Selection Method Research Articles

Detailed Land Use Classification in a Rare Earth Mining Area Using Hyperspectral Remote Sensing Data for Sustainable Agricultural Development

In China, ion-adsorbing rare earth minerals are mainly located in the southern hilly areas and are important strategic resources. Extensive long-term mining has severely damaged the land cover in mining areas, caused soil pollution and terrain fragmentation, disrupted the balance between mining and agriculture, severely restricted agricultural development, and affected ecological development. Precise and detailed classification of land use within mining areas is crucial for monitoring the sustainable development of agricultural ecology in these areas. In this study, we leverage the high spatial and high spectral resolution characteristics of the Zhuhai-1 (OHS) hyperspectral image datasets. We create four types of datasets based on spectral, vegetation, red edge, and texture characteristics. These datasets are optimized for multifaceted features, considering the complex land use scenario in rare earth mining areas. Additionally, we design seven optimal combination schemes for features. This is performed to examine the impact of different schemes on land use classification in rare earth mining areas and the accuracy of identifying agricultural land classes from broken blocks. The results show that (1) the inclusion of texture features has the most obvious effect on the overall classification accuracy; (2) the red edge feature has the worst effect on improving the overall accuracy of the surface classification; however, it has a prominent effect on the identification of agricultural lands such as farmland, orchards, and reclaimed vegetation; and (3), following the combination of various optimization features, the land use classification yielded the highest overall accuracy, at 88.16%. Furthermore, the comprehensive identification of various agricultural land classes, including farmland, orchards, and greenhouse vegetables, yielded the most desirable outcomes. The research results not only highlight the advantages of hyperspectral images for complex terrain classification and recognition but also address the previous limitations in the application of hyperspectral datasets over wide mining areas. Additionally, the results underscore the reliability of feature selection methods in reducing information redundancy and improving classification accuracy. The proposed feature selection combination, based on OHS hyperspectral datasets, offers technical support and guidance for the detailed classification of complex land use in mining areas and the accurate monitoring of agroecological environments.

Fake News Detection Model

The proliferation of fake news across social media and various other platforms is a significant concern due to its potential to cause widespread social and national harm. Extensive research efforts are currently underway to identify and combat this issue. This study conducts an in-depth analysis of research pertaining to the detection of fake news, examining traditional machine learning models to determine the most effective approach. By leveraging supervised machine learning algorithms, a model will be developed to classify news articles as either true or false. This model will utilize tools such as Python's scikit-learn and NLP for textual analysis. The process will involve feature extraction and vectorization, with a focus on utilizing Python's scikit-learn library for text data tokenization and feature extraction, leveraging tools like Count Vectorizer and Tiff Vectorizer. Additionally, feature selection methods will be employed to identify the most relevant features for optimal precision, as determined by the results of the confusion matrix

The freshwater discharge into the Adriatic Sea revisited

The present study reconstructs the river discharge climatology and its respective historical series for all rivers of the Adriatic Sea with averaged climatological daily river discharge above 1 m3s−1, to reach a better representation of the Adriatic rivers in hydrodynamic models and, consequently, to develop a more realistic freshwater balance in the different regions of the hydrographic basin. Based on the European Flood Awareness System (EFAS) data set, a careful method of identification and selection of the Adriatic rivers, followed by a rigorous assessment against observational data, was developed to evaluate the current state of the Adriatic river discharges and their respective trends throughout several climate indicators from 1991 to 2022. Observational data are limited to 85% of the identified rivers, totaling 98% of the overall freshwater input into the Adriatic Sea. The results confirm that the Shallow Northern Adriatic receives the largest freshwater inputs with a daily average exceeding 2,400 m3s−1, which amounts to 61% of the overall Adriatic discharges. Consequently, this region guides the freshwater seasonal cycle of the Adriatic Sea, which presents a well-defined pattern of two flood peaks in late autumn and late spring, separated by a minimum discharge period at mid-summer. From the Central to the Southern Adriatic subregions, the absence of snow-melting effects prevents the secondary flood peak during the spring, shaping the seasonal cycle of river discharges from a single flood peak in late autumn to a drought period in August. The 32 years of continuous river discharge data reveal a negligible trend in the overall Adriatic Sea but a negative trend for the last decade (2013–2022). This decadal decrease is driven by the extreme drought that drastically pounded the northern Adriatic in 2022.

Deep reinforcement learning sensor scheduling for effective monitoring of dynamical systems

Advances in technology have enabled the use of sensors with varied modalities to monitor different parts of systems, each providing diverse levels of information about the underlying system. However, resource limitations and computational power restrict the number of sensors/data that can be processed in real-time in most complex systems. These challenges necessitate the need for selecting/scheduling a subset of sensors to obtain measurements that guarantee the best monitoring objectives. This paper focuses on sensor scheduling for systems modeled by hidden Markov models. Despite the development of several sensor selection and scheduling methods, existing methods tend to be greedy and do not take into account the long-term impact of selected sensors on monitoring objectives. This paper formulates optimal sensor scheduling as a reinforcement learning problem defined over the posterior distribution of system states. Further, the paper derives a deep reinforcement learning policy for offline learning of the sensor scheduling policy, which can then be executed in real-time as new information unfolds. The proposed method applies to any monitoring objective that can be expressed in terms of the posterior distribution of the states (e.g. state estimation, information gain, etc.). The performance of the proposed method in terms of accuracy and robustness is investigated for monitoring the security of networked systems and the health monitoring of gene regulatory networks.

A new surface molecularly imprinted polyacrylamide nanoprobe for trace Cr(VI) with RRS technique

This article was used potassium dichromate as the template molecule, silver nanoclusters as the nano matrix, acrylamide as the monomer, ethylene glycol dimethacrylate (EGDMA) as the crosslinking agent, and azodiisobutyronitrile (AIBN) as the initiator to prepare a new silver nanocluster surface MIP (AgNCs@MIP) nanoprobe for chromate. Upon addition of Cr(VI), it selectively adsorbs on the surface of AgNCs@MIP nanoprobes. The dichromate ion absorption peak at 350 nm overlaps with the AgNCs@MIP RRS peak at 370 nm, resulting in strong RRS energy transfer (RRS-ET) and a decrease in the RRS intensity. The decreased RRS intensity is directly proportional to the concentration of dichromate ions in the range of 0.0025–0.015 µmol/L, with a detection limit of 0.8 nmol/L. Therefore, a simple, fast, sensitive and selective RRS method for the determination of trace Cr(VI) in mineral water has been established, with a relative standard deviation of 9.2–9.8 % and recovery of 95.20 %—103.60 %.

Efficient Feature Learning Model of Motor Imagery EEG Signals with L1-Norm and Weighted Fusion

Brain–computer interface (BCI) for motor imagery is an advanced technology used in the field of medical rehabilitation. However, due to the poor accuracy of electroencephalogram feature classification, BCI systems often misrecognize user commands. Although many state-of-the-art feature selection methods aim to enhance classification accuracy, they usually overlook the interrelationships between individual features, indirectly impacting the accuracy of feature classification. To overcome this issue, we propose an adaptive feature learning model that employs a Riemannian geometric approach to generate a feature matrix from electroencephalogram signals, serving as the model’s input. By integrating the enhanced adaptive L1 penalty and weighted fusion penalty into the sparse learning model, we select the most informative features from the matrix. Specifically, we measure the importance of features using mutual information and introduce an adaptive weight construction strategy to penalize regression coefficients corresponding to each variable adaptively. Moreover, the weighted fusion penalty balances weight differences among correlated variables, reducing the model’s overreliance on specific variables and enhancing accuracy. The performance of the proposed method was validated on BCI Competition IV datasets IIa and IIb using the support vector machine. Experimental results demonstrate the effectiveness and superiority of the proposed model compared to the existing models.

Improving with Hybrid Feature Selection in Software Defect Prediction

Software defect prediction (SDP) is used to identify defects in software modules that can be a challenge in software development. This research focuses on the problems that occur in Particle Swarm Optimization (PSO), such as the problem of noisy attributes, high-dimensional data, and premature convergence. So this research focuses on improving PSO performance by using feature selection methods with hybrid techniques to overcome these problems. The feature selection techniques used are Filter and Wrapper. The methods used are Chi-Square (CS), Correlation-Based Feature Selection (CFS), and Forward Selection (FS) because feature selection methods have been proven to overcome data dimensionality problems and eliminate noisy attributes. Feature selection is often used by some researchers to overcome these problems, because these methods have an important function in the process of reducing data dimensions and eliminating uncorrelated attributes that can cause noisy. Naive Bayes algorithm is used to support the process of determining the most optimal class. Performance evaluation will use AUC with an alpha value of 0.050. This hybrid feature selection technique brings significant improvement to PSO performance with a much lower AUC value of 0.00342. Comparison of the significance of AUC with other combinations shows the value of FS PSO of 0.02535, CFS FS PSO of 0.00180, and CS FS PSO of 0.01186. The method in this study contributes to improving PSO in the SDP domain by significantly increasing the AUC value. Therefore, this study highlights the potential of feature selection with hybrid techniques to improve PSO performance in SDP.

Efficient and non-destructive classification of lily bulbs by laser-induced breakdown spectroscopy combined with machine learning methods

The study presented here is a pioneering attempt to utilize Laser-Induced Breakdown Spectroscopy (LIBS) for the non-destructive testing and classification of lily bulbs. Given the complexities and time-consuming nature of traditional classification methods based on organic matter content detection, the gradual adoption of spectral technologies as an alternative tool is timely. In this research, the Isolation Forest algorithm was employed to eliminate outliers, thereby enhancing data quality. Various variable selection methods combined with five different classification algorithms were used to establish models. Both full spectrum and feature band variable classification models achieved classification accuracy rates exceeding 99% on the prediction set. Among these, the Logistic Regression (LR) model demonstrated the most exceptional performance, with classification accuracy rates of 100% in both cross-validation and prediction sets. Through the analysis of feature variable importance ranking and spectral line elements, elements such as Na, K, Fe, Ca, P, Mg, and N were identified as potentially crucial for the identification of the origin of lily bulbs. These findings demonstrate the effectiveness of using LIBS combined with machine learning to classify lily bulbs, and provide a theoretical reference for origin identification and variety recognition of other medicinal and edible foods.

Optimized data driven fault detection and diagnosis in chemical processes

Fault detection and diagnosis (FDD) is crucial for ensuring process safety and product quality in the chemical industry. Despite the large amounts of process data recorded and stored in chemical plants, most of them are not well-labeled, and their conditions are not adequately specified. In this study, an optimized data-driven FDD model was developed for a chemical process based on automatic clustering algorithms. Due to data preprocessing importance, feature selection was performed by a non-dominated sorting genetic algorithm (NSGAII) based on k-means clustering. The optimal subset of features is selected by comparing clustering results for each subset. The performance of the proposed feature selection method was compared with the Fisher discriminant ratio (FDR), and XGBoost methods. The t-distributed stochastic neighbor embedding (t-SNE), Isomap, and KPCA dimension reduction methods were also employed for feature extraction. Finally, automatic clustering was performed based on metaheuristic algorithms for fault detection and diagnosis. Results were compared with non-automatic clustering methods. The performance of the proposed method was evaluated by examining the Tennessee Eastman and four water tank processes as case studies. The results showed that the proposed method is reliable and capable of online and offline chemical process fault detection and diagnosis. As a result, the findings of this study can be used to stabilize the operation of chemical processes.

Artificial activation of sperm motility in vitro.

The sperm activation method is a modern methodological approach that is used more and more often in practice. The number of studies focused on methods of artificial activation of human sperm motility are constantly increasing. Standard sperm selection methods can fail in some cases, among other things, because very young sperm are isolated that have not yet completed their development. In these cases, artificial stimulation of their movement can have a positive effect and greatly facilitate and faster the process of selecting suitable sperm. Methylxanthines are most often used as activating agents. However, opinions on the safety of using these substances on sperm are not uniform. The aim of the thesis is to present current knowledge about artificial activation of sperm motility for in vitro fertilization and subsequent embryonic development. Research of relevant literature in Web of Science, Scopus, PubMed/Medline databases. The literature analysis shows that this method is safe and effective in the selection of immotile spermatozoa. Scientific studies have been conducted to verify the safety and reliability of this method. The conclusion of these studies is the positive impact of this method of selection, especially in cases of sperm obtained from testicular tissue after method testicular sperm extraction. In these cases, the method of artificial sperm activation facilitated and accelerated the selection of sperm before intracytoplasmic sperm injection. Undamaged spermatozoa, which are immobile due to incomplete maturation, were activated.

International Collaborative Study Comparing Outcomes of Fenestrated Endovascular Aortic Repair in OCtogenarian vs Non-Octogenarian Patients: The FEVOC study.

With an increasing life expectancy, more octogenarian patients are referred with complex aortic aneurysms (cAAA). The aim of this study was to evaluate short and mid-term outcomes following fenestrated aortic repair (FEVAR) in octogenarians. Few studies looking at octogenarian-specific outcomes with diverging results. Retrospective, multicentre cohort study including consecutive patients undergoing elective FEVAR for cAAAs or type IV thoracoabdominal aortic aneurysms between 2007-2022 in eight high-volume centres. Octogenarians vs. non-octogenarians were compared. The primary outcome was 30-day mortality. Secondary outcomes included 1, 2 and 5-year survival and reintervention rates. A total of 729 patients (median age of 74.8 years [IQR 69.2 - 79.14]) were included, 169 (23%) of which were octogenarians, with 316 (43.3%) patients undergoing juxta/pararenal aneurysm repair. Although octogenarians presented less complex but larger (61mm vs. 58mm) aneurysms, the number of fenestrations was similar across groups. No differences in in-hospital mortality (4.1 vs. 3.0%), MAE (16.6% vs 12.2%) or reintervention rates (11.2 vs. 10%) were found. Multivariable logistic regression of in-hospital mortality identified BMI (OR 0.66, 95% CI 0.51-0.95, P=0.003), chronic heart failure (OR 7.70, 95% CI 1.36-36.15, P=0.003) and GFR<45ml/min/1.73m2 (OR 5.25, 95% CI 1.20-22.86, P=0.027) as independent predictors. Median follow-up was 41 months. The 1, 2 and 5-year survival rates were 91.3%, 81.8% and 49.5% in octogenarians vs 90.6%, 86.5% and 68.8% in non-octogenarian patients (Log-rank: =0.001). Freedom from aortic-related death and freedom from reintervention at five-years were similar across groups (log-rank=0.94 and .76, respectively). Age>80 was not an independent predictor of 30-day or long-term mortality on multivariable and Cox regression analysis. Elective FEVAR in octogenarians appears to be safe, with similar outcomes as in younger patients. Future studies looking at improved patient selection methods to ensure long-term survival benefits in both octogenarians and younger patients are warranted.

Multi-model combination in key steps for landslide susceptibility modeling and uncertainty analysis: a case study in Baoji City, China

Reliable landslide susceptibility maps are essential for geohazard risk management. However, the selection of models and methods in the landslide susceptibility modeling (LSM) process is subjective and can lead to uncertainties in susceptibility outcomes. This paper introduces a framework for LSM and uncertainty analysis that leverages multiple models in key steps. The framework dynamically operates and combines models at each LSM step to generate susceptibility evaluation results from various method combinations. The uncertainty of different model combinations is then analyzed by comparing these results. The framework’s effectiveness is validated using Baoji City as a case study, examining the impact of different attribute interval numbers (AIN), non-landslide negative sample selection methods, and prediction models on susceptibility prediction outcomes. The results of each group show that the highest AUC value is 0.963 (AIN = 12, buffer-controlled sampling, Random forest) is about 0.26 higher than the lowest (no AIN, low-slope controlled sampling, Support vector machine). The findings suggest that the combinations involving larger AIN values, buffer-controlled sampling, XGBoost, or Random forest model yield relatively high AUC accuracy and relatively low uncertainty in the susceptibility index.

Quality study of animal-derived traditional Chinese medicinal materials based on spectral technology: Calculus bovis as a case.

Calculus bovis (C. bovis) is a typical traditional Chinese medicine (TCM) derived from animals, which has a remarkable curative effect and high price. Rapid identification of C. bovis from different types was realized based on spectral technology, and a rapid quantitative analysis method for the main quality control indicator bilirubin was established. We conducted a supervised and unsupervised pattern recognition study on 44 batches of different types of C. bovis by five spectral pretreatment methods. Three variable selection methods were used to extract the essential information, and the partial least squares regression (PLSR) quantitative model of bilirubin by near-infrared (NIR) spectroscopy was constructed. The partial least squares discriminant analysis (PLS-DA) model could achieve 100% accuracy in identifying different types of C. bovis. The R2 of the NIR quantitative model was 0.979, which is close to 1, and the root mean square error of calibration (RMSEC) was 2.3515, indicating the good prediction ability of the model. The study was carried out to further improve the basic data of quality control of C. bovis and help the high-quality development of TCM derived from animals.

Structural aspects of personality typology of patients suffering from cardiovascular pathology associated with type 2 diabetes mellitus

BACKGROUND: Diabetes mellitus started as a metabolic disorder, but finished as a vascular pathology, which determines the prognosis and outcomes of disease. The research of personality psychotypes of patients suffering from type 2 diabetes mellitus associated with cardiovascular pathology was undertaken in order to study one of the basic components of system-forming factor of the pathological nervous regulation system.&#x0D; MATERIALS AND METHODS: 148 patients were examined (49 men, 99 women), divided into 2 groups: 1) patients with combined cardiovascular pathology and type 2 diabetes mellitus (monitoring group), 2) patients with cardiovascular pathology without type 2 diabetes mellitus (comparison group). Additional control group: 30 practically healthy volunteers. We conducted a questionnaire and a clarifying interview with profile patients undergoing a planned course of treatment in the therapeutic clinics. The domestic version of the psychological questionnaire test mini-mult AMPQ (Abbreviated Multifactorial Personality Questionnaire), which is an abbreviated form of MMPI, as well as “M. Lusher Color Selection Method” were used. The Nickstormsoftware computer software package “Oskord” (version 1.2) was used for statistical processing of the results.&#x0D; RESULTS: It was found: there are three main psychotypes — asteno-neurotic, epileptoid-exitable, hyperthymic — prevail in the total population of patients suffering from type 2 diabetes mellitus associated with cardiovascular pathology. The most favorable clinical situation is observed in patients with hyperthymic personality type, what demonstrates the systemic level influence of the psychotype. The definition of the personality structure implements a personalized approach to the patient, because it allows you to take into account the characteristics of each psychotype.&#x0D; CONCLUSION: Psychical factors affect the treatment of diabetes and concomitant pathology to the same extent as the treatment itself can affect the psychic state. Such a psychophysiological nature of interaction determines the modeling, possibly organizing role of psychotype components in the work of central link of the nervous regulation pathological system — the basis of the cardiovascular pathology associated with type 2 diabetes mellitus pathogenesis, which requires further scrupulous research, analysis and generalization.

Study on the Method of Selecting Sustainable Food Suppliers Considering Interactive Factors

The existing sustainable supplier selection methods are not sufficient to deal with the problem of sustainable food supplier selection with the interaction of criteria under uncertainty. Therefore, this paper proposes a method of sustainable food supplier selection based on an extended decision model. Firstly, a processing method for supplier evaluation information is constructed using the Pythagorean fuzzy set has the function of processing complex uncertain information. Secondly, to obtain the objective weights of decision experts, a Pythagorean fuzzy weighted distance measure model is constructed, and an expert information fusion method based on a weighted power mean operator is proposed to construct the group decision matrix. Then, the decision experiment and evaluation experiment methods are integrated with the traditional MARCOS method, to construct a sustainable food supplier selection method considering the interaction of factors. This method can effectively deal with the complicated and uncertain problem of sustainable food supplier selection with interactive factors. Finally, the feasibility of the proposed method is verified by an example of sustainable food supplier selection. In addition, parameter sensitivity analysis and multi-method comparative analysis verified the rationality of the proposed selection method for sustainable food supplier selection.

Classification of mental workload using brain connectivity and machine learning on electroencephalogram data

Mental workload refers to the cognitive effort required to perform tasks, and it is an important factor in various fields, including system design, clinical medicine, and industrial applications. In this paper, we propose innovative methods to assess mental workload from EEG data that use effective brain connectivity for the purpose of extracting features, a hierarchical feature selection algorithm to select the most significant features, and finally machine learning models. We have used the Simultaneous Task EEG Workload (STEW) dataset, an open-access collection of raw EEG data from 48 subjects. We extracted brain-effective connectivities by the direct directed transfer function and then selected the top 30 connectivities for each standard frequency band. Then we applied three feature selection algorithms (forward feature selection, Relief-F, and minimum-redundancy-maximum-relevance) on the top 150 features from all frequencies. Finally, we applied sevenfold cross-validation on four machine learning models (support vector machine (SVM), linear discriminant analysis, random forest, and decision tree). The results revealed that SVM as the machine learning model and forward feature selection as the feature selection method work better than others and could classify the mental workload levels with accuracy equal to 89.53% (± 1.36).

Increasing Analytical Quality by Designing a Thin-Layer Chromatography Scanner Method for the Determination of the Radiochemical Purity of Radiopharmaceutical Sodium Iodide 131I Oral Solution.

The goal of this study was to apply the principles of analytical quality by design (AQbD) to the analytical method for determining the radiochemical purity (PQR) of the radiopharmaceutical sodium iodide 131I oral solution, utilizing thin-layer chromatography (TLC) with a radio-TLC scanner, which also enables the evaluation of product quality. For AQbD, the analytical target profile (ATP), critical quality attributes (CQA), risk management, and the method operable design region (MODR) were defined through response surface methodology to optimize the method using MINITAB® 19 software. This study encompassed the establishment of a control strategy and the validation of the method, including the assessment of selectivity, linearity, precision, robustness, detection limit, quantification limit, range, and the stability of the sample solution. Under the experimental conditions, the method parameters of the TLC scanner were experimentally demonstrated and optimized with an injection volume of 3 µL, a radioactive concentration of 10 mCi/mL, and a carrier volume of 40 µL. Statistical analysis confirmed the method's selectivity for the 131I iodide band Rf of 0.8, a radiochemical impurity IO3- Rf of 0.6, a linearity from 6.0 to 22.0 mCi/mL, and an intermediate precision with a global relative standard deviation (RSD) of 0.624%. The method also exhibited robustness, with a global RSD of 0.101%, a detection limit of 0.09 mCi/mL, and a quantification limit of 0.53 Ci/mL, meeting the prescribed range and displaying stability over time (at 0, 2, and 20 h) with a global RSD of 0.362%, resulting in consistent outcomes. The development of a method based on AQbD facilitated the creation of a design space and an operational space, with comprehensive knowledge of the method's characteristics and limitations. Additionally, throughout all operations, compliance with the acceptance criteria was verified. The method's validity was confirmed under the established conditions, making it suitable for use in the manufacturing process of sodium iodide 131I and application in nuclear medicine services.

Feature Selection Techniques for CR Isotope Identification with the AMS-02 Experiment in Space

Isotopic composition measurements of singly charged cosmic rays (CR) provide essential insights into CR transport in the Galaxy. The Alpha Magnetic Spectrometer (AMS-02) can identify singly charged isotopes up to about 10 GeV/n. However, their identification presents challenges due to the small abundance of CR deuterons compared to the proton background. In particular, a high accuracy for the velocity measured by a ring-imaging Cherenkov detector (RICH) is needed to achieve a good isotopic mass separation over a wide range of energies. The velocity measurement with the RICH is particularly challenging for Z=1 isotopes due to the low number of photons produced in the Cherenkov rings. This faint signal is easily disrupted by noisy hits leading to a misreconstruction of the particles’ ring. Hence, an efficient background reduction process is needed to ensure the quality of the reconstructed Cherenkov rings and provide a correct measurement of the particles’ velocity. Machine learning methods, particularly boosted decision trees, are well suited for this task, but their performance relies on the choice of the features needed for their training phase. While physics-driven feature selection methods based on the knowledge of the detector are often used, machine learning algorithms for automated feature selection can provide a helpful alternative that optimises the classification method’s performance. We compare five algorithms for selecting the feature samples for RICH background reduction, achieving the best results with the Random Forest method. We also test its performance against the physics-driven selection method, obtaining better results.

Performance Analysis of Random Forest Algorithm for Network Anomaly Detection using Feature Selection

As the volume and complexity of computer network traffic continue to increase, network administrators face a growing challenge in monitoring and discovering unusual activity. To keep the network safe and functioning, detecting anomalies is essential. Machine learning-based anomaly detection techniques have become increasingly popular in recent years. This is due to the fact that conventional anomaly detection methods make it difficult to detect unknown and complex attacks. This research aims to conduct a performance analysis of two feature selection methods using the random forest algorithm using the UNSW-NB15 dataset to determine which model is most effective in detecting network traffic anomalies. The models evaluated were random forest with the filter method and random forest with the wrapper method. A number of metrics used for model performance assessment are accuracy, F1-score, receiver operating characteristic curve, and precision-recall. Dataset collection, data pre-processing, feature selection, model construction, and evaluation are the main components of the research methodology. The research results show that the Random Forest approach with the Filter method has an accuracy of 0.8950, F1-score of 0.8333, ROC score of 0.8928, and a precision-recall value of 0.8347. Meanwhile, the approach using the Wrapper method obtained an accuracy of 0.9151, F1-score of 0.8510, ROC score of 0.9136, and a precision-recall value of 0.8637. This shows that the performance of Random Forest with the Wrapper method is superior in all assessment metrics. Random Forest with the Wrapper Method is the right choice of model for detecting network traffic anomalies because of its stable performance and ability to handle complex patterns

National-scale digital soil mapping performances are related to covariates and sampling density: Lessons from France

Accurate soil property and class predictions through spatial modelling necessitate a thoughtful selection of explanatory variables and sample size, as their choice greatly impacts model performance. Within the framework of Global Soil Nutrient and Nutrient Budget maps (GSNmap), the FAO Global Soil Partnership (GSP) launched a country-driven digital soil mapping (DSM) approach. The GSP asked the countries if they could implement the DSM prediction of ten soil properties, using their national point data and a set of widely available covariates (GSP_Cov). In this study, we examined the effect of including additional national-based covariates and soil observations on the performance of the prediction models using mainland France as a pilot. The learning soil dataset was based on a systematic 16-to-16 km grid. For a subset of soil properties, we also assessed using repeated k-fold cross-validation the effect of adding to this dataset many other irregularly spread measurements. The GSP_Cov included common widely available covariates that represented information about terrain, climate, and organisms. The second set of covariates consisted of the GSP_Cov, extended to extra covariates available at a national level, such as previously existing soil maps, geological maps, remote sensing products and others. Random Forest approach in combination with the Boruta selection method was employed for mapping ten soil properties: soil organic carbon (SOC), pH (water), total nitrogen (N), available phosphorus (P), available potassium (K), cation exchange capacity (CEC), bulk density (BD), and texture (clay, silt, and sand). The results revealed noteworthy enhancements in prediction performance for more than half of the properties, although, for some of them, the improvements were negligible. The most significant improvements were obtained for pH, CEC and texture, where geological variables and a previous pH map significantly contributed to the increase in accuracy. Adding numerous points (around 25,000) to the learning dataset improved the performance of soil particle-size fractions predictions. By broadening the spectrum of covariates and better covering the feature and geographical spaces considered in soil prediction models, this research underscores the importance of implementing a more diverse range of covariates at a national scale and of densifying soil information to enlarge the feature and geographical spaces of multidimensional soil/covariates combinations. This information should be taken into account in national and continental digital soil mapping endeavours.