the results were quite promising, the focus of this article is not on the model itself or potential performance improvements. Instead, we want to let other researchers participate in the lessons we have learned and the insights we have gained when implementing and evaluating our system in a clinical setting within a highly interdisciplinary pilot project in the cooperation of computer scientists, medical doctors, ethicists, and legal experts.

IIL13Pred: improved prediction of IL-13 inducing peptides using popular machine learning classifiers

BackgroundInflammatory mediators play havoc in several diseases including the novel Coronavirus disease 2019 (COVID-19) and generally correlate with the severity of the disease. Interleukin-13 (IL-13), is a pleiotropic cytokine that is known to be associated with airway inflammation in asthma and reactive airway diseases, in neoplastic and autoimmune diseases. Interestingly, the recent association of IL-13 with COVID-19 severity has sparked interest in this cytokine. Therefore characterization of new molecules which can regulate IL-13 induction might lead to novel therapeutics.ResultsHere, we present an improved prediction of IL-13-inducing peptides. The positive and negative datasets were obtained from a recent study (IL13Pred) and the Pfeature algorithm was used to compute features for the peptides. As compared to the state-of-the-art which used the regularization based feature selection technique (linear support vector classifier with the L1 penalty), we used a multivariate feature selection technique (minimum redundancy maximum relevance) to obtain non-redundant and highly relevant features. In the proposed study (improved IL-13 prediction (iIL13Pred)), the use of the mRMR feature selection method is instrumental in choosing the most discriminatory features of IL-13-inducing peptides with improved performance. We investigated seven common machine learning classifiers including Decision Tree, Gaussian Naïve Bayes, k-Nearest Neighbour, Logistic Regression, Support Vector Machine, Random Forest, and extreme gradient boosting to efficiently classify IL-13-inducing peptides. We report improved AUC, and MCC scores of 0.83 and 0.33 on validation data as compared to the current method.ConclusionsExtensive benchmarking experiments suggest that the proposed method (iIL13Pred) could provide improved performance metrics in terms of sensitivity, specificity, accuracy, the area under the curve - receiver operating characteristics (AUCROC) and Matthews correlation coefficient (MCC) than the existing state-of-the-art approach (IL13Pred) on the validation dataset and an external dataset comprising of experimentally validated IL-13-inducing peptides. Additionally, the experiments were performed with an increased number of experimentally validated training datasets to obtain a more robust model. A user-friendly web server (www.soodlab.com/iil13pred) is also designed to facilitate rapid screening of IL-13-inducing peptides.

Scalable and Data-Independent Multi-Agent Recommender System Using Social Networks Analysis

Nowadays, many online users find the selection of information and required products challenging due to the growing volume of data on the web. Recommender systems are introduced to deal with information overload. Cold start and data sparsity are the two primary issues in these systems, which lead to a decrease in the efficiency of recommender systems. To solve the problems, this paper proposes a novel method based on social network analysis. Our method leverages a multi-agent system for clustering users and items and predicting relationships between them simultaneously. The information on users and items is extracted from the user-item matrix as distinct graphs. Each of the graphs is then treated as a social network, which is further processed and analyzed by community detection and link prediction procedures. The users are grouped into several clusters by the community detection agent, which results in each cluster as a community. Then link prediction agent identifies the latent relationships between users and items. Simulation results show that the proposed method has significantly improved performance metrics as compared to recent techniques.

Application of A* Algorithm Based on Extended Neighborhood Priority Search in Multi-Scenario Maps

The robustness of the traditional A* algorithm of path planning is poor due to its excessive number of traversal nodes, slow search speed, and large turning angle. Aiming to solve the above problems, a multi-scenario adaptive A* algorithm based on extended neighborhood priority search is proposed. Firstly, this algorithm designs the heuristic function that can adapt to various scene changes by quantifying the scene map information, and the search weight is adjusted adaptively to enhance the robustness and adaptability of the algorithm. Secondly, the search strategy based on extended neighborhood priority is adopted to improve the orientation of the algorithm, and the redundant node removal strategy is used to smooth the path to reduce the number of traversed nodes and the turning angle. Finally, simulation tests are conducted in several representative map environments. The test results show that the proposed algorithm is superior to the traditional A* algorithm due to its stronger robustness and significantly improved performance metrics, with an 84.95% reduction in the number of traversal nodes, an 83.84% reduction in the number of path nodes, a 62.28% reduction in turning points on the path, a 77.38% reduction in the total turning angle, and a 58.47% reduction in the search time.

Deep Survival Analysis With Clinical Variables for COVID-19.

Millions of people have been affected by coronavirus disease 2019 (COVID-19), which has caused millions of deaths around the world. Artificial intelligence (AI) plays an increasing role in all areas of patient care, including prognostics. This paper proposes a novel predictive model based on one dimensional convolutional neural networks (1D CNN) to use clinical variables in predicting the survival outcome of COVID-19 patients. We have considered two scenarios for survival analysis, 1) uni-variate analysis using the Log-rank test and Kaplan-Meier estimator and 2) combining all clinical variables ([Formula: see text]=44) for predicting the short-term from long-term survival. We considered the random forest (RF) model as a baseline model, comparing to our proposed 1D CNN in predicting survival groups. Our experiments using the univariate analysis show that nine clinical variables are significantly associated with the survival outcome with corrected p < 0.05. Our approach of 1D CNN shows a significant improvement in performance metrics compared to the RF and the state-of-the-art techniques (i.e., 1D CNN) in predicting the survival group of patients with COVID-19. Our model has been tested using clinical variables, where the performance is found promising. The 1D CNN model could be a useful tool for detecting the risk of mortality and developing treatment plans in a timely manner. The findings indicate that using both Heparin and Exnox for treatment is typically the most useful factor in predicting a patient's chances of survival from COVID-19. Moreover, our predictive model shows that the combination of AI and clinical data can be applied to point-of-care services through fast-learning healthcare systems.

Optimized Resource Allocation and Queue Management for Traffic Control in MANET

A set of mobile devices that employs wireless transmission for communication is termed Mobile Ad hoc Networks (MANETs). Offering better communication services among the users in a centralized organization is the primary objective of the MANET. Due to the features of MANET, this can directly End-to-End Delay (EED) the Quality of Service (QoS). Hence, the implementation of resource management becomes an essential issue in MANETs. This paper focuses on the efficient Resource Allocation (RA) for many types of Traffic Flows (TF) in MANET. In Mobile Ad hoc Networks environments, the main objective of Resource Allocation (RA) is to process consistently available resources among terminals required to address the service requirements of the users. These three categories improve performance metrics by varying transmission rates and simulation time. For solving that problem, the proposed work is divided into Queue Management (QM), Admission Control (AC) and RA. For effective QM, this paper develops a QM model for elastic (EL) and inelastic (IEL) Traffic Flows. This research paper presents an AC mechanism for multiple TF for effective AC. This work presents a Resource Allocation Using Tokens (RAUT) for various priority TF for effective RA. Here, nodes have three cycles which are: Non-Critical Section (NCS), Entry Section (ES) and Critical Section (CS). When a node requires any resources, it sends Resource Request Message (RRM) to the ES. Elastic and inelastic TF priority is determined using Fuzzy Logic (FL). The token holder selects the node from the inelastic queue with high priority for allocating the resources. Using Network Simulator-2 (NS-2), simulations demonstrate that the proposed design increases Packet Delivery Ratio (PDR), decrease Packet Loss Ratio (PLR), minimise the Fairness and reduce the EED.

Recent progress in soft electronics and robotics based on magnetic nanomaterials

Recent advancements in soft electronics and robotics have expanded the possibilities beyond the capabilities of traditional rigid devices, indicating promise for a range of applications in electronic skins, wireless biomedical devices, and others. Magnetic materials exploited in these soft systems can further broaden the modalities in sensing and actuation. These magnetic materials, when constructed in the forms of nanoparticles, nanomembranes, or other types of nanostructures, exhibit some unique characteristics, such as the magnetoresistance effect and size-dependent coercivity. Soft electronics and robotics employing such magnetic nanomaterials offer a variety of functions, including the detection of the intensity and direction of magnetic fields, measurement of various types of mechanical deformations, manipulation and transport at small scales, and multimodal complex locomotion in a controllable fashion. Despite recent advancements in soft electronics and robotics, challenges remain in developing advanced materials and manufacturing schemes to improve performance metrics and facilitate integration with other devices. This review article aims to summarize the progress made in soft electronics and robotics based on magnetic nanomaterials, with an emphasis on introducing material and device performance. The discussions focus on soft electronics and robotics based on magnetic nanomembranes/nanostructures and magnetic composites. As a concluding remark, this article summarizes the current status of the field and discusses opportunities that underpin future progress.

IoT-Driven Healthcare Monitoring with Explainable Machine Learning Models

The need for this work arises from the critical importance of IoT-driven healthcare monitoring. In an era marked by technological advancements, the healthcare sector has not been left untouched. The ability to monitor patients' health remotely through IoT devices has emerged as a promising solution, offering real-time data for healthcare providers. However, amidst this promise, there remain significant challenges. Existing approaches in this domain have limitations. They often lack transparency and interpretability, making it challenging to trust the decisions made by machine learning models. Moreover, their performance metrics often fall short of achieving optimal precision, accuracy, recall, AUC, and speed, which are crucial in healthcare applications where timely and accurate decisions. In response to these challenges, this paper presents a novel approach. The proposed model leverages Convolutional Neural Networks (CNN) and integrates Deep Shap and GridCAM++ techniques to offer a more explainable and interpretable solution for IoT-driven healthcare monitoring. This fusion of methods enhances the model's transparency, allowing healthcare professionals to understand the rationale behind its decisions. The advantages of this approach are multifold. First and foremost, it enhances the precision, accuracy, recall, and AUC by 5.5%, 5.9%, 6.5%, and 8.3%, respectively, when compared to existing methods. These improvements translate to more reliable diagnoses and decisions in healthcare. Additionally, the model achieves a 4.9% boost in speed, ensuring that critical decisions can be made swiftly, reducing the time between data collection and actions. The impact of this work is substantial, it paves the way for more trustworthy and efficient IoT-driven healthcare monitoring systems, addressing the limitations of existing approaches. With improved performance metrics and enhanced explainability, healthcare professionals can make more informed decisions, leading to better patient outcomes. Ultimately, this paper contributes to the advancement of healthcare technology, bringing us closer to a future where IoT-enabled monitoring plays a pivotal role in improving patient care sets.

Modeling and analysis of field-oriented control based permanent magnet synchronous motor drive system using fuzzy logic controller with speed response improvement

&lt;span lang="EN-US"&gt;The permanent magnet synchronous motor (PMSM) acts as an electrical motor mainly used in many diverse applications. The controlling of the PMSM drive is necessary due to frequent usage in various systems. The conventional proportional-integral-derivative (PID) controller’s drawbacks are overcome with fuzzy logic controller (FLC) and adopted in the PMSM drive system. In this manuscript, an efficient field-oriented control (FOC) based PMSM drive system using a fuzzy logic controller (FLC) is modeled to improve the speed and torque response of the PMSM. The PMSM drive system is modeled using abc to &lt;/span&gt;&lt;em&gt;&lt;span lang="EN-US"&gt;αβ&lt;/span&gt;&lt;/em&gt;&lt;span lang="EN-US"&gt; and &lt;/span&gt;&lt;em&gt;&lt;span lang="EN-US"&gt;αβ&lt;/span&gt;&lt;/em&gt;&lt;span lang="EN-US"&gt; to abc transformation, 2-level space vector pulse width modulation (SVPWM), AC to DC rectifier with an inverter, followed by PMSM drive, &lt;a name="_Hlk107566912"&gt;&lt;/a&gt;proportional integral (PI) controller along with FLC. The FLC’s improved fuzzy rule set is adopted to provide faster speed response, less % overshoot time, and minimal steady-state error of the PMSM drive system. The simulation results of speed response, torque response, speed error, and phase currents are analyzed. The FLC-based PMSM drive is compared with the conventional PID-based PMSM drive system with better improvements in performance metrics.&lt;/span&gt;

Design and performance evaluation of successive interference cancellation based Slotted Aloha MAC protocol

As a result of densification, the performance of the wireless networks has become highly interference-limited and energy inefficient. To overcome this problem, interference mitigation techniques such as Successive Interference Cancellation (SIC) can be used to decode multiple packets simultaneously at the receiver. In this context, we analyze a SIC-based Slotted Aloha (SIC-SA) Medium Access Control (MAC) protocol for wireless networks. We derive expressions for packets decoding probability and optimal transmission probability of the nodes of the SIC-SA MAC protocol. Our derivation is based on the order statistics of Independent and Identical/non-Identical exponentially distributed received-signal-powers from the nodes under the Rayleigh channel condition. Throughput, delay, and energy efficiency of the SIC-SA MAC protocol have been derived and validated against simulation. The effect of path loss exponent, SINR threshold, and the number of nodes on the performance of SIC-SA have been studied. The performance of SIC-SA in a network of nodes distributed randomly according to the Poisson Point Process has been analyzed. Extension of our analysis to Power Domain Non-Orthogonal Multiple Access (NOMA) has been demonstrated. We also analyzed the impact of imperfect estimation of channel state information and imperfect SIC at the receiver. Results show an improvement in performance metrics of SIC-SA over the traditional Slotted Aloha.

Efficient Hybrid Iterative Method for Signal Detection in Massive MIMO Uplink System over AWGN Channel

Massive multiple input multiple output (massive MIMO) is a key technology in fifth-generation (5G) and beyond fifth-generation (B5G) networks. It improves performance metrics such as gain, energy efficiency, spectral efficiency, and bit error rate (BER). Because of the large number of users and antennas, sophisticated processing is required to detect the transmitted message signal. One of the challenges in massive MIMO systems is transmitted message signal detection. To respond to these challenges, several detection algorithms have been developed, including minimum mean squared error (MMSE), zero forcing (ZF), matched filter (MF), conjugate-gradient (CG), gauss-seidel (GS), and optimized coordinate descent (OCD). Although the ZF and MMSE algorithms perform well, their computational complexity is high due to direct matrix inversion. When the number of users is much lower than the number of antennas, the MF algorithm performs well. However, as the number of users increases, the performance of the MF algorithm degrades. Although the OCD, CG, and GS algorithms have less computational complexity than the MMSE algorithm, they perform poorly in comparison. To address and resolve the shortcomings of existing methods, an efficient iterative algorithm has been proposed in this manuscript, which is a hybrid method possessing the combination of MMSE with the alternating direction method of multipliers (ADMM) technique and Gauss-Seidel method. The initial vector has a large influence on the performance, complexity, and convergence rate of such iterative algorithms. The proposed detector’s initial solution is determined using the diagonal matrix and MMSE with the ADMM technique. The proposed algorithm’s performance and complexity are compared with existing algorithms based on BER and the real number of multiplications, respectively. The numerical results revealed that the proposed algorithm achieves the desired performance with a small number of iterations and a significant reduction in computational complexity. At 8QAM, SNR = 20 dB, 80 × 120 massive MIMO antenna configuration, and n = 2, the percentage performance improvement of the proposed detector from the GS detector is 99.82%. At 32QAM, SNR = 25 dB, 120 × 180 antenna configuration, and n = 5, performance improvement of the proposed detector is 99.89%. At 64QAM, SNR = 28 dB, 80 × 120 antenna configuration, and n = 3, performance improvement of the proposed detector is 99.93%.

Yttrium-based Double Perovskite Nanorods for Electrocatalysis.

Herein, we investigate the effect of the chemical composition of double perovskite nanorods on their versatile electrocatalytic activity not only as supports for the oxidation of small organic molecules but also as catalysts for the oxygen evolution reaction. Specifically, Y2CoMnO6 and Y2NiMnO6 nanorods with average diameters of 300 nm were prepared by a two-step hydrothermal method, in which the individual effects of synthetic parameters, such as the pH, annealing temperature, and precursor ratios on both the composition and morphology, were systematically investigated. When used as supports for Pt nanoparticles, Y2CoMnO6/Pt catalysts exhibited an electrocatalytic activity for the methanol oxidation reaction, which is 2.1 and 1.3 times higher than that measured for commercial Pt/C and Y2NiMnO6/Pt, respectively. Similarly, the Co-based catalyst support material displayed an ethanol oxidation activity, which is 2.3 times higher than both Pt/C and Y2NiMnO6/Pt. This clear enhancement in the activity for Y2CoMnO6 can largely be attributed to strong metal-support interactions, as evidenced by a downshift in the binding energy of the Pt 4f bands, measured by X-ray photoelectron spectroscopy (XPS), which is often correlated not only with a downshift in the d-band center but also to a decreased adsorption of poisoning adsorbates. Moreover, when used as catalysts for the oxygen evolution reaction, Y2CoMnO6 displayed a much greater activity as compared with Y2NiMnO6. This behavior can largely be attributed not only to a preponderance of comparatively more favorable oxidation states and electronic configurations but also to the formation of an active layer on the surface of the Y2CoMnO6 catalyst, which collectively gives rise to improved performance metrics and greater stability as compared with both IrO2 and Y2NiMnO6. Overall, these results highlight the importance of both the chemical composition and the electronic structure of double perovskites, especially when utilized in multifunctional roles as either supports or catalysts.

It's Time to Focus! Maximizing the Effectiveness and Outcomes of Your Infection Prevention Committee

Background One out of every 25 patients is impacted by a healthcare-acquired infection (HAI). Mitigation and prevention efforts require both a skilled infection preventionist (IP) and a high functioning infection prevention committee (IPC) by which the diffusion of information related to HAI risks is reviewed. In addition, there exist numerous regulatory requirements for infection prevention programs (IPP). This breadth of IPP elements led to deferred discussions of critical items, information overload, and limited time for detailed case reviews. Member attendance was sporadic therefore, situational awareness of practice gaps was missing. IPP action planning was ineffective due to this continual cycle of inefficiencies, thus putting patients and the organization at risk. Methods Within this healthcare system, the formation of IPP quality assurance, process improvement (QAPI) was implemented using standardized tools and a meeting consent agenda format. The IPC was educated on the consent agenda method and member responsibilities. Regulatory program requirements are all listed on a one-page checklist, providing clarity with all IPP components. These items and accompanying performance metrics are sent to committee members one week before the meeting. Most of the items are placed on the consent agenda which leaves time for focused discussions on critical items. Results Observed benefits across this network of hospitals include focused discussion on critical items and improved performance metrics. Other benefits include clarity for attendee expectations regarding meeting preparedness and readiness to discuss items of pertinence, meeting flexibility for the IPC by maneuvering consent agenda items based on need and shortened meeting time. Improved meeting attendance and meeting satisfaction was reported. Conclusions The IPP effectiveness and outcomes are based on an organized QAPI process and engaged committee members. Effective meeting facilitation and tools assist the IP in managing the broad scope of IPP elements ensuring patient harm is mitigated and regulatory requirements are met. One out of every 25 patients is impacted by a healthcare-acquired infection (HAI). Mitigation and prevention efforts require both a skilled infection preventionist (IP) and a high functioning infection prevention committee (IPC) by which the diffusion of information related to HAI risks is reviewed. In addition, there exist numerous regulatory requirements for infection prevention programs (IPP). This breadth of IPP elements led to deferred discussions of critical items, information overload, and limited time for detailed case reviews. Member attendance was sporadic therefore, situational awareness of practice gaps was missing. IPP action planning was ineffective due to this continual cycle of inefficiencies, thus putting patients and the organization at risk. Within this healthcare system, the formation of IPP quality assurance, process improvement (QAPI) was implemented using standardized tools and a meeting consent agenda format. The IPC was educated on the consent agenda method and member responsibilities. Regulatory program requirements are all listed on a one-page checklist, providing clarity with all IPP components. These items and accompanying performance metrics are sent to committee members one week before the meeting. Most of the items are placed on the consent agenda which leaves time for focused discussions on critical items. Observed benefits across this network of hospitals include focused discussion on critical items and improved performance metrics. Other benefits include clarity for attendee expectations regarding meeting preparedness and readiness to discuss items of pertinence, meeting flexibility for the IPC by maneuvering consent agenda items based on need and shortened meeting time. Improved meeting attendance and meeting satisfaction was reported. The IPP effectiveness and outcomes are based on an organized QAPI process and engaged committee members. Effective meeting facilitation and tools assist the IP in managing the broad scope of IPP elements ensuring patient harm is mitigated and regulatory requirements are met.

Evaluating the digestible lysine requirements of male Cobb MV × Cobb 500 broilers during the first fourteen days of age and the carryover effect of feeding varying levels of digestible lysine on performance and processing

Digestible lysine ( dLys ) requirements in broilers is heavily researched due to lysine's importance in diet formulation, body protein accretion, and dietary protein cost. As such, when novel broiler strains are established, it is important to evaluate their dLys requirements. The current objective was to determine the dLys requirement of male Cobb MV × Cobb 500 broilers from d 0 to 14 and evaluate the carryover effect (d 14–41) of feeding varying starter levels of dLys. Two basal diets were formulated (0.88% dLys and 1.44% dLys), then blended to create 6 intermediate dLys treatments (ranging from 0.96 to 1.36% dLys), creating a total of 8 dLys experimental diets. A control diet (treatment 9–1.28% dLys) was separately made to confirm blending techniques. Treatments were arranged as a randomized complete block design ( RCBD ) to 96 pens, with each pen having 14 male chicks. In general, increasing dLys resulted in improvements in performance metrics and processing weights. From d 0 to 14, feeding ≥1.20% dLys increased BW and BWG, feeding ≥0.96% dLys increased FI, with an exception in birds fed 1.44% dLys; stepwise reductions in FCR were observed as dLys increased from 0.88 until 1.20% dLys. Estimated dLys requirements ranged from 1.17 to 1.30% for BWG and 1.29 to 1.49% for FCR (using multiple regression methods). In general, carryover effect data (d 0–41) demonstrated that increasing starter dLys improved performance and d 42 processing. Future research should evaluate higher dLys levels than those used in this study and the dLys requirements of female Cobb MV × Cobb 500 broilers during the starter phase.

Optimizing hybrid metaheuristic algorithm with cluster head to improve performance metrics on the IoT

The Internet of Things (IoT) has subsequently been applied to a variety of sectors, including smart grids, farming, weather prediction, power generation, wastewater treatment, and so on. So if the Internet of Things has enormous promise in a wide range of applications, there still are certain areas where it may be improved. Designers had focused our present research on reducing the energy consumption of devices in IoT networks, which will result in a longer network lifetime. The far more suitable Cluster Head (CH) throughout the IoT system is determined in this study to optimize energy consumption. Whale Optimization Algorithm (WOA) with Evolutionary Algorithm (EA) is indeed a mixed meta-heuristic algorithm used during the suggested study. Various quantifiable metrics, including the variety of adult nodes, workload, temperatures, remaining energy, and a target value, were utilized IoT network groups. The suggested method then is contrasted to several cutting-edge optimization techniques, including the Artificial Bee Colony method, Neural Network, Adapted Gravity Simulated annealing. The findings show that the suggested hybrid method outperforms conventional methods.

Ensembled Transfer Learning and Multiple Kernel Learning for Phonocardiogram Based Atherosclerotic Coronary Artery Disease Detection.

Conventional machine learning has paved the way for a simple, affordable, non-invasive approach for Coronary artery disease (CAD) detection using phonocardiogram (PCG). It leaves a scope to explore improvement of performance metrics by fusion of learned representations from deep learning. In this study, we propose a novel, multiple kernel learning (MKL) for their fusion using deep embeddings transferred from pre-trained convolutional neural network (CNN). The proposed MKL, finds optimal kernel combination by maximizing the similarity with ideal kernel and minimizing the redundancy with other basis kernels. Experiments are performed on 960 PCG epochs collected from 40 CAD and 40 normal subjects. The transferred embeddings attain maximum subject-level accuracy of 89.25% with kappa of 0.7850. Later, their fusion with handcrafted features using the proposed MKL gives an accuracy of 91.19% and kappa 0.8238. The study shows the potential of development of high accuracy CAD detection system by using easy to acquire, non-invasive PCG signal.

Enhancing Time Synchronization for Home Automation Systems

Clock synchronization techniques are in highly demand for smart home applications with the approach of the Internet of Things (IoT). The developing innovation of smart home applications built on the Internet of Things is generally constrained and scattered. Synchronization is the basic characteristic of the emerging field of IoT in smart home applications which is recommended for improving the acceptance and its use. For the clock synchronization between systems, Network Time Protocol (NTP) is used. But due to inconsistent routing and limited computing resources there can be the dominant source of error that limits time transfer uncertainty. This paper proposes an Elastic Timer Protocol for home automation system networks that performs better in terms of consumption of energy and time of convergence than other clock synchronization protocols. The proposed protocol improves the elasticity and efficiency of the network, resulting in a faster network convergence time. Furthermore, simulation results revealed that the proposed work has improved performance metrics and proved superior performance characteristics when compared to standard synchronization protocols.

A Study on Performance Metrics for Anomaly Detection Based on Industrial Control System Operation Data

Recently, OT (operational technology) networks of industrial control systems have been combined with IT networks. Therefore, OT networks have inherited the vulnerabilities and attack paths existing in IT networks. Consequently, attacks on industrial control systems are increasing, and research on technologies combined with artificial intelligence for detecting attacks is active. Current research focuses on detecting attacks and improving the detection accuracy. Few studies exist on metrics that interpret anomaly detection results. Different analysis metrics are required depending on the characteristics of the industrial control system data used for anomaly detection and the type of attack they contain. We focused on the fact that industrial control system data are time series data. The accuracy and F1-score are used as metrics for interpreting anomaly detection results. However, these metrics are not suitable for evaluating anomaly detection in time series data. Because it is not possible to accurately determine the start and end of an attack, range-based performance metrics must be used. Therefore, in this study, when evaluating anomaly detection performed on time series data, we propose a range-based performance metric with an improved algorithm. The previously studied range-based performance metric time-series aware precision and recall (TaPR) evaluated all attacks equally. In this study, improved performance metrics were studied by deriving ambiguous instances according to the characteristics of each attack and redefining the algorithm of the TaPR metric. This study provides accurate assessments when performing anomaly detection on time series data and allows predictions to be evaluated based on the characteristics of the attack.