Standard Algorithm Research Articles

Navigating complex care pathways-healthcare workers' perspectives on health system barriers for children with tuberculous meningitis in Cape Town, South Africa.

Tuberculous meningitis (TBM) occurs when tuberculosis (TB) bacilli disseminate and seed into the meninges, triggering a severe inflammatory response that often leads to brain infarction. It is the most severe and debilitating form of childhood TB with high mortality, and children who survive TBM often suffer lifelong physical and neuro-disability resulting in emotional, social, and economic burdens for families. In the early stages the symptoms may be non-specific and so the diagnosis is often made late when the patient already has significant brain injury. To facilitate earlier diagnosis, it is important to understand how patients are evaluated. This study aimed to chart health systems for paediatric TBM care at both primary healthcare (PHC) and hospital level in Cape Town, South Africa. We conducted fourteen in-depth interviews and eight days of semi-structured observations of patient flow across eight healthcare facilities. We found that children with TBM navigate multiple levels of care categorised into pre-admission and primary care, hospital admission and inpatient care, and post-discharge follow-up care. Healthcare workers identified the following health system barriers along the TBM care pathway for children: limited post-training and mentorship opportunities to manage TBM, overburdened facilities, time constraints, lack of recognition of TBM symptoms, delays in referral between PHC and hospital, lack of standardized diagnostic algorithms, limited diagnostic tests and a lack of child-friendly, easy-to-administer treatment. Regular and compulsory training on TB and TBM in children, including continuous mentoring and support to healthcare workers working in child health and TB services in high TB-burden settings, can facilitate early recognition of symptoms and rapid referral for diagnosis. Algorithms outlining referral criteria for patients with possible TBM at both PHC facilities and district level hospitals can guide healthcare providers and facilitate timely referral between different levels of healthcare services. An integrated data system and alert functions could flag multiple healthcare visits and improve communication between different healthcare facilities during diagnosis and treatment. Children and families affected by TBM are an especially vulnerable sub-population requiring high priority attention and support.

Improving derivative-free optimization algorithms through an adaptive sampling procedure

Black-box optimization plays a pivotal role in addressing complex real-world problems where the underlying mathematical model is unknown or expensive to evaluate. In this context, this work presents a method to enhance the performance of derivative-free optimization algorithms by integrating an adaptive sampling process. The proposed methodology aims to overcome the limitations of traditional methods by intelligently guiding the search towards promising regions of the search space. To achieve this, we utilize machine learning models, which effectively substitute first principles models. Furthermore, we employ the error maximization approach to steer the exploration towards areas where the surrogate model deviates significantly from the true model. Moreover, we involve a heuristic method, an adaptive sampling procedure, that repeats calls to a widely-used derivative-free optimization algorithm, SNOBFIT, allowing for the creation of new and improved surrogate models. To evaluate the efficiency of the proposed method, we conduct a comparative analysis across a benchmark set of 776 continuous problems. Our findings indicate that our approach successfully solved 93% of the problems. Notably, for larger problems, our method outperformed the standard SNOBFIT algorithm by achieving a 19% increase in problem-solving rate, and when, we introduced an additional termination criterion to enhance computational efficiency, the proposed method achieved a 31% time reduction compared to SNOBFIT.

Detection of Subtle ECG Changes Despite Superimposed Artifacts by Different Machine Learning Algorithms

Analyzing electrocardiographic (ECG) signals is crucial for evaluating heart function and diagnosing cardiac pathology. Traditional methods for detecting ECG changes often rely on offline analysis or subjective visual inspection, which may overlook subtle variations, particularly in the case of artifacts. In this theoretical, proof-of-concept study, we investigated the potential of five different machine learning algorithms [random forests (RFs), gradient boosting methods (GBMs), deep neural networks (DNNs), an ensemble learning technique, as well as logistic regression] to detect subtle changes in the morphology of synthetically generated ECG beats despite artifacts. Following the generation of a synthetic ECG beat using the standardized McSharry algorithm, the baseline ECG signal was modified by changing the amplitude of different ECG components by 0.01–0.06 mV. In addition, a Gaussian jitter of 0.1–0.3 mV was overlaid to simulate artifacts. Five different machine learning algorithms were then applied to detect differences between the modified ECG beats. The highest discriminatory potency, as assessed by the discriminatory accuracy, was achieved by RFs and GBMs (accuracy of up to 1.0), whereas the least accurate results were obtained by logistic regression (accuracy approximately 10% less). In a second step, a feature importance algorithm (Boruta) was used to discriminate which signal parts were responsible for difference detection. For all comparisons, only signal components that had been modified in advance were used for discretion, demonstrating that the RF model focused on the appropriate signal elements. Our findings highlight the potential of RFs and GBMs as valuable tools for detecting subtle ECG changes despite artifacts, with implications for enhancing clinical diagnosis and monitoring. Further studies are needed to validate our findings with clinical data.

Inferring fault structures and overburden depth in 3D from geophysical data using machine learning algorithms – A case study on the Fenelon gold deposit, Quebec, Canada

AbstractWe apply a machine learning approach to automatically infer two key attributes – the location of fault or shear zone structures and the thickness of the overburden – in an 18 km2 study area within and surrounding the Archean Fenelon gold deposit in Quebec, Canada. Our approach involves the inversion of carefully curated borehole lithological and structural observations truncated at 480 m below the surface, combined with magnetic and Light Detection and Ranging survey data. We take a computationally low‐cost approach in which no underlying model for geological consistency is imposed. We investigated three contrasting approaches: (1) an inferred fault model, in which the borehole observations represent a direct evaluation of the presence of fault or shear zones; (2) an inferred overburden model, using borehole observations on the overburden‐bedrock contact; (3) a model with three classes – overburden, faulted bedrock and unfaulted bedrock, which combines aspects of (1) and (2). In every case, we applied all 32 standard machine learning algorithms. We found that Bagged Trees, fine K‐nearest neighbours and weighted K‐nearest neighbour were the most successful, producing similar accuracy, sensitivity and specificity metrics. The Bagged Trees algorithm predicted fault locations with approximately 80% accuracy, 70% sensitivity and 73% specificity. Overburden thickness was predicted with 99% accuracy, 77% sensitivity and 93% specificity. Qualitatively, fault location predictions compared well to independently construct geological interpretations. Similar methods might be applicable in other areas with good borehole coverage, providing that criteria used in borehole logging are closely followed in devising classifications for the machine learning training set and might be usefully supplemented with a variety of geophysical survey data types.

AN EFFICIENT MACHINE LEARNING-BASED DETECTION AND PREDICTION MECHANISM FOR CYBER THREATS USING INTELLIGENT FRAMEWORK IN IOTS

The dangers that Internet of Things (IoT) devices pose to large corporate corporations and smart districts have been dissected by several academics. Given the ubiquitous use of IoT and its unique characteristics, such as mobility and normalization restrictions, intelligent frameworks that can independently detect suspicious activity in privately linked IoT devices are crucial. The IoTs have led an explosion in traffic through the network, bringing information processing techniques for attack detection. The increase in traffic poses challenges in detecting attacks and differentiating traffic that is harmful. In this work, we have proposed a mechanism that uses the standard algorithms in a system that is designed to detect, track, measure and identify online traffic from organizations with malignant transmission: Random Forest (RF), gradient-boosted decision trees (GBDT), and support vector machines (SVM) gives an optimal accuracy of 80.34%,87.5%, and 88.6% while the random forest-based supervised approach is 5.5% better than the previous techniques. To facilitate comparisons between training time, prediction time, specificity, and accuracy, the proposed approach leverages the NSL KDD dataset accuracy.

AN EFFICIENT MACHINE LEARNING-BASED DETECTION AND PREDICTION MECHANISM FOR CYBER THREATS USING INTELLIGENT FRAMEWORK IN IOTS

The dangers that Internet of Things (IoT) devices pose to large corporate corporations and smart districts have been dissected by several academics. Given the ubiquitous use of IoT and its unique characteristics, such as mobility and normalization restrictions, intelligent frameworks that can independently detect suspicious activity in privately linked IoT devices are crucial. The IoTs have led an explosion in traffic through the network, bringing information processing techniques for attack detection. The increase in traffic poses challenges in detecting attacks and differentiating traffic that is harmful. In this work, we have proposed a mechanism that uses the standard algorithms in a system that is designed to detect, track, measure and identify online traffic from organizations with malignant transmission: Random Forest (RF), gradient-boosted decision trees (GBDT), and support vector machines (SVM) gives an optimal accuracy of 80.34%,87.5%, and 88.6% while the random forest-based supervised approach is 5.5% better than the previous techniques. To facilitate comparisons between training time, prediction time, specificity, and accuracy, the proposed approach leverages the NSL KDD dataset accuracy.

Automated phenotyping empowered by deep learning for genomic prediction of body size in the tiger pufferfish, Takifugu rubripes

Although genomic selection (GS) is accelerating the genetic improvement of economically important traits in aquaculture species, the accurate phenotyping of thousands of individuals remains a bottleneck in GS programs. Computer vision (CV) technologies with the assistance of deep learning may provide a cost-efficient solution for automated phenotyping of fish body size. In this study, we designed a CV-based automated phenotyping system based on the deep learning model Mask R-CNN (region-based convolutional neural networks) and tested its ability to predict standard length (SL) of tiger pufferfish (Takifugu rubripes) fed low fishmeal (LFM) diet. Simultaneously, we investigated the impact of the automated phenotyping on genetic parameter estimation and genomic prediction for growth-related traits. We raised a test population (n = 1001) that was fed the LFM diet from three months old until harvest at 22 months. Manual phenotyping of the fish was performed at 10-, 14-, 17-, and 22-months-old. At the last two measurement times, images of each fish were collected to train Mask R-CNN for fish detection and pixel-level segmentation of the fish body excluding the tail fin. An automated phenotyping pipeline combining the trained Mask R-CNN and standard CV algorithms were designed to estimate SL. We found a high correlation (0.965–0.971) and low relative difference (0.015–0.027) between the CV-derived and manually measured SL, indicating high precision of our CV system. A high correlation (0.912–0.956) was also observed between CV-derived body area and manual body weight. The heritability estimations and genomic predictions on manual SLs were conducted using a Genomic Best Linear Unbiased Prediction (GBLUP) model using a total of 16,471 single nucleotide polymorphism (SNP). Estimated heritability at the four sampling points ranged from 0.573 to 0.622, and the predictive ability of genomic prediction for harvest size was 0.627. These values are comparable to those reported in previous studies collected from the population fed a normal fishmeal diet. Our heritability estimation and genomic prediction derived from automated SL measurements differed negligibly to those based on manual measurements. This finding indicates the utility of our CV-based automated phenotyping system for GS breeding programs.

Decoherence ensures convergence of non-adiabatic molecular dynamics with number of states.

Non-adiabatic (NA) molecular dynamics (MD) is a powerful approach for studying far-from-equilibrium quantum dynamics in photophysical and photochemical systems. Most NA-MD methods are developed and tested with few-state models, and their validity with complex systems involving many states is not well studied. By modeling intraband equilibration and interband recombination of charge carriers in MoS2, we investigate the convergence of three popular NA-MD algorithms, fewest switches surface hopping (FSSH), global flux surface hopping (GFSH), and decoherence induced surface hopping (DISH) with the number of states. Only the standard DISH algorithm converges with the number of states and produces Boltzmann equilibrium. Unitary propagation of the wave function in FSSH and GFSH violates the Boltzmann distribution, leads to internal inconsistency between time-dependent Schrödinger equation state populations and trajectory counts, and produces non-convergent results. Introducing decoherence in FSSH and GFSH by collapsing the wave function fixes these problems. The simplified version of DISH that omits projecting out the occupied state and is applicable to few-state systems also causes problems when the number of states is increased. We discuss the algorithmic application of wave function collapse and Boltzmann detailed balance and provide detailed FSSH, GFSH, and DISH flow charts. The use of convergent NA-MD methods is highly important for modeling complicated quantum processes involving multiple states. Our findings provide the basis for investigating quantum dynamics in realistic complex systems.

A method to standardize the temperature for near infrared spectra of the indigo pigment in non-dairy cream based on symbolic regression

Near infrared (NIR) spectroscopy is sensitive to physical conditions such as sample temperature, meaning that rapid detection methods based on NIR spectroscopy are significantly influenced by temperature. To address this challenge, symbolic regression was employed to mitigate the effects of temperature. The Weighted Windowed Adaptive Optimization algorithm was proposed and combined with the Sequential Projection Algorithm to extract temperature-related feature points and remove redundant data. Subsequent 3D modeling of these feature points revealed that absorbance alterations due to temperature comprised two distinct segments. Consequently, based on symbolic regression, the temperature standardization algorithm was devised to generate piecewise equations. This algorithm surpassed genetic programming and non-segmented methods in performance metrics. The piecewise function equations generated by the algorithm were used to regress the absorbance at different temperatures to the standard temperature. Non-dairy cream, with different indigo pigment contents, was temperature standardized using a piecewise function to obtain spectra at two standard temperatures; 18°C and 28°C. The r2 for the quantitative regression model improved from 0.71 to 0.95 at 18°C and from 0.63 to 0.85 at 28°C. The temperature standardization method offers interpretable equations for spectra that model the complex changes with temperature, factoring out the temperature variation, thereby facilitating the practical use of NIR spectroscopy in rapid detection applications.

Improving quality of service for Internet of Things(IoT) in real life application: A novel adaptation based Hybrid Evolutionary Algorithm

In this paper, we address critical challenges in IoT sensor lifespan, service latency, and coverage area, all impacting energy consumption in smart agriculture applications. To enhance the quality of service (QoS) while prolonging the energy efficiency of smart sensors, a novel optimization algorithm is introduced. Referred to as the ”Adaptation-Based Hybrid Evolutionary Algorithm,” this innovative approach combines the strengths of Grey Wolf Optimizers (GWO) and Differential Evolution (DE) algorithms. The methodology involves a new adaptation-based strategy and incorporates a hybrid algorithm that synergizes the exploratory and exploitative capabilities of both GWO and DE algorithms. This hybrid approach is leveraged to meticulously select optimal mutation new adaptation services, drawing from the GWO and DE algorithm frameworks. Notably, the algorithm’s control parameters autonomously adjust through insights gained from prior evolutionary searches. Furthermore, we enhance the DE-based crossover technique by integrating the proficient search capabilities of the GWO algorithm, renowned for tackling continuous global optimization problems. To validate our approach, we apply it to IoT scenarios and optimize QoS through a fitness function that comprehensively accounts for energy consumption, coverage rate, lifespan, and latency. Comparative evaluations against standard algorithms underscore the superior performance of our proposed methodology, particularly evident in its application to IoT-smart agriculture settings.

Treatment concepts for the medial clavicle and the sternoclavicular joint

Medial clavicle fractures and injuries to the sternoclavicular joint are rare injuries but can have life-threatening consequences. There are no standardized treatment algorithms or guidelines for the diagnostics and treatment. This article provides an overview of the individual topographies as well as the conservative and surgical treatment strategies.Conservative treatment is preferred for medial clavicle fractures. The indications for surgical treatment are variable but this is frequently carried out if there is a fracture displacement > 1 cm or1 shaft width and high functional demands. In the case of accompanying injuries to neurovascular structures, an open fracture or the threat of perforation of the skin, surgical treatment is mandatory. Open reduction and internal fixation using (locking) plates is currently the preferred form of treatment.In the case of posterior dislocation of the sternoclavicular joint, an immediate closed reduction must be attempted with the patient under analgosedation and with emergency treatment on standby. This temporal urgency does not exist for anterior and superior dislocations. Surgical treatment is indicated in cases of unsuccessful reduction, persistent symptomatic instability or injuries of the neurovascular bundle. From a multitude of treatment options, arthrodesis with suture cerclage has shown good results. Tendon grafts as well as special hook plates are increasingly being used due to better biomechanical qualities. The surgical treatment of combined injuries is determined by the individual injury pattern.Despite the variety of treatment strategies, the long-term outcome has consistently been positively described.

Affect regulation and allostatic load over time

ObjectiveEmerging work suggests that affect regulation strategies (e.g., active coping, anger expression) predict disease and mortality risk, with sometimes divergent estimates by sex or education levels. However, few studies have examined potential underlying biological mechanisms. This study assessed the longitudinal association of affect regulation with future allostatic load. MethodIn 2004–2006, 574 participants from the Midlife in the United States study completed validated scales assessing use of nine general and emotion-specific regulatory strategies (e.g., denial, anger expression). As a proxy for how flexibly participants regulate their affect, variability in the use of regulatory strategies was operationalized using a standard deviation-based algorithm and considered categorically (i.e., lower, moderate, greater variability) to assess non-linear effects. Participants also provided data on relevant covariates and 24 allostatic load biomarkers (e.g., cortisol, blood pressure). In 2017–2021, these biomarkers were again collected. Linear regressions modeled betas (β) and 95 % confidence intervals (CI) examining associations of affect regulatory constructs with future allostatic load. ResultsIn fully-adjusted models including initial allostatic load, general regulatory strategies were unrelated to future allostatic load. Yet, greater versus moderate affect regulation variability levels predicted lower allostatic load (β=−0.14; 95 %CI: −0.27, −0.01). Only among more educated participants, greater use of anger expression predicted lower allostatic load, while the reverse was noted with anger control (βexpression=−0.12; 95 %CI: −0.20, −0.05; βcontrol=0.14; 95 %CI: 0.05, 0.24). ConclusionsWhile general regulatory strategies appeared unrelated to allostatic load, greater variability in their use and anger-related strategies showed predictive value. Subsequent studies should examine these associations in larger, more diverse samples.

Internet-of-Things for smart irrigation control and crop recommendation using interactive guide-deep model in Agriculture 4.0 applications

ABSTRACT The rapid advancements in Agriculture 4.0 have led to the development of the continuous monitoring of the soil parameters and recommend crops based on soil fertility to improve crop yield. Accordingly, the soil parameters, such as pH, nitrogen, phosphorous, potassium, and soil moisture are exploited for irrigation control, followed by the crop recommendation of the agricultural field. The smart irrigation control is performed utilizing the Interactive guide optimizer-Deep Convolutional Neural Network (Interactive guide optimizer-DCNN), which supports the decision-making regarding the soil nutrients. Specifically, the Interactive guide optimizer-DCNN classifier is designed to replace the standard ADAM algorithm through the modeled interactive guide optimizer, which exhibits alertness and guiding characters from the nature-inspired dog and cat population. In addition, the data is down-sampled to reduce redundancy and preserve important information to improve computing performance. The designed model attains an accuracy of 93.11 % in predicting the minerals, pH value, and soil moisture thereby, exhibiting a higher recommendation accuracy of 97.12% when the model training is fixed at 90%. Further, the developed model attained the F-score, specificity, sensitivity, and accuracy values of 90.30%, 92.12%, 89.56%, and 86.36% with k-fold 10 in predicting the minerals that revealed the efficacy of the model.

Task-Driven Reinforcement Learning With Action Primitives for Long-Horizon Manipulation Skills.

It is an interesting open problem to enable robots to efficiently and effectively learn long-horizon manipulation skills. Motivated to augment robot learning via more effective exploration, this work develops task-driven reinforcement learning with action primitives (TRAPs), a new manipulation skill learning framework that augments standard reinforcement learning algorithms with formal methods and parameterized action space (PAS). In particular, TRAPs uses linear temporal logic (LTL) to specify complex manipulation skills. LTL progression, a semantics-preserving rewriting operation, is then used to decompose the training task at an abstract level, informs the robot about their current task progress, and guides them via reward functions. The PAS, a predefined library of heterogeneous action primitives, further improves the efficiency of robot exploration. We highlight that TRAPs augments the learning of manipulation skills in both learning efficiency and effectiveness (i.e., task constraints). Extensive empirical studies demonstrate that TRAPs outperforms most existing methods.

Secure transmission of IRS-UAV buffer-aided relaying system with delay constraint

This paper investigates the secure communication between legitimate users in the presence of eavesdroppers, where the Intelligent Reflective Surface-Unmanned Aerial Vehicle (IRS-UAV) and Buffer-Aided (BA) relaying techniques are utilized to enhance secrecy performance. By jointly optimizing the link selection strategy, the UAV position, and the reflection coefficient of the IRS, we aim to maximize the long-term average secrecy rate. Specifically, we propose a novel buffer in/out stabilization scheme based on the Lyapunov framework, which transforms the long-term average secrecy rate maximization problem into two per-slot drift-plus-penalty minimization problems with different link selection factors. The hybrid Particle Swarm Optimization-Artificial Fish Swarm Algorithm (PSO-AFSA) is adopted to optimize the UAV position, and the IRS reflection coefficient optimization problem is solved by iterative optimization in which auxiliary variables and standard convex optimization algorithms are introduced. Finally, the delay constraint is set to ensure the timeliness of information packets. Simulation results demonstrate that our proposed scheme outperforms the comparison schemes in terms of average secrecy rate. Specifically, the addition of BA improves the average secrecy rate by 1.37 bps/Hz, and the continued optimizations of IRS reflection coefficients and UAV positions improve the average secrecy rate by 2.46 bps/Hz and 3.75 bps/Hz, respectively.

Photovoltaic MPPT based on improved Cuckoo algorithm

Abstract Photovoltaic power generation is susceptible to external factors such as light, resulting in a reduction in the actual efficiency of photovoltaic power generation. However, the traditional maximum power point tracking technology has the characteristics of slow convergence speed and poor accuracy. In order to improve the above problems, a new photovoltaic maximum power control algorithm with an improved cuckoo algorithm is proposed. By adjusting the iterative step size in the cuckoo algorithm and changing the probability of finding the bird nest, the iterative convergence speed is accelerated, and the global optimization ability is increased. Finally, the algorithm is applied to the maximum power tracking, and the simulation model is built on the Matlab/Simulink platform. After comparing the standard Cuckoo algorithm with the improved Cuckoo algorithm under static and dynamic conditions, the simulation results show that the improved Cuckoo algorithm has more advantages in maximum power tracking speed and accuracy.

Fuzzy PID control based on genetic algorithm optimization inverted pendulum system

Abstract For the first-order inverted pendulum control system, a fuzzy PID control system based on the optimization of the genetic algorithm is proposed. The traditional genetic algorithm has the problem that the difference in the fuzzy subset parameter leads to a decrease in the interpretative ability of the fuzzy system. The main problem of the current genetic algorithm is the complexity of the computation and the low efficiency. Based on this problem, this paper proposes an improved genetic algorithm, i.e., it adopts the variance operator and adaptive change of the variance index and elite retention strategy, which solves the premature and local convergence problems of the standard genetic algorithm, in order to optimize the fuzzy system. The experimental results show that the optimized genetic algorithm gives full play to the advantages of fuzzy control in terms of interpretability and robustness, and at the same time guarantees the prediction accuracy, which provides a new research idea in the field of artificial intelligence control.

Application of optimizing advanced encryption standard encryption algorithm in secure communication of vehicle controller area network bus

Introduction: As the main means of information exchange within vehicles, the safety of the controller area network bus directly affects the safe operation of the vehicle and the safety of passengers’ lives and property.Methods: To enhance its secure communication function, this study utilizes advanced encryption standard algorithms and improves the S-box of the algorithm to solve problems such as extended processing time. A secure communication system for the local area network bus of the vehicle controller is designed based on optimized advanced encryption standard algorithms.Results and Discussion: The results showed that when the file size was 200MB, the encryption and decryption time spent by the research method was 469.8 s and 528.5 s, respectively, which are significantly lower than traditional methods. In the simulation results, under both non-encrypted and encrypted transmission, the information remained intact throughout the entire transmission process. This indicated that the optimization algorithm effectively reduced encryption processing time and system resource consumption while ensuring data confidentiality and integrity. The new system meets the security requirements of the local area network bus of vehicle-mounted controllers.Conclusion: This study not only enhances the security of in-vehicle networks but also promotes the application and development of related encryption technologies in the field of vehicle networking. It provides strong technical support for the further development of vehicle networking and the safe operation of intelligent vehicles.

Entropy‐based hybrid sampling (EHS) method to handle class overlap in highly imbalanced dataset

AbstractClass imbalance and class overlap create difficulties in the training phase of the standard machine learning algorithm. Its performance is not well in minority classes, especially when there is a high class imbalance and significant class overlap. Recently it has been observed by researchers that, the joint effects of class overlap and imbalance are more harmful as compared to their direct impact. To handle these problems, many methods have been proposed by researchers in past years that can be broadly categorized as data‐level, algorithm‐level, ensemble learning, and hybrid methods. Existing data‐level methods often suffer from problems like information loss and overfitting. To overcome these problems, we introduce a novel entropy‐based hybrid sampling (EHS) method to handle class overlap in highly imbalanced datasets. The EHS eliminates less informative majority instances from the overlap region during the undersampling phase and regenerates high informative synthetic minority instances in the oversampling phase near the borderline. The proposed EHS achieved significant improvement in F1‐score, G‐mean, and AUC performance metrics value by DT, NB, and SVM classifiers as compared to well‐established state‐of‐the‐art methods. Classifiers performances are tested on 28 datasets with extreme ranges in imbalance and overlap.

Localized density-based K-means UAV communication coverage algorithm for UV collaboration

In response to the problem of UAV-assisted communication, utilizing the characteristics of wireless ultraviolet with strong anti-jamming performance and good confidentiality, providing emergency communication services by covering ground users with UAVs as aerial base stations is crucial. A K-means clustering coverage algorithm based on the local density is proposed, which increases the connectivity term of UAVs on the basis of minimizing the squared error; secondly, the definition formula of the local density is proposed, and a heuristic decision function is used to determine the selection of the initial cluster center. The simulation results show that the initial coverage of the initial cluster center selected by the heuristic function is increased by 31.6%. Compared to the standard K-means algorithm and FSFDP algorithm, the coverage rate of the proposed algorithm in this paper is improved by 7.2% and 4.3%, respectively.