Classification Of Events Research Articles

Vector to matrix representation for CNN networks for classifying astronomical data

Choosing the right classifier is crucial for effective classification in various astronomical datasets aimed at pattern recognition. While the literature offers numerous solutions, the support vector machine (SVM) continues to be a preferred choice across many scientific fields due to its user-friendliness. In this study, we introduce a novel approach using convolutional neural networks (CNNs) as an alternative to SVMs. CNNs excel at handling image data, which is arranged in a grid pattern. Our research explores converting one-dimensional vector data into two-dimensional matrices so that CNNs pre-trained on large image datasets can be applied. We evaluate different methods to input data into standard CNNs by using two-dimensional feature vector formats. In this work, we propose a new method of data restructuring based on a set of wavelet transforms. The robustness of our approach is tested across two benchmark datasets/problems: brown dwarf identification and threshold crossing event (Kepler data) classification. The proposed ensembles produce promising results on both datasets. The MATLAB code of the proposed ensemble is available at https://github.com/LorisNanni/Vector-to-matrix-representation-for-CNN-networks-for-classifying-astronomical-data

Enhancing Sound Source Localization via False Negative Elimination.

Sound source localization aims to localize objects emitting the sound in visual scenes. Recent works obtaining impressive results typically rely on contrastive learning. However, the common practice of randomly sampling negatives in prior arts can lead to the false negative issue, where the sounds semantically similar to visual instance are sampled as negatives and incorrectly pushed away from the visual anchor/query. As a result, this misalignment of audio and visual features could yield inferior performance. To address this issue, we propose a novel audio-visual learning framework which is instantiated with two individual learning schemes: self-supervised predictive learning (SSPL) and semantic-aware contrastive learning (SACL). SSPL explores image-audio positive pairs alone to discover semantically coherent similarities between audio and visual features, while a predictive coding module for feature alignment is introduced to facilitate the positive-only learning. In this regard SSPL acts as a negative-free method to eliminate false negatives. By contrast, SACL is designed to compact visual features and remove false negatives, providing reliable visual anchor and audio negatives for contrast. Different from SSPL, SACL releases the potential of audio-visual contrastive learning, offering an effective alternative to achieve the same goal. Comprehensive experiments demonstrate the superiority of our approach over the state-of-the-arts. Furthermore, we highlight the versatility of the learned representation by extending the approach to audio-visual event classification and object detection tasks. Code and models are available at: https://github.com/zjsong/SACL.

Hypnotic Use and the Risk of Cardiovascular Diseases in Insomnia Patients.

We aimed to examine the association between hypnotic agents and cardiovascular outcomes in general individuals with insomnia. In a propensity score matched cohort of UK Biobank (UKB) participants with insomnia, Cox proportional hazard model was used to estimate the association between regular use of hypnotic agents and predetermined cardiovascular outcomes including incident coronary heart diseases (CHD), heart failure (HF), stroke, and cardiovascular death. Inverse probability of treatment weighting, competing risk models, and shared frailty models were further performed during sensitivity analysis. Drug-target Mendelian randomization (MR) analyses were employed for further evaluation of the association between therapeutic targets of hypnotics and cardiovascular diseases. During a median follow-up of 14.3 years, the matched cohort documented a total of 929 CHD cases, 360 HF cases, 262 stroke cases, and 180 cardiovascular deaths. No significant association was detected between Z-meds and CHD, stroke, and cardiovascular mortality. Benzodiazepine use was significantly associated with the increased risk of CHD, HF, and cardiovascular mortality. The inverse probability of treatment weighting, competing risk models, and shared frailty models didn't alter the above associations. Moreover, drug-target MR analyses corroborated the safety of Z-meds in the general population regarding cardiovascular health. Our findings suggested the heterogeneous associations between different categories of hypnotics and incident cardiovascular events in individuals with insomnia. Both observational and genetic evidence raised safety concerns regarding the cardiovascular impact of benzodiazepines. No cardiovascular hazard of Z-meds was discovered in the UKB population with insomnia.

Optimization of activity-driven event detection for long-term ambulatory urodynamics.

Lower urinary tract dysfunction (LUTD) is a debilitating condition that affects millions of individuals worldwide, greatly diminishing their quality of life. The use of wireless, catheter-free implantable devices for long-term ambulatory bladder monitoring, combined with a single-sensor system capable of detecting various bladder events, has the potential to significantly enhance the diagnosis and treatment of LUTD. However, these systems produce large amounts of bladder data that may contain physiological noise in the pressure signals caused by motion artifacts and sudden movements, such as coughing or laughing, potentially leading to false positives during bladder event classification and inaccurate diagnosis/treatment. Integration of activity recognition (AR) can improve classification accuracy, provide context regarding patient activity, and detect motion artifacts by identifying contractions that may result from patient movement. This work investigates the utility of including data from inertial measurement units (IMUs) in the classification pipeline, and considers various digital signal processing (DSP) and machine learning (ML) techniques for optimization and activity classification. In a case study, we analyze simultaneous bladder pressure and IMU data collected from an ambulating female Yucatan minipig. We identified 10 important, yet relatively inexpensive to compute signal features, with which we achieve an average 91.5% activity classification accuracy. Moreover, when classified activities are included in the bladder event analysis pipeline, we observe an improvement in classification accuracy, from 81% to 89.0%. These results suggest that certain IMU features can improve bladder event classification accuracy with low computational overhead.Clinical Relevance: This work establishes that activity recognition may be used in conjunction with single-channel bladder event detection systems to distinguish between contractions and motion artifacts for reducing the incorrect classification of bladder events. This is relevant for emerging sensors that measure intravesical pressure alone or for data analysis of bladder pressure in ambulatory subjects that contain significant abdominal pressure artifacts.

Negative Life Events and Emotional Symptoms From Ages 2 to 30 Years

Exposure to different types of negative life events, including traumatic events, is common across the lifespan and associated with increased mental health symptoms. To assess whether vulnerability to negative life events varies across 5 developmental periods from preschool to young adulthood. This cohort study analyzed data from 3 community-representative studies set in the southeastern US (1992-2015) with harmonized assessment approaches that included a total of 13 775 assessments of individuals aged 2 to 30 years with up to 21 years of follow-up. Data analysis occurred from July 2023 to June 2024. Each study assessed lifetime exposure to (1) traumatic events (ie, severe events associated with posttraumatic stress disorder) and (2) recent stressful events (eg, loss of a friend or moving). All assessments were completed with structured interviews with participants and/or their caregivers. The primary outcome was emotional symptoms (ie, anxiety and depressive symptoms). Associations of both categories of life events with emotional symptoms were compared across preschool (<7 years), childhood (7-12 years), adolescence (13-17 years), late adolescence (18-22 years), and young adulthood (23-30 years). Analyses were based on 13 775 assessments of 3258 participants (1519 female [weighted percentage, 50.0%]). Recent stressful events were associated with emotional symptoms across each developmental period, ranging from a low in preschool (B =0.14; SE = 0.05) to a high in young adulthood (B = 0.57; SE = 0.12) in cross-sectional analyses and ranging from a low in childhood (B = 0.10; SE = 0.06) to a high adolescence (B = 0.19; SE = 0.05) in longitudinal analyses. Lifetime traumatic events were associated with emotional symptoms across each developmental period, ranging from a low in preschool (B = 0.18; SE = 0.05) to a high in adolescence (B = 0.28; SE = 0.04) in cross-sectional analyses and ranging from a low in childhood (B = 0.09; SE = 0.06) to a high in late adolescence (B = 0.21; SE = 0.05) in longitudinal analyses. Associations had overlapping 95% CIs across the different developmental periods with one exception: stressful events had a larger-magnitude cross-sectional association with emotional symptoms in young adulthood than in other developmental periods. Results were consistent with additive, rather than interactive, associations of traumatic and stressful events with emotional symptoms at each developmental period. In this cohort study of 3 community-representative samples, vulnerability to traumatic and stressful events was generally similar across the first 3 decades of life; both types of events had an independent association with emotional functioning. These findings suggest response to stressful events is similar from childhood to adulthood.

Boundedness supports children's event representations.

Natural languages distinguish between telic predicates that denote events leading to an inherent endpoint (e.g., draw a balloon) and atelic predicates that denote events with no inherent endpoint (e.g., draw balloons). Telicity distinctions in many languages are already partly available to 4-5-year-olds. Here, using exclusively nonlinguistic tasks and a sample of English-speaking children, we ask whether young learners use corresponding temporal notions to characterize event structure-that is, whether children represent events in cognition as bounded temporal entities with a specified endpoint or unbounded temporal units that could in principle extend indefinitely. We find that 4-5-year-old children in our sample compute boundedness during an event categorization task (Experiment 1) and distinguish event boundedness from event completion (Experiment 2). Furthermore, 4-5-year-olds in our sample evaluate interruptions at event endpoints versus midpoints differently-but only for events that are construed as bounded, presumably because in such construals, events truly culminate (Experiment 3). We conclude that young children represent events in terms of foundational and abstract temporal properties. These properties could support the acquisition of linguistic aspectual distinctions and further scaffold the way children conceptualize and process their dynamic experiences. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Detecting information from Twitter on landslide hazards in Italy using deep learning models

BackgroundMass media are a new and important source of information for any natural disaster, mass emergency, pandemic, economic or political event, or extreme weather event affecting one or more communities in a country. Several techniques have been developed for data mining in social media for many natural events, but few of them have been applied to the automatic extraction of landslide events. In this study, Twitter has been investigated to detect data about landslide events in Italian-language. The main aim is to obtain an automatic text classification on the basis of information about natural hazards. The text classification for landslide events in Italian-language has still not been applied to detect this type of natural hazard.ResultsOver 13,000 data were extracted within Twitter considering five keywords referring to landslide events. The dataset was classified manually, providing a solid base for applying deep learning. The combination of BERT + CNN has been chosen for text classification and two different pre-processing approaches and bert-model have been applied. BERT-multicase + CNN without preprocessing archived the highest values of accuracy, equal to 96% and AUC of 0.96.ConclusionsTwo advantages resulted from this studio: the Italian-language classified dataset for landslide events fills that present gap of analysing natural events using Twitter. BERT + CNN was trained to detect this information and proved to be an excellent classifier for the Italian language for landslide events.

Interval life events are an important determinant of heterogeneity in outcomes in a randomised trial: a novel, simple method of assessment

ObjectiveAlthough life events are clearly important to health, most of the scientific focus has been on baseline life events that occur prior to a study. Life events that occur after...

Ensemble Learning Improves the Efficiency of Microseismic Signal Classification in Landslide Seismic Monitoring.

A deep-seated landslide could release numerous microseismic signals from creep-slip movement, which includes a rock-soil slip from the slope surface and a rock-soil shear rupture in the subsurface. Machine learning can effectively enhance the classification of microseismic signals in landslide seismic monitoring and interpret the mechanical processes of landslide motion. In this paper, eight sets of triaxial seismic sensors were deployed inside the deep-seated landslide, Jiuxianping, China, and a large number of microseismic signals related to the slope movement were obtained through 1-year-long continuous monitoring. All the data were passed through the seismic event identification mode, the ratio of the long-time average and short-time average. We selected 11 days of data, manually classified 4131 data into eight categories, and created a microseismic event database. Classical machine learning algorithms and ensemble learning algorithms were tested in this paper. In order to evaluate the seismic event classification performance of each algorithmic model, we evaluated the proposed algorithms through the dimensions of the accuracy, precision, and recall of each model. The validation results demonstrated that the best performing decision tree algorithm among the classical machine learning algorithms had an accuracy of 88.75%, while the ensemble algorithms, including random forest, Gradient Boosting Trees, Extreme Gradient Boosting, and Light Gradient Boosting Machine, had an accuracy range from 93.5% to 94.2% and also achieved better results in the combined evaluation of the precision, recall, and F1 score. The specific classification tests for each microseismic event category showed the same results. The results suggested that the ensemble learning algorithms show better results compared to the classical machine learning algorithms.

Trajectory-based fish event classification through pre-training with diffusion models

This study contributes to advancing the field of automatic fish event recognition in natural underwater videos, addressing the current gap in studying fish interaction and competition, including predator-prey relationships and mating behaviors. We used the corkwing wrasse (Symphodus melops) as a model, a marine species of commercial importance that reproduces in sea-weed nests built and cared for by a single male. These nests attract a wide range of visitors and are the focal point for behavior such as spawning, chasing, and maintenance. We propose a deep learning methodology to analyze the movement trajectories of the nesting male and classify the associated events observed in their natural habitat. Our approach leverages unsupervised pre-training based on diffusion models, leading to improved feature learning. Additionally, we introduce a dataset comprising 16,937 trajectories across 12 event classes, making it the largest in terms of event class diversity. Our results demonstrate the superior performance of our method compared to several deep architectures. The code for the proposed method and the trajectories can be found at https://github.com/NoeCanovi/Fish_Behaviors_Generative_Models.

Home-video EEG monitoring in a pediatric setting

IntroductionPediatric video-EEG monitoring is a standard procedure in epilepsy clinics, typically conducted in in-hospital settings.However, hospitalizationis sometimesunnecessary and imposes a burden on children and their families. In response to the rise of telehealth, home video-EEG monitoring has emerged, utilizing portable EEG equipment and video-cameras. ObjectiveThe aim of this study was to assess the feasibility of home video-EEGin a pediatric population. MethodsWe conducteda prospective pilot study of twentyhome video-EEG tests in children. We evaluated the quality of EEG and video recordings using a 5-point scale.Demographic, clinical and quality data were comparedto a similar group undergoing in-hospital video-EEG monitoring. ResultsTwenty children aged 2.1–17.2 years (mean 9.57 ± 1.01), 12 females (60 %), underwent home video-EEG. A higher proportion of children with intellectual disability/autism were observed in the home-EEG group compared to the in-hospital group: 12 patients (60 %) vs. 5 (25 %) (p < 0.05*, Fisher exact test). In the ambulatory group patients with developmental and epileptic encephalopathy were overrepresented (7 i.e., 35 % vs. 0), while those withself-limited childhood epilepsy were more prevalent in the in-hospital group (5 i.e., 25 % vs 0) (p < 0.05*, Chi square). In the ambulatory group the reasons for referral were seizure localization/classification in 11 patients (55 %), paroxysmal event classification in 5 (25 %) and quantification of sleep epileptic activity in 4(20 %),similar to the in-hospital group (40 %, 40 % and 20 % respectively, p > 0.05, Chi square). The quality of the EEG recording was higher compared to in-hospital tests: median 5 [IQR 3.25–5] vs 4[IQR 3–4] (p < 0.05*, Mann-Whitney U test), while the quality of video recording was lower compared to in-hospital recordings: median 3[IQR 2.25–4] vs 5[IQR4-5] (p < 0.01**, Mann-Whitney U test). ConclusionsHome video-EEG monitoring is apromising option forlong-termpediatric EEG monitoring, particularlyfor children with special needs.

Patient Harm Events and Associated Cost Outcomes Reported to a Patient Safety Organization.

The aims of the study were to describe inpatient harm events detected via an automatic electronic trigger system (ETS) and to consider their financial consequences. Over a 27-month period, inpatient harm events were identified and documented in 1 healthcare system with 37 acute care facilities. Patients who experienced harm (all harm or preventable harm only) were compared with similar patients who did not. Clinical, financial, and demographic data were used to identify labor-adjusted direct variable costs (DVC) and potential differences in length of stay (LOS) associated with all-harm, preventable-harm, and nonharmed cohorts. Age-adjusted Charlson Comorbidity Index, case mix index, diagnosis-related groups, major diagnostic category, sex, age, location, diagnosis, adverse event category and subcategory, preventability, and harm severity were used to compare cohorts. Total harm events reported via the ETS and the health system's voluntary event reporting system were compared. Nearly 93,000 encounters for all-harm (n = 25,665) and nonharmed cohorts (n = 67,217) were compared by random sampling of diagnosis-related group-matched all-harm and nonharmed groups to ensure similar clinical conditions, as measured by Charlson Comorbidity Index and case mix index. Sampling (2 groups, n = 100 and n = 200) showed that increased LOS was associated with harm; yet other clinical comparators were similar across groups. the preventable-harm subcohort had longer LOS (10.7 versus 5.9 days) and higher DVC ($13,442 versus $8024) than the nonharmed cohort. Identification of harm events was nearly 6-fold higher with the ETS than with the voluntary event reporting system. Patients with preventable harm had increased LOS that was associated with higher DVC per preventable-harm encounter in a large US healthcare system.

First insights into the new particle formation and growth at a north-eastern Romanian urban site, Iasi. Potential health risks from ultrafine particles

New particle formation (NPF) process brings significant contribution to the global atmospheric particle number concentrations. Evidence on the formation and growth of NPF is reported for the first time at a Romanian urban site, i.e., Iasi, north-eastern Romania. Size-dependent aerosol number concentrations were obtained during two short-term campaigns undertaken in 2017 by using a scanning mobility particle sizer. The existence of two categories of events can be highlighted by investigating the net maximum increase in the nucleation mode particle number concentration and the maximum size of the geometric median diameter of new particles. The variability of meteorological parameters showed that the solar radiation peak was usually associated with NPF events, while relative humidity was anti-correlated with those events. Calculated nucleation rate values for events in May, median of 4.5 cm−3 s−1, was lower than the corresponding median values of 8.6 cm−3 s−1 for December. The particle growth rates showed a similar trend with a median of 4.0 nm h−1 and 7.7 nm h−1, in May and December, respectively. The obtained results suggest that regional emission sources might bring some contributions on the particle nucleation and growth processes, particularly over the cold season. Regarding the deposition of particles in the respiratory system, it appears that ultrafine particles are among the most significant contributors to the alveolar deposits. Moreover, evidences were obtained that the impact on the particle deposition at the alveolar region is not mandatory in direct relation to the intensity of the NPF event, the pollution burden of the particles most probably playing an important role.

Contribution of El Niño Southern Oscillation (ENSO) Diversity to Low‐Frequency Changes in ENSO Variance

AbstractEl Niño Southern Oscillation (ENSO) diversity is characterized based on the longitudinal location of maximum sea surface temperature anomalies (SSTA) and amplitude in the tropical Pacific, as Central Pacific events are typically weaker than Eastern Pacific events. SSTA pattern and intensity undergo low‐frequency modulations, affecting ENSO prediction skill and remote impacts, and resulting in low‐frequency changes in ENSO variance. Yet, how different ENSO types contribute to these decadal variance changes remains unclear. Here, we decompose the low‐frequency changes of ENSO variance into contributions from ENSO diversity categories. We propose a fuzzy clustering of monthly SSTA to allow for non‐binary event category memberships, where each event can belong to different clusters. Our approach identifies two La Niña and three El Niño categories and shows that the major shift of ENSO variance in the mid‐1970s was associated with an increasing likelihood of strong La Niña and extreme El Niño events.

Hydrodynamic modelling for simulating nearshore waves and sea levels: classification of extreme events from the English Channel to the Normandy coasts

Hydrodynamic modelling for simulating nearshore waves and sea levels: classification of extreme events from the English Channel to the Normandy coasts

Artificial neural networks for PIO events classification comparing different data collection procedures

Artificial neural networks for PIO events classification comparing different data collection procedures

Analysis of Attack Intensity on Autonomous Mobile Robots

Autonomous mobile robots (AMRs) combine a remarkable combination of mobility, adaptability, and an innate capacity for obstacle avoidance. They are exceptionally well-suited for a wide range of applications but usually operate in uncontrolled, non-deterministic environments, so the analysis and classification of security events are very important for their safe operation. In this regard, we considered the influence of different types of attacks on AMR navigation systems to subdivide them into classes and unified the effect of attacks on the system through their level of consequences and impact. Then, we built a model of an attack on a system, taking into account five methods of attack implementation and identified the unified response thresholds valid for any type of parameter, which allows for creating universal correlation rules and simplifies this process, as the trigger threshold is related to the degree of impact that the attack has on the finite subsystem. Also, we developed a methodology for classifying incidents and identifying key components of the system based on ontological models, which makes it possible to predict risks and select the optimal system configuration. The obtained results are important in the context of separating different types of destructive effects based on attack classes. Our study showed that it is sometimes difficult to divide spoofing attacks into classes by assessing only one parameter since the attacker can use a complex attack scenario, mixing the stages of the scenarios. We then showed how adding an attack intensity factor can make classification more flexible. The connections between subsystems and parameters, as well as the attack impact patterns, were determined. Finally, a set of unique rules was developed to classify destructive effects with uniform response thresholds for each parameter. In this case, we can increase the number of parameters as well as the type of parameter value.

Research on the classification of complex noise-mixed microseismic events based on machine vision

Event classification is important for accurately monitoring and warning against rockburst hazards using microseismic technology. Here, we propose an automatic classification method for microseismic events based on machine vision. The method uses Histogram of Oriented Gradient (HOG) integrated with Support Vector Machine (SVM) as the core model (HOG-SVM, HSVM) to classify microseismic events. First, the method uses as input spectrograms generated from microseismic event signals recorded in the field. Next, the HOG method is used to accurately extract the spectral feature information of the useful signals of microseismic events under the interference of noisy signal. Finally, the extracted feature data is used to train SVM, after the training is completed, the SVM is used to classify the microseismic events. The performance of the method for categorizing microseismic events was tested using multiple independent test sets built from data monitored in the field of a mine in Shandong Province. The results show that the method can effectively extract the spectral feature information of useful signals of microseismic events contaminated with noise, with good classification accuracy and robustness to noise. It classifies microseismic events with high accuracy and efficiency compared to well-performing classification methods based on seismic source parameters and typical depth models. The method can provide technical support for the effective classification of microseismic events in complex construction sites, especially in noisy deep underground construction environments.

ChickenSense: A Low-Cost Deep Learning-Based Solution for Poultry Feed Consumption Monitoring Using Sound Technology

This research proposes a low-cost system consisting of a hardware setup and a deep learning-based model to estimate broiler chickens’ feed intake, utilizing audio signals captured by piezoelectric sensors. The signals were recorded 24/7 for 19 consecutive days. A subset of the raw data was chosen, and events were labeled in two classes, feed-pecking and non-pecking (including singing, anomaly, and silence samples). Next, the labeled data were preprocessed through a noise removal algorithm and a band-pass filter. Then, the spectrogram and the signal envelope were extracted from each signal and fed as inputs to a VGG-16-based convolutional neural network (CNN) with two branches for 1D and 2D feature extraction followed by a binary classification head to classify feed-pecking and non-pecking events. The model achieved 92% accuracy in feed-pecking vs. non-pecking events classification with an f1-score of 91%. Finally, the entire raw dataset was processed utilizing the developed model, and the resulting feed intake estimation was compared with the ground truth data from scale measures. The estimated feed consumption showed an 8 ± 7% mean percent error on daily feed intake estimation with a 71% R2 score and 85% Pearson product moment correlation coefficient (PPMCC) on hourly intake estimation. The results demonstrate that the proposed system estimates broiler feed intake at each feeder and has the potential to be implemented in commercial farms.

Analysis of informativeness of features of classification of dangerous weather events based on radar observation results

One of the crucial factors affecting the safety and regularity of state and civil aviation flights is the meteorological situation. The European territory of Russia is most characterized by dangerous meteorological phenomena associated with cumulonimbus clouds: shower, thunderstorms, hail, accompanied by high atmospheric turbulence. Currently, meteorological radar stations are an indispensable source of information about the weather situation for air transport. The criteria for the classification of meteorological phenomena used in modern radar stations are formed for each event separately and are based on knowledge only about the altitude distribution of radar reflectivity and air temperature, despite the fact that radar data assess the wind characteristics of the atmosphere. It is shown that optimization of the classification criteria for the mentioned meteorological phenomena should be realized by generalization of the criteria and their construction in accordance with the theory of statistical hypothesis distinction, as well as by additional use of information on atmospheric turbulence. Based on the analysis of radar signals reflected from the meteorological events of shower, thunderstorm, and hail, probability distributions of reflectivity and specific dissipation rate of turbulent energy were obtained. Statistical analysis of probability distribution densities was carried out for: the maximum value of reflectivity Zmax, its dependence on height H(Zmax), as well as the maximum specific dissipation rate of turbulent energy EDRmax and the value H(EDRmax). The classification criterion based on the maximum probability functional was chosen to determine the structure of classification algorithms and decision rules. At the same time under the acceptable confidence is accepted the value of the probability of correct classification not lower than 0.8. For the accepted criterion the decision thresholds are constructed and the complete matrices of classification probabilities are calculated. The results of calculations showed that the worst informativeness in the classification of dangerous meteorological events of cumulonimbus cloudiness have parameters H(Zmax), H(EDRmax). Parameters Zmax, EDRmax have greater separating ability, but even for them the confidence of classification is unacceptable. In the article to increase the confidence of classification the joint use of features in the form of multivariate probability distribution densities of information parameters was applied. The best results are achieved when three p(Zmax,H(Zmax),EDRmax) and four p(Zmax,H(Zmax),EDRmax,H(EDRmax)) features are used. In the probability matrices for these cases, the maximum and acceptable at 0.8 level of probabilities of correct classification are achieved. Thus, the expansion of the feature space due to atmospheric turbulence is justified in the problem under consideration. These results will be refined with increasing observation time and will vary for different climatic zones. In general, the decision thresholds for classifying dangerous meteorological events of cumulonimbus cloudiness should be adaptive.