Application Of Support Vector Machine Research Articles

Analysing the causes of tourists’ emotional experience related to tourist attractions from a binary emotions perspective utilising machine learning models

ABSTRACT This study constructs a theoretical framework to analyse the causes of tourists' binary emotional experiences. It applies Support Vector Machine (SVM) and Latent Dirichlet allocation (LDA) machine learning models, combined with geospatial analysis methods, to online reviews of five types of tourist attractions in Dali, China. The results indicate that positive sentiments predominated across Dali Prefecture, though some attractions in Dali City received negative ratings. Furthermore, service experience and price were common influences on tourists' sentiments. This study reveals the causes of tourists' varied emotional experiences at tourist attractions from a binary emotional perspective.

Prediction of Air Pressure using Support Vector Machine Compare with Multi Layer Perception Model in an Area of North Twenty-four Parganas, West Bengal

Lifesaving requires the capacity to foresee the weather. Applications for weather prediction in agriculture, transportation, floods, and other natural phenomena are significant. The prediction of Air Pressure is crucial for weather forecasting. Multiple linear regression is one of the techniques available for Air Pressure prediction. model. To determine how one independent variable and one dependent variable are related to one another In statistics, the linear regression model is an extremely helpful tool. Multiple linear regression is an extension of the linear regression model. We utilized MATLAB to add the variables to our experiment. Weather forecasting relies heavily on Air Pressure prediction. For the purpose of forecasting rainfall, we have chosen Kolkata, or more specifically, the Dum Dum Metropolitan region in North 24 Parganas. We have investigated additional atmospheric forecasting parameters. applying a multilayer perceptron to predict Air Pressure. Once we get the coefficient, we have the intended and calculated Air Pressure. We have identified the error. Now we are applying Support Vector Machine over Multi Linear Preceptron to get better results.

Predicting land cover driven ecosystem service value using artificial neural network model

Understanding the synergies and trade-offs of major cities' ecosystem services is vital to mitigating regional ecological and environmental risks and enhancing human well-being in this era of rapid urbanization and global climate change. This study aimed to assess and predict the land use- and land cover (LULC)-driven ecosystem service value (ESV) dynamics in Arkansas’s capital city, Little Rock. Historical LULC data were derived by applying support vector machine learning algorithms to Landsat satellite imagery. The benefit transfer method was utilized to identify nine types of ecosystem services and their corresponding economic values. A cellular automata artificial neural network model was used to simulate future potential LULC and ESV patterns. Vegetation accounted for more than 94% of total ESV over the past two decades. However, a 38.40% expansion of built-up areas resulted in a 45.28% decrease in vegetated areas, which reduced total ESV from $3619.73 × 106 to $2563.81 × 106 during 2003–2023. By 2033, the city’s urban area will expand to 72.75% of the total area and will witness further declines of 30.35 km2 in vegetation, 19.30 km2 in barren soil, and 1.69 km2 in waterbody areas. Consequently, the ESVs of these natural landscapes will decline by $708.58 × 106, $44.87 × 106, and $15.69 × 106, respectively. Provisioning services will be most affected, followed by supporting, regulating, and cultural services. The study findings provide reference information to policymakers and the local government for use in adopting sustainable land management policies, thereby promoting the ecological value of Little Rock.

Impermeability Evaluation Of Concrete With Fly Ash Aggregate And Prediction With Modelling

Concrete is the most synthesized material in construction sector which it has aggregate as one of its components. The use of natural aggregate in concrete preparation uses a significant amount of non-renewable resources and energy, having a significant environmental impact. Multiple research projects have been conducted to safeguard natural reserves, seeking a solution to the waste disposal issue, and reduce construction costs by utilizing waste materials. FA(Fly Ash) aggregate is one such material that can be a replacement for natural aggregate. Durability parameters of concrete with Fly Ash (FA) aggregate are studied in this work as an alternative for fine aggregate. In this study, 5 concrete mixes were prepared utilizing FA aggregate in percentage substitution of 0%, 10%, 20%, 30%, and 40% for each. The quantity of cement, compaction, curing rate, concrete cover, and porosity all influence the durability of the concrete. Concrete attributes such as strength in compression, retaliation to abrasion and half-cell potential were investigated. Durability parameters of the specimens were tested after 90-day curing. The results revealed that concrete with 30% FA aggregate had the highest compressive strength, improved resistance towards abrasion and least half cell potential values. Experimentation data were used to develop comprehensive prediction models by applying support vector machine (SVM) algorithm. The SVM model analyses R2 values with an accuracy of over 97%. As a result, we can use SVM to efficiently execute prediction modelling in construction area.

Dingo algorithm-based forwarder selection and huffman coding to improve authentication

<span>In wireless sensor network (WSN), the high volume of observe and transmitted data among sensor nodes make it requires to maintain the security. Even though numerous secure data transmission approaches designed over a network, an inadequate resource and the complex environment cause not able to used in WSNs. Moreover, secure data communication is a big challenging problem in WSNs especially for the military application. This paper proposes a dingo algorithm-based forwarder selection and huffman coding (DAHC) to improve authentication in internet of things (IoT) WSN. Initially, it detects the anomalous nodes by applying support vector machine (SVM) algorithm based on sensor node energy, node selfishness, and signal to noise ratio (SNR). Next, we using the dingo algorithm to select the forwarder node. This dingo algorithm computes the fitness function based on node degreee, node distance and node energy. Finally, the huffman coding to provide end to end authentication established on node energy from sender to receiver. During data transmission, the huffman coding to build the binary hop count value, it improves the authentication in the WSN. Performance results specify that this approach enhances the detection ratio and throughput.</span>

Machine Learning Algorithms Applied to Predict Autism Spectrum Disorder Based on Gut Microbiome Composition.

The application of machine learning (ML) techniques stands as a reliable method for aiding in the diagnosis of complex diseases. Recent studies have related the composition of the gut microbiota to the presence of autism spectrum disorder (ASD), but until now, the results have been mostly contradictory. This work proposes using machine learning to study the gut microbiome composition and its role in the early diagnosis of ASD. We applied support vector machines (SVMs), artificial neural networks (ANNs), and random forest (RF) algorithms to classify subjects as neurotypical (NT) or having ASD, using published data on gut microbiome composition. Naive Bayes, k-nearest neighbors, ensemble learning, logistic regression, linear regression, and decision trees were also trained and validated; however, the ones presented showed the best performance and interpretability. All the ML methods were developed using the SAS Viya software platform. The microbiome's composition was determined using 16S rRNA sequencing technology. The application of ML yielded a classification accuracy as high as 90%, with a sensitivity of 96.97% and specificity reaching 85.29%. In the case of the ANN model, no errors occurred when classifying NT subjects from the first dataset, indicating a significant classification outcome compared to traditional tests and data-based approaches. This approach was repeated with two datasets, one from the USA and the other from China, resulting in similar findings. The main predictors in the obtained models differ between the analyzed datasets. The most important predictors identified from the analyzed datasets are Bacteroides, Lachnospira, Anaerobutyricum, and Ruminococcus torques. Notably, among the predictors in each model, there is the presence of bacteria that are usually considered insignificant in the microbiome's composition due to their low relative abundance. This outcome reinforces the conventional understanding of the microbiome's influence on ASD development, where an imbalance in the composition of the microbiota can lead to disrupted host-microbiota homeostasis. Considering that several previous studies focused on the most abundant genera and neglected smaller (and frequently not statistically significant) microbial communities, the impact of such communities has been poorly analyzed. The ML-based models suggest that more research should focus on these less abundant microbes. A novel hypothesis explains the contradictory results in this field and advocates for more in-depth research to be conducted on variables that may not exhibit statistical significance. The obtained results seem to contribute to an explanation of the contradictory findings regarding ASD and its relation with gut microbiota composition. While some research correlates higher ratios of Bacillota/Bacteroidota, others find the opposite. These discrepancies are closely linked to the minority organisms in the microbiome's composition, which may differ between populations but share similar metabolic functions. Therefore, the ratios of Bacillota/Bacteroidota regarding ASD may not be determinants in the manifestation of ASD.

A novel method for petroleum and natural gas resource potential evaluation and prediction by support vector machines (SVM)

Petroleum and natural gas resources (PNGR) are some of the major forms of fossil energy that are important for the development of industry and energy security. Along with the growing demand of petroleum consumption and the requirement for enhancing drilling success rate, reducing the exploration risk and saving exploration cost, prediction method for PNGR potential with high accuracy and wide practicability is needed. However, the existing PNGR evaluation and prediction methods based on traditional statistical principles are far from meeting the requirements of the present petroleum exploration and exploitation. Therefore, this study introduces a novel method for PNGR potential prediction by applying support vector machines (SVM) in the context of the rapid development of artificial intelligence and machine learning. This novel machine learning methodology first proposed a combination of support vector classification (SVC) for hydrocarbon accumulation probability prediction and then support vector regression (SVR) for reserve abundance prediction. The combining use of classification and regression model can fully utilize the professional knowledge of petroleum geology and the powerful data processing capabilities of machine learning algorithms and hence significantly improve the performance of the method. Furthermore, the dataset is set based on petroleum geology knowledge with the feature variables of source rock, sandstone reservoirs, sealing capacity and hydrocarbon migration, whose distribution are predictable and thus ensures the predictive effect in practical petroleum exploration. The results show that the testing accuracy of the hydrocarbon accumulation probability evaluation model by SVC ranges from 80% to 100% with an average of 88.92%. The performance of the SVR model for evaluating reserve abundance also performs well with the highest correlation coefficient of 0.767. In addition, several validation ways are applied for testing the reliability and stability of the model. For a hold-out test for a new zone, the model provides precise prediction of hydrocarbon accumulation probability and reserve abundance with an accuracy of 72.5% and a correlation coefficient of 0.744. The evaluation metric of the F1-score shows an average of 0.91 for the SVC models, the 4-fold cross-validation shows an average correlation coefficient of 0.663 for SVR model, which indicates the good performance of the SVC and SVR model. To conclude, this study not only provides an intelligent ML method system for PNGR potential precisely evaluation and prediction with the combination of SVC and SVR which is firstly used by application of ML in petroleum industry field, but is also significant for the application of ML in petroleum and natural gas exploration and exploitation.

Applying support vector machine (SVM) using GPS-TEC and Space Weather parameters to distinguish ionospheric disturbances possibly related to earthquakes

The effort to identify and comprehend potential earthquake-related phenomena shares a common goal: successful earthquake forecasting. Advancements in science and technology have made this goal multidisciplinary. Currently, possibly earthquake-related anomalies in the Vertical Total Electron Content (VTEC) of the Earth’s ionosphere are being investigated. Global Navigation Satellite Systems (GNSS) can be used to calculate this ionospheric parameter. In this research work, GPS VTEC was calculated for periods between the years 2015 and 2019. The selection of this periods considered both seismically active and non-seismically areas in Mexico. The Mw ≥ 5 earthquakes under study were registered by the National Seismological Service. Moreover, different geomagnetic storm and solar activity parameters, such as the geomagnetic equatorial Dst index and the F10.7 index, were analyzed. Additionally, the daily average and monthly mean number of sunspots (R, SSN, respectively) were included as a direct, long-term record of the development of the solar cycle. To the periods under study different statistical methods were applied, such as Mean-Square Error (MSE) and cross-correlation. The above aims to apply a machine learning technique capable of classifying between periods with seismic and non-seismic activity. The features were constructed using statistical data and results from the implemented analysis. Furthermore, Principal Component Analysis (PCA) was applied to reduce the feature vector dimensions, and accuracy scores were compute using k-fold cross-validation. The results from the Support Vector Machine (SVM) model indicated an accuracy of 88.9% for the training set, and an accuracy of 80% was obtained for the test set. One of the limitations of the current study was the sample size. However, the present initial approach for classifying seismic events from non-seismic periods using SVM demonstrated promising results when considering the indicated parameters and the days under study.

Modified Cuttlefish Swarm Optimization with Machine Learning-Based Sustainable Application of Solid Waste Management in IoT

The internet of things (IoT) paradigm roles an important play in enhancing smart city tracking applications and managing city procedures in real time. The most important problem connected to smart city applications has been solid waste management, which can have adverse effects on society’s health and environment. Waste management has developed a challenge faced by not only evolving nations but also established and developed counties. Solid waste management is an important and stimulating problem for environments across the entire world. Therefore, there is the need to develop an effective technique that will remove these problems, or at least decreases them to a minimal level. This study develops a modified cuttlefish swarm optimization with machine learning-based solid waste management (MCSOML-SWM) in smart cities. The MCSOML-SWM technique aims to recognize different categories of solid wastes and enable smart waste management. In the MCSOML-SWM model, a single shot detector (SSD) model allows effectual recognition of objects. Then, a deep convolutional neural network-based MixNet model was exploited to produce feature vectors. Since trial-and-error hyperparameter tuning is a tedious process, the MCSO algorithm was applied for automated hyperparameter tuning. For accurate waste classification, the MCSOML-SWM technique applies support vector machine (SVM) in this study. A comprehensive set of simulations demonstrate the improved classification performance of the MCSOML-SWM model with maximum accuracy of 99.34%.

Clinical Hematochemical Parameters in Differential Diagnosis between Pediatric SARS-CoV-2 and Influenza Virus Infection: An Automated Machine Learning Approach.

The influenza virus and the novel beta coronavirus (SARS-CoV-2) have similar transmission characteristics, and it is very difficult to distinguish them clinically. With the development of information technologies, novel opportunities have arisen for the application of intelligent software systems in disease diagnosis and patient triage. A cross-sectional study was conducted on 268 infants: 133 infants with a SARS-CoV-2 infection and 135 infants with an influenza virus infection. In total, 10 hematochemical variables were used to construct an automated machine learning model. An accuracy range from 53.8% to 60.7% was obtained by applying support vector machine, random forest, k-nearest neighbors, logistic regression, and neural network models. Alternatively, an automated model convincingly outperformed other models with an accuracy of 98.4%. The proposed automated algorithm recommended a random tree model, a randomization-based ensemble method, as the most appropriate for the given dataset. The application of automated machine learning in clinical practice can contribute to more objective, accurate, and rapid diagnosis of SARS-CoV-2 and influenza virus infections in children.

Prediction of severity of aviation landing accidents using support vector machine models

The purpose of this study was to apply support vector machine (SVM) models to predict the severity of aircraft damage and the severity of personal injury during an aircraft approach and landing accident and to evaluate and rank the importance of 14 accident factors across 39 sub-categorical factors. Three new factors were introduced using the theory of inattentional blindness: The presence of visual area surface penetrations for a runway, the Federal Aviation Administration’s (FAA) visual area surface penetration policy timeframe, and the type of runway approach lighting. The study comprised 1,297 aircraft approach and landing accidents at airports within the United States with at least one instrument approach procedure. Support vector machine models were developed in using the linear, polynomial, radial basis function (RBF), and sigmoid kernels for the severity of aircraft damage and additional SVM models were developed for the severity of personal injury.The SVM models using the RBF kernel produced the best machine learning models with a 96% accuracy for predicting the severity of aircraft damage (0.94 precision, 0.95 recall, and 0.95 F1-score) and a 98% accuracy for predicting the severity of personal injury (0.99 precision, 0.98 recall, and 0.99 F1-score). The top predictors across both models were the pilot’s total flight hours, time of the accident, pilot’s age, crosswind component, landing runway number, single-engine land certificate, and any obstacle penetration. This study demonstrates the benefit of SVM modeling using the RBF kernel for accident prediction and for datasets with categorical factors.

Bioelectronic tongue dedicated to the analysis of milk using enzymes linked to carboxylated-PVC membranes modified with gold nanoparticles

Bioelectronic tongues (bioET) made of sensors combining enzymes and nanomaterials have been shown to be advantageous due to the specificity offered by the biosensors and the enhanced sensitivity provided by the nanomaterials. In this work, an innovative bioET for milk analysis is developed using potentiometric biosensors based on lactic dehydrogenase, galactose oxidase and urease specific for the detection of compounds of interest in milk (lactic acid, galactose and urea). The performance of the biosensors has been fostered by covalently immobilizing the enzymes on membranes of carboxylated polyvinyl chloride combined with gold nanoparticles. The design and composition of the biosensors contributes to preserving the enzymatic activity, allowing limits of detection in the range of 10 −5 – 10 −6 M with excellent sensitivity and reproducibility (variation coefficients ranged from 1 to 5.1%). The three biosensors, combined in a single device and coupled to a pattern recognition software, can discriminate efficiently twelve classes of milk with different fat content (skimmed, semi-skimmed and whole milk) and nutritional characteristics (calcium enriched, lactose free and folic acid-enriched). The bioET shows an excellent classification capability with an accuracy of up to 99.7%. By applying Support Vector Machine (SVM) analysis, the BioET can perform the simultaneous assessment of eight physicochemical parameters (acidity, fat, proteins, lactose, density, urea, dry matter and nonfat dry matter) with satisfactory correlation coefficients and low residual errors. The results are further improved by implementing ensemble methodologies. The proposed strategy has been demonstrated to be useful for improving the performance of bioETs in the dairy industry. • A potentiometric bioET specifically dedicated to milk analysis was developed. • Enzymes were covalently linked to membranes combining Carboxilated-PVC and AuNPs. • The effective enzymatic immobilization helped to retain the enzymatic activity. • Using SVM and ensemble methods, nine physicochemical parameters can be determined simultaneously.

A Simple and Effective Intrusion Detection System for Manets

This work proposes a simple and effective Intrusion Detection System (IDS) to classify different attacks in MANETs. IDS extracts four features for every traffic pattern and applies Support Vector Machine algorithm over them for the classification. Before applying the feature extraction, the input traffic pattern is subjected to pre-processing as it is composed of non-uniform features. IDS classifies the input traffic pattern into three classes; they are normal, blackhole and wormhole. Finally, this work analyses the feasibility of machine learning algorithms for the detection of security attacks in MANETs. For experimental validation, we have referred a self-created dataset which was acquired from the observations of blackhole and wormhole attacked node’s traffic patterns. Moreover, we have also validated the proposed method through NSL-KDD dataset.

Application of machine learning to improve the efficiency of electrophysiological simulations used for the prediction of drug-induced ventricular arrhythmia

Background and ObjectiveIn silico prediction of drug-induced ventricular arrhythmia often requires computationally intensive simulations, making its application tedious and non-interactive. This inconvenience can be mitigated using matrices of precomputed simulation results, allowing instantaneous computation of biomarkers such as action potential duration at 90% of the repolarisation (APD90). However, preparing such matrices can be computationally intensive for the method developers, limiting the range of simulated conditions. In this work, we aim to optimise the generation of these matrices so that they can be obtained with less effort and for a broader range of input values. MethodsMachine learning methods were applied, building models trained with only a small fraction of the originally simulated results. The predictive performances of the models were assessed by comparing their predicted values with the actual simulation results, using percentual mean absolute error and mean relative error, as well as the percentage of data with a relative error below 5%. ResultsOur method obtained highly accurate estimations of the original values, leading to a nearly one hundred-fold decrease in computation time. This method also allows precomputing more complex matrices, describing the effect of more ion channels on the APD90. The best results were obtained by applying Support Vector Machine models, which yielded errors below 1% in most cases. This approach was further validated by predicting the APD90 of a set of 12 CiPA compounds and exporting the optimal settings for predicting APD90 using a different set of ion channels, always with satisfactory results. ConclusionsThe proposed method effectively reduces the computational effort required to generate matrices of precomputed electrophysiological simulation values. The same approach can be applied in other fields where computationally costly simulations are applied repeatedly using slightly different input values.

Combining Transformer Embeddings with Linguistic Features for Complex Word Identification

Identifying which words present in a text may be difficult to understand by common readers is a well-known subtask in text complexity analysis. The advent of deep language models has also established the new state-of-the-art in this task by means of end-to-end semi-supervised (pre-trained) and downstream training of, mainly, transformer-based neural networks. Nevertheless, the usefulness of traditional linguistic features in combination with neural encodings is worth exploring, as the computational cost needed for training and running such networks is becoming more and more relevant with energy-saving constraints. This study explores lexical complexity prediction (LCP) by combining pre-trained and adjusted transformer networks with different types of traditional linguistic features. We apply these features over classical machine learning classifiers. Our best results are obtained by applying Support Vector Machines on an English corpus in an LCP task solved as a regression problem. The results show that linguistic features can be useful in LCP tasks and may improve the performance of deep learning systems.

Evaluation of plastic packaging waste degradation in seawater and simulated solar radiation by spectroscopic techniques

The goal of this study was to evaluate the degradation of plastic packaging waste in seawater and under simulated solar radiation by different spectroscopic techniques in view of their potential recycling. Selected packaging products made of polypropylene (PP), high-density polyethylene (HDPE) and polystyrene (PS) were weathered in a Solar Box for 4 cycles of 504 h each (21 days), for a total of 2016 h (84 days). At each aging cycle, digital images of the polymer samples were acquired to investigate surface changes in terms of cracking and fragmentation. Furthermore, color measurements were carried out by reflectance spectrophotometry in the visible range (VIS: 400–700 nm). Finally, hyperspectral imaging (HSI) in the shortwave infrared region (SWIR: 1000–2500 nm) coupled with chemometric techniques was applied to monitor the aging process in terms of reflectance spectral variations and to evaluate the possibility of predicting the polymer aging time.The results showed that plastic packaging samples are characterized by different degradation behaviors, such as surface alteration (i.e. fractures, grooves and deep cracks), color variation (i.e. yellowing and discoloration) and fragmentation and microplastic formation, corresponding to changes in reflectance and chemical bonds vibrations. The degradation phenomena are influenced by both thickness and polymer type, being more evident for PP and PS samples rather than for HDPE ones.Concerning the results based on SWIR-HSI, principal component analysis (PCA) highlighted the spectral variation of polymers occurring during the artificial aging process through the different clustering in the score plot. Furthermore, good correlations were obtained between polymer spectral signature and aging time applying Support Vector Machines (SVM) prediction model. The use of the proposed spectroscopic techniques, as well as the development of predictive models based on HSI, contributes to understand the behavior of polymers when exposed to marine environment. This approach represents a promising tool for the definition of the degradation status of polymers and for the selection of marine plastics characterized by a level of degradation compatible with their recycling, fully in line with the principles of circular and sustainable economy.

Clinical Predictive Modeling of Anastomotic Leaks Using Hyperspectral Imaging

INTRODUCTION: Anastomotic leaks (AL) remain virtually unchanged despite advancements in minimally invasive surgical techniques and postoperative care. Adequate intraoperative bowel perfusion assessment remains a standard of surgical care to prevent AL. Although several platforms offer objective assessment of bowel perfusion, their reliable use has not demonstrated AL prevention and involves cumbersome steps through sterile preparation or use of dyes. We harnessed machine learning to address these 2 concerns by generating a clinical-predictive modeling tool for AL using a well-established platform of tissue perfusion called hyperspectral imaging. METHODS: We analyzed oxyhemoglobin, deoxyhemoglobin, and oxygen saturation levels of pre- and postanastomotic bowel segments intraoperatively at Brigham and Women’s Hospital using the contactless, point-and-shoot platform called HyperView. Patients 18 years or older undergoing open or laparoscopic-assisted abdominal surgery were included in the study. We followed patients postoperatively for development of AL and perianastomotic abscesses, and preliminarily analyzed the data using random forests to assess variable importance. We will perform a final analysis of the cohort by applying support vector machine to train a model with 80% training and 20% testing partition using a multiple kernel approach. RESULTS: Preliminary results demonstrate a strong correlation of oxyhemoglobin and oxygen saturation levels in preanastomotic intestinal limbs with the incidence of AL. The variation in oxygen saturation and mean oxyhemoglobin levels in the common channel postanastomoses is correlated with development of perianastomotic abscesses. CONCLUSION: We report an innovative intraoperative assessment tool using a noninvasive, contactless imaging technology for intestinal perfusion assessment that can clinically predict incidence of AL and guide intraoperative decisions.

On Subsampling Procedures for Support Vector Machines

Herein, theoretical results are presented to provide insights into the effectiveness of subsampling methods in reducing the amount of instances required in the training stage when applying support vector machines (SVMs) for classification in big data scenarios. Our main theorem states that under some conditions, there exists, with high probability, a feasible solution to the SVM problem for a randomly chosen training subsample, with the corresponding classifier as close as desired (in terms of classification error) to the classifier obtained from training with the complete dataset. The main theorem also reflects the curse of dimensionalityin that the assumptions made for the results are much more restrictive in large dimensions; thus, subsampling methods will perform better in lower dimensions. Additionally, we propose an importance sampling and bagging subsampling method that expands the nearest-neighbors ideas presented in previous work. Using different benchmark examples, the method proposed herein presents a faster solution to the SVM problem (without significant loss in accuracy) compared with the available state-of-the-art techniques.

A machine-learning approach for predicting impaired consciousness in absence epilepsy.

Behavior during 3–4 Hz spike‐wave discharges (SWDs) in absence epilepsy can vary from obvious behavioral arrest to no detectible deficits. Knowing if behavior is impaired is crucial for clinical care but may be difficult to determine without specialized behavioral testing, often inaccessible in practice. We aimed to develop a pure electroencephalography (EEG)‐based machine‐learning method to predict SWD‐related behavioral impairment. Our classification goals were 100% predictive value, with no behaviorally impaired SWDs misclassified as spared; and maximal sensitivity. First, using labeled data with known behavior (130 SWDs in 34 patients), we extracted EEG time, frequency domain, and common spatial pattern features and applied support vector machines and linear discriminant analysis to classify SWDs as spared or impaired. We evaluated 32 classification models, optimized with 10‐fold cross‐validation. We then generalized these models to unlabeled data (220 SWDs in 41 patients), where behavior during individual SWDs was not known, but observers reported the presence of clinical seizures. For labeled data, the best classifier achieved 100% spared predictive value and 93% sensitivity. The best classifier on the unlabeled data achieved 100% spared predictive value, but with a lower sensitivity of 35%, corresponding to a conservative classification of 8 patients out of 23 as free of clinical seizures. Our findings demonstrate the feasibility of machine learning to predict impaired behavior during SWDs based on EEG features. With additional validation and optimization in a larger data sample, applications may include EEG‐based prediction of driving safety, treatment adjustment, and insight into mechanisms of impaired consciousness in absence seizures.

Classification of uranium ore concentrates applying support vector machine to spectrophotometric and textural features

Uranium ore concentrates (UOCs) are produced in the early stages of the nuclear fuel cycle, prior to conversion to uranium hexafluoride. Because of their high uranium content and the large-scale production, UOCs diversion from civilian use and proliferation are potential risks. This implies the necessity to develop methods able to recognise characteristic parameters correlating each UOC powder to its history and origin. Here, a novel methodology is proposed: first the reflectance spectra of 79 commercial UOCs are acquired and clustered by means of Ward's clustering analysis, then classified by Support Vector Machine (SVM). Second, SVM classification is applied to the image textural features extracted with the Grey Level Co-occurrence Matrix (GLCM) and the Angle Measure Technique (AMT) algorithms for powders in two different colour groups. The developed SVM models present good classification quality: a Matthews correlation coefficient (MCC) of 0.95 is obtained for the classification based on colours while macro-F1 is generally greater than 0.81 (MCC larger than 0.75) for the texture-based classification. These results reveal the potentiality of the present automated classification for the scopes of nuclear forensics in the identification of an unknown uranium ore concentrate sample.