Classification Rate Research Articles

On the Comparative Study of the Supervised Machine Learning Models

This research empirically compared the performance of three supervised machine learning models which are Multinomial Logistic Regression (MLR), Multilayer back propagated Neural Networks (MNN) and Multinomial Decision Trees (MDT) based on a tested data set for the opinion of Nigerian citizens during the period of naira redesign policy using classification matrix criterion. About 600 copies of questionnaires on the opinion of Nigerian citizens on their welfare at the period of naira redesign. The result showed that ANN outperformed other models with 94.4% correct classification rates, followed by the MLR with correct classification rates of 93.5% and lastly by MDT with correct classification rates of 90.0%.

Assessment of fuzzy logic to enhance species distribution modelling of two cryptic wood boring beetle species in Australia.

Fuzzy logic presents a promising approach for Species Distribution Modelling by generating a value that can be used for comparative purposes termed 'environmental favourability'. In contrast to 'presence probability', 'environmental favourability' remains robust regardless of species prevalence. This characteristic facilitates effective comparisons across species with varying levels of prevalence. In this study, presence probability was predicted using three commonly used Species Distribution Models: Generalised Linear Model, Generalised Additive Modelling, and Boosted Regression Trees for two beetle species, Euwallacea fornicatus and Euwallacea perbrevis in Australia. Fuzzy logic was then employed to derive environmental favourability values based on these models. Additionally, Maxent modelling was included to compare prediction outputs and facilitate a comprehensive analysis. Model performance was evaluated using standard metrics (Area under the receiver operating characteristic curve, True statistical skill, Correct classification rate), as well as Hosmer-Lemeshow test. The research explored fuzzy similarity, fuzzy intersection and potential biotic interaction of these closely related borers, and revealed a favourable distribution pattern for Euwallacea fornicatus across Australia. This study supports the efficacy of fuzzy logic in Species Distribution Modelling and highlights the value of environmental favourability function in enhancing the comparative analysis of the geographical relationship across species. This approach offers a more nuanced perspective on Species Distribution Modelling.

Domain‐independent adaptive histogram‐based features for pomegranate fruit and leaf diseases classification

AbstractDisease identification for fruits and leaves in the field of agriculture is important for estimating production, crop yield, and earnings for farmers. In the specific case of pomegranates, this is challenging because of the wide range of possible diseases and their effects on the plant and the crop. This study presents an adaptive histogram‐based method for solving this problem. Our method describe is domain independent in the sense that it can be easily and efficiently adapted to other similar smart agriculture tasks. The approach explores colour spaces, namely, Red, Green, and Blue along with Grey. The histograms of colour spaces and grey space are analysed based on the notion that as the disease changes, the colour also changes. The proximity between the histograms of grey images with individual colour spaces is estimated to find the closeness of images. Since the grey image is the average of colour spaces (R, G, and B), it can be considered a reference image. For estimating the distance between grey and colour spaces, the proposed approach uses a Chi‐Square distance measure. Further, the method uses an Artificial Neural Network for classification. The effectiveness of our approach is demonstrated by testing on a dataset of fruit and leaf images affected by different diseases. The results show that the method outperforms existing techniques in terms of average classification rate.

Confronting the idealised ‘Nordic model’ in education with contemporary realities of special education in Sweden

ABSTRACT For decades, the Nordic countries have been admired for their inclusive education systems within egalitarian societies, bolstered by PISA studies and global calls to ensure inclusive education as a human right. This paper contrasts idealised perspectives on Nordic education with internal experiences, exploring recent challenges to the relationship between special and inclusive education. Using Sweden as a case study, it highlights internal diversity within the Nordic region, showing differing paths and gaps between research, policy and school realities. The risks of idealising past educational models are discussed, noting that Sweden, once known for its egalitarian education and low special education needs classification rates, now faces higher classification rates and more separate provision due to policy changes. Recent trends have shown a shift towards increased diagnosis and classification, influenced by marketisation and decentralisation policies. The paper focuses on neurodivergence, analysing shifts in government priorities, teacher education, school practices and research agendas. Analysis of recent policies, reports and educational data reveals winners and losers in school participation, providing insights into the implications for educational discourse, the future of the ‘Nordic model’, and Swedish education in a global context.

MACHINE LEARNING, DEEP LEARNING AND ENSEMBLE LEARNING BASED APPROACHES FOR INTRUSION DETECTION ENHANCEMENT

With the extensive application of Internet of Things and wireless sensor networks (WSNs) in various real-time fields, they have become increasingly vulnerable to several types of attacks that could have a serious impact in their functionalities. Recently, intrusion detection systems have become one of the crucial security components. We propose in this work an approach based on machine and deep learning for DoS and DDoS attack detection. We have evaluated, analyzed, and compared the efficiency of three learning models separately, including deep neural network (DNN), random forest, and decision tree, using the standard metrics of evaluation such as accuracy, precision, F1-score, and recall. In our contribution, an approach ensemble learning was introduced in order to enhance the accuracy rate of classification. Our study was carried out using two well-known benchmark and real-time datasets, CICIoT-2023 and WSN-DS, intended for wireless sensor networks and IoT. These datasets containing various types of attacks. CICIoT2023 dataset contains 33 attacks divided into 7 classes, while WSN-DS dataset contains four types of DoS attacks, including Blackhole, Grayhole, Flooding, and TDMA. The experiment result demonstrate the effectiveness of our approach in attacks detection with high accuracy achieved close to perfect. Our approach result demonstrates that using the ensemble learning technique works better than each ML architecture independently.

Rapid bacterial identification through volatile organic compound analysis and deep learning

BackgroundThe increasing antimicrobial resistance caused by the improper use of antibiotics poses a significant challenge to humanity. Rapid and accurate identification of microbial species in clinical settings is crucial for precise medication and reducing the development of antimicrobial resistance. This study aimed to explore a method for automatic identification of bacteria using Volatile Organic Compounds (VOCs) analysis and deep learning algorithms.ResultsAlexNet, where augmentation is applied, produces the best results. The average accuracy rate for single bacterial culture classification reached 99.24% using cross-validation, and the accuracy rates for identifying the three bacteria in randomly mixed cultures were SA:98.6%, EC:98.58% and PA:98.99%, respectively.ConclusionThis work provides a new approach to quickly identify bacterial microorganisms. Using this method can automatically identify bacteria in GC-IMS detection results, helping clinical doctors quickly detect bacterial species, accurately prescribe medication, thereby controlling epidemics, and minimizing the negative impact of bacterial resistance on society.

Crop Leaf Type Classification and Multi-Class Leaf Disease Identification with Tangent Hunter Prey Optimization-Based LeNet

The early detection and accurate rate of classification of plant leaf disease are more important for reliable and good agriculture, which prevents unwanted financial loss and resources. In this study, Tangent Hunter Prey Optimization-based LeNet (THPO-LeNet) is utilized for crop leaf classification and multi-class plant leaf disease identification. In this case, the input image that goes through the image pre-processing step is obtained from the plant village dataset. An adaptive Wiener filter is applied at the pre-processing stage to reduce needless mistakes and improve image quality. The pre-processed image is then sent to the leaf segmentation step, where a Mask Region-based Convolution Neural Network (Mask R-CNN) performs the segmentation. Consequently, image augmentation is performed using techniques such as scaling, rotation, translation, flipping, contrast, saturation, and hue. Afterwards, first-level classification plant leaf classification is processed by LeNet, which is optimized by Tangent Hunter Prey Optimization (THPO). The THPO is the incorporation of a Tangent Search Algorithm (TSA) and Hunter–Prey Optimizer (HPO). At last, the second-level classification of plant leaf disease is conducted by THPO-LeNet. Furthermore, the efficiency of THPO-LeNet is examined based on measures, like accuracy, Negative Predictive Value (NPV), True Positive Rate (TPR), True Negative Rate (TNR), Positive Predictive Value (PPV), loss value, and False Positive Rate (FPR), and the value attained is 95.68%, 92.50%, 91.48%, 92.25%, 92.54%, 8.321%, and 7.754%, respectively.

An Analysis of SRSS-IE Externalizing Cut Scores to Facilitate Data-Informed Decision-Making in K–12 Schools

We report the results of a convergent validity study examining the externalizing subscale (SRSS-E5, five items) of the adapted Student Risk Screening Scale for Internalizing and Externalizing (SRSS-IE 9) with the externalizing subscale of the Teacher Report Form (TRF) with two samples of K–12 students. Results of logistic regression and receiver operating characteristic curves analyzed with a sample of 195 K–5 students from 13 schools across three states suggested the following preliminary cutting scores for the SRSS-E5 for use at the elementary level: 0 to 1 for low risk, 2 to 3 for moderate risk, and 4 to 15 for high risk for externalizing behaviors. Results of the same analyses with a sample of 227 students in Grades 6 to 12 suggested the following preliminary cutting scores for the SRSS-E5 for use in middle and high (secondary) schools: 0 to 4 for low risk, 5 to 6 for moderate risk, and 7 to 15 for high risk for externalizing behaviors. For both samples, sensitivity and specificity levels were acceptable, as was the correct classification rate. We discuss findings and limitations and emphasize the importance of replication before shifting screening practices from the SRSS-IE 12 (12 items) to the newly adapted SRSS-IE 9 (9 items).

Evaluating MIR and NIR Spectroscopy Coupled with Multivariate Analysis for Detection and Quantification of Additives in Tobacco Products.

The detection and quantification of additives in tobacco products are critical for ensuring consumer safety and compliance with regulatory standards. Traditional analytical techniques, like gas chromatography-mass spectrometry (GC-MS), liquid chromatography-mass spectrometry (LC-MS), and others, although effective, suffer from drawbacks, including complex sample preparation, high costs, lengthy analysis times, and the requirement for skilled operators. This study addresses these challenges by evaluating the efficacy of mid-infrared (MIR) spectroscopy and near-IR (NIR) spectroscopy, coupled with multivariate analysis, as potential solutions for the detection and quantification of additives in tobacco products. So, a representative set of tobacco products was selected and spiked with the targeted additives, namely caffeine, menthol, glycerol, and cocoa. Multivariate analysis of MIR and NIR spectra consisted of principal component analysis (PCA), hierarchical clustering analysis (HCA), partial least squares-discriminant analysis (PLS-DA) and soft independent modeling of class analogy (SIMCA) to classify samples based on targeted additives. Based on the unsupervised techniques (PCA and HCA), a distinction could be made between spiked and non-spiked samples for all four targeted additives based on both MIR and NIR spectral data. During supervised analysis, SIMCA achieved 87-100% classification accuracy for the different additives and for both spectroscopic techniques. PLS-DA models showed classification rates of 80% to 100%, also demonstrating robust performance. Regression studies, using PLS, showed that it is possible to effectively estimate the concentration levels of the targeted molecules. The results also highlight the necessity of optimizing data pretreatment for accurate quantification of the target additives. Overall, NIR spectroscopy combined with SIMCA provided the most accurate and robust classification models for all target molecules, indicating that it is the most effective single technique for this type of analysis. MIR, on the other hand, showed the overall best performance for quantitative estimation.

A Neural Network Approach to Understanding Employee Retention Dynamics: Insights from Feature Importance Analysis

In today’s business environment, employee retention has become a significant challenge, as employee turnover can lead to decreased productivity, increased costs, and reduced morale. This study aims to leverage neural network technology to predict employees' retention intentions and conduct in-depth analyses based on employee data. A classification model was established considering various factors, including gender, marital status, number of children, education level, years of service, weekly working hours, career development opportunities, salary, and bonuses, to identify the potential risk of employee turnover. The results indicated that the model achieved an accurate classification rate of 95.12% on the test set, demonstrating high effectiveness in identifying employee retention intentions. Feature importance analysis revealed that education level (29.63%), bonuses (27.50%), and the number of children (21.81%) were the primary factors influencing retention decisions. Additionally, working hours, marital status, and career development opportunities also impacted employees’ retention intentions. This research not only provides insights into employee mobility but also offers data support for enterprises to develop effective retention strategies, suggesting that businesses prioritize enhancing education and training opportunities, designing competitive bonus systems, and providing flexible work arrangements and benefits tailored to employees with children.

Authenticating the Geographical Origin of Jingbai Pear in Northern China by Multiple Stable Isotope and Elemental Analysis.

The Jingbai pear is one of the best pear species in China with high quality and nutrition values which are closely linked to its geographical origin. With the purpose of discriminating the PGI Mentougou Jingbai pear from three other producing regions, the stable isotope ratios and elemental profiles of the pears (n = 52) and the corresponding soils and groundwater were determined using isotope ratio mass spectrometry (IRMS) and inductively coupled plasma mass spectrometry (ICP-MS), respectively. The results revealed that δ15N, δ18OJ, and Li were significantly different (p < 0.05) in samples from different regions, which indicated their potential to be used in the geographical origin classification of the Jingbai pear. The nitrogen isotopic values of the pear pulp were positively correlated with the δ15N value and nitrogen content of the corresponding soils, whilst the B, Na, K, Cr, and Cd contents of the pear pulps were positively correlated with their corresponding soils. Orthogonal partial least squares discriminant analysis (OPLS-DA) was performed in combination with analysis of the stable isotopes and elemental profiles, making it possible to distinguish the cultivation regions from each other with a high prediction accuracy (a correct classification rate of 92.3%). The results of this study highlight the potential of stable isotope ratios and elemental profiles to trace the geographical origin of pears at a small spatial scale.

A Multi-Isotopic Chemometric Approach for Tracing Hazelnut Origins.

High-value products, such as hazelnuts, are particularly vulnerable to fraud due to their price dependence on geographical origin. Guaranteeing hazelnuts' authenticity is essential for consumer trust and safety. Stable isotope analysis has become a reference method for origin authentication as it is reliable, robust, and easily transferable across laboratories. However, multiple isotopic markers coupled with chemometric techniques are often needed to authenticate food provenance accurately. In this study, we focused on assessing the potential of bulk δ18O, along with δ2H and δ13C of the main fatty acids, as hazelnut-origin authenticity markers. PLS-DA classification models were developed to differentiate samples (n = 207) according to their region of origin. This multi-isotopic approach provided promising external validation results, achieving a 94% global correct classification rate in discriminating hazelnuts from regions with distinct geographical and environmental conditions. This study lays the groundwork for further model development and evaluation across additional production areas and harvest years.

Improving Automatic Coronary Stenosis Classification Using a Hybrid Metaheuristic with Diversity Control.

This study proposes a novel Hybrid Metaheuristic with explicit diversity control, aimed at finding an optimal feature subset by thoroughly exploring the search space to prevent premature convergence. Background/Objectives: Unlike traditional evolutionary computing techniques, which only consider the best individuals in a population, the proposed strategy also considers the worst individuals under certain conditions. In consequence, feature selection frequencies tend to be more uniform, decreasing the probability of premature convergent results and local-optima solutions. Methods: An image database containing 608 images, evenly balanced between positive and negative coronary stenosis cases, was used for experiments. A total of 473 features, including intensity, texture, and morphological types, were extracted from the image bank. A Support Vector Machine was employed to classify positive and negative stenosis cases, with Accuracy and the Jaccard Coefficient used as performance metrics. Results: The proposed strategy achieved a classification rate of 0.92 for Accuracy and 0.85 for the Jaccard Coefficient, obtaining a subset of 16 features, which represents a discrimination rate of 0.97 from the 473 initial features. Conclusions: The Hybrid Metaheuristic with explicit diversity control improved the classification performance of coronary stenosis cases compared to previous literature. Based on the achieved results, the identified feature subset demonstrates potential for use in clinical practice, particularly in decision-support information systems.

An AutoEncoder enhanced light gradient boosting machine method for credit card fraud detection

Online financial transactions bring convenience to people’s lives, but also present vulnerabilities for criminals to embezzle users’ accounts and trick users into credit card fraud. Although machine learning methods have been adopted to detect anomalous transactions, it’s hard for a single machine learning method to achieve satisfying results with the increasing scale and dimensionality of financial datasets. In addition, for anomaly detection of financial data, there is an obvious imbalance between normal records and abnormal. In this situation, the experimental results cannot be objectively evaluated only by the traditional metrics, such as precision, recall, and accuracy. This paper proposes an AutoEncoder enhanced LightGBM method for credit card detection. The method inherits the advantages of each component, using an AutoEncoder for feature reconstruction on the dataset, and integrating the LightGBM algorithm for improving the GBDT (Gradient Boosting Decison Tree) to detect abnormal data more accurately and efficiently. Besides the traditional evaluation metrics, F-measure, area under curve (AUC), Matthew’s correlation coefficient (MCC), and balanced classification rate (BCR) are also adopted as the evaluation metrics. Two financial datasets were used to validate the performance and robustness of the proposed model. Results obtained from the credit card fraud dataset containing 31 features indicate that our model significantly outperforms other models with a recall of 94.85%, representing a 10.70% improvement compared to the best detection performance model with a recall of only 86%. Additionally, our model’s BCR score is also significantly better than other models, with a BCR score of 97%, as opposed to the best detection performance model’s BCR score of 92%, representing a 5% improvement by our model. Various sampling methods and model combinations were considered in this study. It was found that the SMOTE algorithm combined with the proposed model produced the best results, with an AUC value of 96.83% and an F-measure score of 80.27%. The Santander bank transaction record dataset is a high dimensional large dataset containing 200 features. Experimental results on this dataset reveal that compared to other models, our model significantly improved recall and F-measure results, raising the recall to 94.14% and the F-measure score by 11.51%, surpassing the second-best-performing model. Overall, these findings demonstrate the robustness and superiority of our model in detecting fraudulent transactions and highlight the effectiveness of the SMOTE algorithm in combination with the proposed model.

Discriminant analysis using feature extraction from spectral domain responses to achieve accurate delineation for robust evaluation or classification of soil properties

ABSTRACT Soil spectroscopy offers a method for quantitatively analysing soil chromophores and employing a screening approach. However, utilizing general models with this data at once doesn’t consistently yield accurate results in determining soil parameters. Ambiguities in response within certain domains of the reflectance spectrum are observed. To address this issue, a method is proposed where quantities associated with spectral response are outputted within well-defined domains corresponding to specific parameters. This article introduces a novel approach centred on identifying features from reflectance spectra to assess soil properties and mitigate issues arising from a heterogeneous soil dataset. The focus is on swiftly automating the classification of soil spectral data by extracting features directly related to the varied composition of chromophores in the samples. These features are categorized for each chromophore parameter and analysed using linear discriminant analysis (LDA). The study employs the ‘309 soil samples’ from the GEO-CRADLE open spectral soil library in Greece. Specifically, relevant data from six spectral responses are extracted to separate and quantify certain soil parameters such as clay content, NO3− levels, and SOM (Soil Organic Matter), despite their interrelations. We developed a method to visually represent the determination outcomes quantitatively by establishing three threshold parameter values: 1.6% for SOM, 19% for clay, and 13.6 ppm for NO3−. These determinations will span the explored ranges of ’0.3 to 4.18%’ for SOM, “3 to 48%” for clay, and ’0 to 661.2 ppm’ for NO3−. Utilizing supervised LDA proves notable classification rates of 93.3% for the SOM task, 90% for the Clay task, and 86.6% for NO3− with the specific cut-off value parameter. The final result is presented as a map showing the positions of various prediction samples, highlighting the percentages of clay and organic matter relative to 19% and 1.6%, respectively, as well as the nitrate concentration in these samples relative to 13.6 ppm.

Integrating Whole Slide Imaging and Artificial Intelligence Provides Higher Accuracy in Thyroid Cytology

Abstract Introduction/Objective The aim of the study is to evaluate the potential advantages of integrating Whole Slide Imaging (WSI) and Artificial Intelligence (AI) to increase efficiency and diagnostic accuracy in thyroid cytology, especially in indeterminate lesions. Methods/Case Report A retrospective database search from 2000-2023 was performed, and the concordance with the pathologists was evaluated based on the manual or AI-based Region of Interest (ROI) identification, the best- suited AI, algorithm design, color enhancement, and combination of AI and Electronic Medical Records (EMR). Results (if a Case Study enter NA) The cohort included 3285 WSI and 121,384 ROI. In 109 WSI with manually diagnosed ROI, Deep learning successfully predicted benign and malignant cases, including 11 indeterminate cases, with zero false positives or negatives for 35% of cases. In the second study with 908 WSI, 4494 ROI were identified with Convolutional Neural Network (CNN). A higher area under the curve (AUC) resulted when EMR and Machine Learning algorithm (MLA) were integrated: EMR: 0.931, MLA: 0932, EMR+ MLA; 0.962, Sensitivity: 92.0%, Specificity 90.5%. In the next study with 145 WSI, the CNN-identified ROI was presented to the same pathologist after 117 days. Perfect concordance was seen between the two methods (K score 0.924, time to diagnosis;81.59 Second). In 360 WSI, YOLOV4 algorithm was trained to detect the ROI, and detection and classification rates were used to evaluate the accuracy. It had a precision of 81.84% (3.16% more than the single detection networks of 78.68%). In the last study with 964 WSI, color augmentation improved the AUC for indeterminate cases from 88.3% to 96.2%. Conclusion CNN is a better-suited AI for whole slide imaging due to its compatibility with image data and ability to automatically detect important features without human supervision. It also removes the component of subjective bias in interpretation. Color augmentation and integration of EMR can overcome the unique challenges in thyroid indeterminate cytology.

Effectiveness of Generative AI for Post-Earthquake Damage Assessment

After an earthquake, rapid assessment of building damage is crucial for emergency response, reconstruction planning, and public safety. This study evaluates the performance of various Generative Artificial Intelligence (GAI) models in analyzing post-earthquake images to classify structural damage according to the EMS-98 scale, ranging from minor damage to total destruction. Correct classification rates for masonry buildings varied from 28.6% to 64.3%, with mean damage grade errors between 0.50 and 0.79, while for reinforced concrete buildings, rates ranged from 37.5% to 75.0%, with errors between 0.50 and 0.88. Fine-tuning these models could substantially improve accuracy. The practical implications are significant: integrating accurate GAI models into disaster response protocols can drastically reduce the time and resources required for damage assessment compared to traditional methods. This acceleration enables emergency services to make faster, data-driven decisions, optimize resource allocation, and potentially save lives. Furthermore, the widespread adoption of GAI models can enhance resilience planning by providing valuable data for future infrastructure improvements. The results of this work demonstrate the promise of GAI models for rapid, automated, and precise damage evaluation, underscoring their potential as invaluable tools for engineers, policymakers, and emergency responders in post-earthquake scenarios.

Identifying the Vertical Stratification of Sediment Samples by Visible and Near-Infrared Spectroscopy

Vertical stratification in marine sediment profiles indicates physical and chemical sedimentary processes and, thus, is the first step in sedimentary research and in studying their relationship with global climate change. Traditional technologies for studying vertical stratification have low efficiency; thus, new technologies are highly needed. Recently, visible and near-infrared spectroscopy (VNIR) has been explored to rapidly determine sediment parameters, such as clay content, particle size, total carbon (TC), total nitrogen (TN), and so on. Here, we explored vertical stratification in a sediment column in the South China Sea using VNIR. The sediment column was 160 cm and divided into 160 samples by 1 cm intervals. All samples were classified into three layers by depth, that is, 0–50 cm (the upper layer), 50–100 cm (the middle layer), and 100–160 cm (the bottom layer). Concentrations of TC and TN in each sample were measured by Elementa Vario EL III. Visible and near-infrared reflectance spectra of each sample were collected by Agilent Cary 5000. A global model and several classification models for vertical stratification in sediments were established by a Support Vector Machine (SVM) after the characteristic spectra were identified using Competitive Adaptive Reweighted Sampling. In the classification models, K-means clustering and Density Peak Clustering (DPC) were employed as the unsupervised clustering algorithms. The results showed that the stratification was successful by VNIR, especially when using the combination of unsupervised clustering and machine learning algorithms. The correct classification rate (CCR) was much higher in the classification models than in the global model. And the classification models had a higher CCR using K-means combined with SVM (94.8%) and using DPC combined with SVM (96.0%). The higher CCR might be derived from the chemical classification. Indeed, similar results were also found in the chemical stratification. This study provided a theoretical basis for the rapid and synchronous measurement of chemical and physical parameters in sediment profiles by VNIR.

Multivariate Modelling Based on Isotopic, Elemental, and Fatty Acid Profiles to Distinguish the Backyard and Barn Eggs.

The ability to trace the origin of eggs from backyard-raised hens is important due to their higher market value compared to barn-raised eggs. This study aimed to differentiate eggs from these two rearing systems using isotopic, elemental, and fatty acid profiles of egg yolks. A total of 90 egg yolk samples were analyzed, analytical results being followed by statistical tests (Student's t-test) showing significant differences in δ18O, several elements (Mg, K, Sc, Mn, Fe, Ni, Cu, Zn, As, Cd, Ba, Pb), and fatty acids compositions (C23:0, C17:0, C18:0, C16:1n7, C18:1n9, C18:2n6, C20:1n7, C20:4n6, C20:5n3, C22:6n3), as well as in the ratios of SFA, PUFA, and UFA. The results indicated a nutritional advantage in backyard eggs due to their lower n-6 polyunsaturated fatty acid content and a more favorable n-6 to n-3 ratio, linked to differences in the hens' diet and rearing systems. To classify the production system (backyard vs. barn), three pattern recognition methods were applied: linear discriminant analysis (LDA), k-nearest neighbor (k-NN), and multilayer perceptron artificial neural networks (MLP-ANN). LDA provided perfect initial separation, achieving 98.9% accuracy in cross-validation. k-NN yielded classification rates of 98.4% for the training set and 85.7% for the test set, while MLP-ANN achieved 100% accuracy in training and 92.3% in testing, with minor misclassification. These results demonstrate the effectiveness of fusion among isotopic, elemental, and fatty acid profiles in distinguishing backyard eggs from barn eggs and highlight the nutritional benefits of the backyard-rearing system.

Adaptive Cybersecurity Neural Networks: An Evolutionary Approach for Enhanced Attack Detection and Classification

The increasing sophistication and frequency of cyber threats necessitate the development of advanced techniques for detecting and mitigating attacks. This paper introduces a novel cybersecurity-focused Multi-Layer Perceptron (MLP) trainer that utilizes evolutionary computation methods, specifically tailored to improve the training process of neural networks in the cybersecurity domain. The proposed trainer dynamically optimizes the MLP’s weights and biases, enhancing its accuracy and robustness in defending against various attack vectors. To evaluate its effectiveness, the trainer was tested on five widely recognized security-related datasets: NSL-KDD, CICIDS2017, UNSW-NB15, Bot-IoT, and CSE-CIC-IDS2018. Its performance was compared with several state-of-the-art optimization algorithms, including Cybersecurity Chimp, CPO, ROA, WOA, MFO, WSO, SHIO, ZOA, DOA, and HHO. The results demonstrated that the proposed trainer consistently outperformed the other algorithms, achieving the lowest Mean Square Error (MSE) and highest classification accuracy across all datasets. Notably, the trainer reached a classification rate of 99.5% on the Bot-IoT dataset and 98.8% on the CSE-CIC-IDS2018 dataset, underscoring its effectiveness in detecting and classifying diverse cyber threats.