Multilayer Perceptron Classifier Research Articles

Speech Emotion Recognition using Machine Learning With Real-time Audio Analysis

This research paper presents a Speech Emotion Recognition (SER) system utilizing a Multilayer Perceptron (MLP) classifier and real-time audio analysis. The system records audio samples, extracts relevant features, and employs machine learning techniques to predict emotions in spoken language. The study focuses on the development of an intuitive Graphical User Interface (GUI) using the Kivy framework, providing a user-friendly platform for real-time emotion analysis. In shortly In this project, we attempt to detect underlying emotions such as (sad, happy, neutral, angry, disgust, surprised, fearful and calm) in recorded speech by analysing the acoustic features of the audio data of recordings and Created an application to implement the same on user input

An ensemble learning-based framework for breast cancer prediction

Breast cancer is a disease characterised by abnormal cell growth in the breast, with various types and unique characteristics depending on the malignant cells involved. It predominantly affects women, and researchers in the field of clinical sciences are increasingly interested in leveraging Artificial Intelligence (AI) techniques to predict and diagnose breast cancer. This study utilises the widely-used Wisconsin Breast Cancer Dataset (WBCD) from the University of California Irvine (UCI) machine learning repository. The dataset contains 30 features, including mean, standard error, and “worst” values, providing insights into different aspects of breast cancer. A stacked ensemble classifier is proposed to evaluate the effectiveness of the proposed model, incorporating multiple machine learning algorithms such as the Decision Tree Classifier, AdaBoost Classifier, Gaussian Naive Bayes (GaussianNB), and Multi-layer Perceptron Classifier (MLPClassifier). Various performance measures, including the Receiver Operating Characteristic Curve (ROC curve), Area Under the Curve (AUC), specificity, F1-score, sensitivity, and accuracy, are used to assess the model’s performance. The results demonstrate that the proposed ensemble technique achieves an accuracy of 97.66%, surpassing the performance reported in existing literature. This highlights the superior performance of the developed model compared to previous approaches. The study underscores the potential of AI techniques and machine learning algorithms in advancing breast cancer prediction and diagnosis, offering promising avenues for improving clinical decision-making and patient outcomes.

Enhanced focal cortical dysplasia detection in pediatric frontal lobe epilepsy with asymmetric radiomic and morphological features.

Focal cortical dysplasia (FCD) is the most common pathological cause for pediatric epilepsy, with frontal lobe epilepsy (FLE) being the most prevalent in the pediatric population. We attempted to utilize radiomic and morphological methods on MRI and PET to detect FCD in children with FLE. Thirty-seven children with FLE and 20 controls were included in the primary cohort, and a five-fold cross-validation was performed. In addition, we validated the performance in an independent site of 12 FLE children. A two-stage experiments including frontal lobe and subregions were employed to detect the lesion area of FCD, incorporating the asymmetric feature between the left and right hemispheres. Specifically, for the radiomics approach, we used gray matter (GM), white matter (WM), GM and WM, and the gray-white matter boundary regions of interest to extract features. Then, we employed a Multi-Layer Perceptron classifier to achieve FCD lesion localization based on both radiomic and morphological methods. The Multi-Layer Perceptron model based on the asymmetric feature exhibited excellent performance both in the frontal lobe and subregions. In the primary cohort and independent site, the radiomics analysis with GM and WM asymmetric features had the highest sensitivity (89.2 and 91.7%) and AUC (98.9 and 99.3%) in frontal lobe. While in the subregions, the GM asymmetric features had the highest sensitivity (85.6 and 79.7%). Furthermore, relying on the highest sensitivity of GM and WM asymmetric features in frontal lobe, when integrated with the subregions results, our approach exhibited overlaps with GM asymmetric features (55.4 and 52.4%), as well as morphological asymmetric features (54.4 and 53.8%), both in the primary cohort and at the independent site. This study demonstrates that a two-stage design based on the asymmetry of radiomic and morphological features can improve FCD detection. Specifically, incorporating regions of interest for GM, WM, GM, and WM, and the gray-white matter boundary significantly enhances the localization capabilities for lesion detection within the radiomics approach.

Mental Stress Classification from Brain Signals using MLP Classifier

INTRODUCTION: The most common and widespread mental condition that unavoidably affects people's mood and conduct is stress. The physiological reaction to powerful emotional, intellectual, and physical obstacles might be viewed as stress. As a result, early stress detection can result in solutions for potential improvements and ultimate event suppression. OBJECTIVES: To classify mental stress from the EEG signals of humans using an MLP classifier. METHODS: We examine the EEG signal analysis techniques currently in use for detecting mental stress using Multi-layer Perceptron (MLP). RESULTS: The suggested technique has a 95% classification accuracy performance. CONCLUSION: In our study, the use of MLP classifiers for stress detection from EEG signals has shown promising results. The high accuracy and precision of the classifiers, as well as the informative nature of certain EEG frequency bands, suggest that this approach could be a valuable tool for stress detection and management.

PSIV-23 Activity Classification of Grazing Cows on Desert Rangeland in the Southwestern United States

Abstract Monitoring cattle on rangelands is a daunting task that can be improved by using wearable sensors that are capable of transmitting motion and position data in real time and at low cost. This study tested the performance of machine learning (ML) classifiers to discriminate among foraging activities of cows based on triaxial accelerometer data collected in real-time by LoRa WAN networks. Trials were conducted at the New Mexico State University Chihuahuan Desert Rangeland Research Center and the USDA Jornada Experimental Range in Doña Ana County, NM. A total of 24 Brangus, Brahman, Raramuri Criollo and Angus x Hereford mature cows fitted with LoRa WAN tracking collars housing GPS and triaxial accelerometers were monitored across four periods during the 2022 summer and fall seasons on desert rangeland pastures. Trackers integrated and transmitted activity count (Ac) data from accelerometers at one-minute intervals. Video recording of focal cows (n = 24) was undertaken during daylight hours (0630 to 2000 h) from a distance of ~30 m to minimize interference with natural behaviors. A total of 168 hours of video were recorded and inspected by an experienced observer to label video files according to a classification tree of four main activities: grazing (GR), walking (WA), resting (RE) and ruminating (RU), and two states: active (AC) or static (ST). Individualized activities and states were considered when cows performed the same predefined activity or state for more than 30 secs. Retrieved sensor data from collar trackers were labeled by state and activity according to labels collected from video records. This classification resulted in a dataset containing 9,222 events, including 3,928 for GR, 2,286 for WA, 2,032 for RE, and 976 for RU, as well as 6,214 labels for AC and 3,008 labels for ST. Deep learning through Multilayer Perceptron Classifiers (MLPC) were coded and implemented using a split configuration of 70% of the data for training and 30% for testing, respectively. In preliminary runs, models had reduced ability to properly discriminate among RE (F1 = 0.42) and RU (F1 = 0.43) Thus, RE and RU were merged on subsequent tests, resulting in 3,928 labels for GR, 2,286 labels for WA, and 3,008 labels for merged RE. Deep learning models successfully classified between AC vs. ST behavior with an overall F1 performance score of 0.96. Further use of the same deep learning models successfully classified among GR, WA, and RE activities with an overall F1 performance score of 0.91, suggesting satisfactory application of the trained models to monitor cattle grazing activities on desert rangeland.

Temporal and Spectral Analysis of EMG for Classification of Muscular Paralysis

Loss of muscle function is the condition referred to as paralysis. Parts of the body may be completely paralysed or only partially. The quality of life is enhanced by the early identification of paralysis. In people with paralysis and neuromuscular diseases, EMG signals can be used to analyse muscular activation. In this study, EMG signals are analysed by feature extraction and divided into two categories: normal and paralysed. The obtained findings demonstrate that the extracted features in the suggested work perform better for EMG signal categorization. The conditions of Amyotrophic Lateral Sclerosis (ALS) and Myopathy are taken into consideration in this study to examine the paralysis state. Using time and frequency domain approaches, characteristics were retrieved from the EMG of healthy and paralysed participants. Multilayer Perceptron (MLP), Support Vector Machine (SVM), Random Forest (RF), Gradient Boosting (XGBOOST), and K-Nearest Neighbour (KNN) Classifiers are the classifier models used in the study. With time domain EMG information, classifiers like MLP, SVM, RF, XGBOOST, and KNN are used. The frequency domain EMG characteristics are applied to the MLP, RF, XGBOOST, and KNN classifiers. With time domain EMG features, MLP achieved a classification accuracy of 76.5%, SVM with 77.2%, RF with 76.1%, XGBOOST with 77.1%, and KNN with 75.8%. In comparison to classifier models employing time domain EMG information, the SVM classifier performs better. The classification accuracy for MLP, RF, XGBOOST, and KNN using frequency domain EMG features is 77.7%, 76.6%, and 75%, respectively. In comparison to other classifier models with frequency domain features, the MLP and RF classifiers perform better. Time and frequency domains of the EMG of Normal, ALS, and Myopathy diseases are investigated. It has been noted that the EMG signal and its characteristics differ significantly (p<0.05) between the Normal and Paralysis conditions. EMG is utilised in the current study to analyse and categorise paralysis, which helps with early diagnosis and improved treatment options.

Multi-Layer Perceptron Classifier with the Proposed Combined Feature Vector of 3D CNN Features and Lung Radiomics Features for COPD Stage Classification.

Computed tomography (CT) has been regarded as the most effective modality for characterizing and quantifying chronic obstructive pulmonary disease (COPD). Therefore, chest CT images should provide more information for COPD diagnosis, such as COPD stage classification. This paper proposes a features combination strategy by concatenating three-dimension (3D) CNN features and lung radiomics features for COPD stage classification based on the multi-layer perceptron (MLP) classifier. First, 465 sets of chest HRCT images are automatically segmented by a trained ResU-Net, obtaining the lung images with the Hounsfield unit. Second, the 3D CNN features are extracted from the lung region images based on a truncated transfer learning strategy. Then, the lung radiomics features are extracted from the lung region images by PyRadiomics. Third, the MLP classifier with the best classification performance is determined by the 3D CNN features and the lung radiomics features. Finally, the proposed combined feature vector is used to improve the MLP classifier's performance. The results show that compared with CNN models and other ML classifiers, the MLP classifier with the best classification performance is determined. The MLP classifier with the proposed combined feature vector has achieved accuracy, mean precision, mean recall, mean F1-score, and AUC of 0.879, 0.879, 0.879, 0.875, and 0.971, respectively. Compared to the MLP classifier with the 3D CNN features selected by Lasso, our method based on the MLP classifier has improved the classification performance by 5.8% (accuracy), 5.3% (mean precision), 5.8% (mean recall), 5.4% (mean F1-score), and 2.5% (AUC). Compared to the MLP classifier with lung radiomics features selected by Lasso, our method based on the MLP classifier has improved the classification performance by 5.0% (accuracy), 5.1% (mean precision), 5.0% (mean recall), 5.1% (mean F1-score), and 2.1% (AUC). Therefore, it is concluded that our method is effective in improving the classification performance for COPD stage classification.

Quantification of vesicoureteral reflux using machine learning

The radiographic grading of voiding cystourethrogram (VCUG) images is often used to determine the clinical course and appropriate treatment in patients with vesicoureteral reflux (VUR). However, image-based evaluation of VUR remains highly subjective, so we developed a supervised machine learning model to automatically and objectively grade VCUG data. A total of 113 VCUG images were gathered from public sources to compile the dataset for this study. For each image, VUR severity was graded by four pediatric radiologists and three pediatric urologists (low severity scored 1-3; high severity 4-5). Ground truth for each image was assigned based on the grade diagnosed by a majority of the expert assessors. Nine features were extracted from each VCUG image, then six machine learning models were trained, validated, and tested using 'leave-one-out' cross-validation. All features were compared and contrasted, with the highest-ranked then being used to train the final models. F1-score is a metric that is often used to indicate performance accuracy of machine learning models. When using the highest-ranked VCUG image features, F1-scores for the support vector machine (SVM) and multi-layer perceptron (MLP) classifiers were 90.27% and 91.14%, respectively, indicating a high level of accuracy. When using all features combined, F1 scores were 89.37% for SVM and 90.27% for MLP. These findings indicate that a distorted pattern of renal calyces is an accurate predictor of high-grade VUR. Machine learning protocols can be enhanced in future to improve objective grading of VUR.

Predicting rock-paper-scissors choices based on single-trial EEG signals.

Decision prediction based on neurophysiological signals is of great application value in many real-life situations, especially in human-AI collaboration or counteraction. Single-trial analysis of electroencephalogram (EEG) signals is a very valuable step in the development of an online decision-prediction system. However, previous EEG-based decision-prediction methods focused mainly on averaged EEG signals of all decision-making trials to predict an individual's general decision tendency (e.g., risk seeking or aversion) over a period rather than on a specific decision response in a single trial. In the present study, we used a rock-paper-scissors game, which is a common multichoice decision-making task, to explore how to predict participants' single-trial choice with EEG signals. Forty participants, comprising 20 females and 20 males, played the game with a computer player for 330 trials. Considering that the decision-making process of this game involves multiple brain regions and neural networks, we proposed a new algorithm named common spatial pattern-attractor metagene (CSP-AM) to extract CSP features from different frequency bands of EEG signals that occurred during decision making. The results showed that a multilayer perceptron classifier achieved an accuracy significantly exceeding the chance level among 88.57% (31 of 35) of participants, verifying the classification ability of CSP features in multichoice decision-making prediction. We believe that the CSP-AM algorithm could be used in the development of proactive AI systems.

Competing leaders grey wolf optimizer and its application for training multi-layer perceptron classifier

The multi-layer perceptron (MLP) is a highly popular artificial neural network used for classification across various applications. Swarm intelligence algorithms, such as the grey wolf optimizer, battle royale optimizer, sine cosine algorithm, and their improved variants, have been widely utilized for training MLP classifiers, addressing the challenging optimization problem that covers various difficulties in determining the optimal network parameters to minimize errors while processing datasets. Nevertheless, these prevailing algorithms frequently encounter problems like stagnation in local optima and inefficient search, leading to a deficiency in their ability to deliver efficient and accurate solutions. In order to overcome this challenge and provide a powerful tool for MLP classifier training, the competing leaders grey wolf optimizer (CGWO) is proposed in this study. CGWO benefits from a novel mechanism of competing leaders that provides a flexible wolf pack leadership hierarchy to avoid stagnation in local optima and accelerate convergence speed. In addition, a population diversity-enhanced initiation method is designed to help improve the efficiency of the mechanism of competing leaders. The performance of CGWO in training MLP classifiers is evaluated in an experiment comparing CGWO with nine state-of-the-art algorithms on eleven classification datasets representing a wide range of optimization challenges. CGWO substantially outperforms its competitors regarding optimal fitness, classification accuracy, and convergence speed; and shows competitive performance in computation time comparison.

Establishment of a Prediction Model Based on Preoperative MRI Radiomics for Diffuse Astrocytic Glioma, IDH-Wildtype, with Molecular Features of Glioblastoma.

In 2021, the WHO central nervous system (CNS) tumor classification criteria added the diagnosis of diffuse astrocytic glioma, IDH wild-type, with molecular features of glioblastoma, WHO grade 4 (DAG-G). DAG-G may exhibit the aggressiveness and malignancy of glioblastoma (GBM) despite the lower histological grade, and thus a precise preoperative diagnosis can help neurosurgeons develop more refined individualized treatment plans. This study aimed to establish a predictive model for the non-invasive identification of DAG-G based on preoperative MRI radiomics. Patients with pathologically confirmed glioma in Huashan Hospital, Fudan University, between September 2019 and July 2021 were retrospectively analyzed. Furthermore, two external validation datasets from Wuhan Union Hospital and Xuzhou Cancer Hospital were also utilized to verify the reliability and accuracy of the prediction model. Two regions of interest (ROI) were delineated on the preoperative MRI images of the patients using the semi-automatic tool ITK-SNAP (version 4.0.0), which were named the maximum anomaly region (ROI1) and the tumor region (ROI2), and Pyradiomics 3.0 was applied for feature extraction. Feature selection was performed using a least absolute shrinkage and selection operator (LASSO) filter and a Spearman correlation coefficient. Six classifiers, including Gauss naive Bayes (GNB), K-nearest neighbors (KNN), Random forest (RF), Adaptive boosting (AB), and Support vector machine (SVM) with linear kernel and multilayer perceptron (MLP), were used to build the prediction models, and the prediction performance of the six classifiers was evaluated by fivefold cross-validation. Moreover, the performance of prediction models was evaluated using area under the curve (AUC), precision (PRE), and other metrics. According to the inclusion and exclusion criteria, 172 patients with grade 2-3 astrocytoma were finally included in the study, and a total of 44 patients met the diagnosis of DAG-G. In the prediction task of DAG-G, the average AUC of GNB classifier was 0.74 ± 0.07, that of KNN classifier was 0.89 ± 0.04, that of RF classifier was 0.96 ± 0.03, that of AB classifier was 0.97 ± 0.02, that of SVM classifier was 0.88 ± 0.05, and that of MLP classifier was 0.91 ± 0.03, among which, AB classifier achieved the best prediction performance. In addition, the AB classifier achieved AUCs of 0.91 and 0.89 in two external validation datasets obtained from Wuhan Union Hospital and Xuzhou Cancer Hospital, respectively. The prediction model constructed based on preoperative MRI radiomics established in this study can basically realize the prospective, non-invasive, and accurate diagnosis of DAG-G, which is of great significance to help further optimize treatment plans for such patients, including expanding the extent of surgery and actively administering radiotherapy, targeted therapy, or other treatments after surgery, to fundamentally maximize the prognosis of patients.

Hyperspectral Images Weakly Supervised Classification with Noisy Labels

The deep network model relies on sufficient training samples to achieve superior processing performance, which limits its application in hyperspectral image (HSI) classification. In order to perform HSI classification with noisy labels, a robust weakly supervised feature learning (WSFL) architecture combined with multi-model attention is proposed. Specifically, the input noisy labeled data are first subjected to multiple groups of residual spectral attention models and multi-granularity residual spatial attention models, enabling WSFL to refine and optimize the extracted spectral and spatial features, with a focus on extracting clean samples information and reducing the model’s dependence on labels. Finally, the fused and optimized spectral-spatial features are mapped to the multilayer perceptron (MLP) classifier to increase the constraint of the model on the noisy samples. The experimental results on public datasets, including Pavia Center, WHU-Hi LongKou, and HangZhou, show that WSFL is better at classifying noise labels than excellent models such as spectral-spatial residual network (SSRN) and dual channel residual network (DCRN). On Hangzhou dataset, the classification accuracy of WSFL is superior to DCRN by 6.02% and SSRN by 7.85%, respectively.

STAMarker: determining spatial domain-specific variable genes with saliency maps in deep learning.

Spatial transcriptomics characterizes gene expression profiles while retaining the information of the spatial context, providing an unprecedented opportunity to understand cellular systems. One of the essential tasks in such data analysis is to determine spatially variable genes (SVGs), which demonstrate spatial expression patterns. Existing methods only consider genes individually and fail to model the inter-dependence of genes. To this end, we present an analytic tool STAMarker for robustly determining spatial domain-specific SVGs with saliency maps in deep learning. STAMarker is a three-stage ensemble framework consisting of graph-attention autoencoders, multilayer perceptron (MLP) classifiers, and saliency map computation by the backpropagated gradient. We illustrate the effectiveness of STAMarker and compare it with serveral commonly usedcompeting methods on various spatial transcriptomic data generated by differentplatforms. STAMarker considers all genes at once and is more robust when the dataset is very sparse. STAMarker could identify spatial domain-specific SVGs for characterizing spatial domains and enable in-depth analysis of the region of interest in the tissue section.

KSFinder-a knowledge graph model for link prediction of novel phosphorylated substrates of kinases.

Aberrant protein kinase regulation leading to abnormal substrate phosphorylation is associated with several human diseases. Despite the promise of therapies targeting kinases, many human kinases remain understudied. Most existing computational tools predicting phosphorylation cover less than 50% of known human kinases. They utilize local feature selection based on protein sequences, motifs, domains, structures, and/or functions, and do not consider the heterogeneous relationships of the proteins. In this work, we present KSFinder, a tool that predicts kinase-substrate links by capturing the inherent association of proteins in a network comprising 85% of the known human kinases. We also postulate the potential role of two understudied kinases based on their substrate predictions from KSFinder. KSFinder learns the semantic relationships in a phosphoproteome knowledge graph using a knowledge graph embedding algorithm and represents the nodes in low-dimensional vectors. A multilayer perceptron (MLP) classifier is trained to discern kinase-substrate links using the embedded vectors. KSFinder uses a strategic negative generation approach that eliminates biases in entity representation and combines data from experimentally validated non-interacting protein pairs, proteins from different subcellular locations, and random sampling. We assess KSFinder's generalization capability on four different datasets and compare its performance with other state-of-the-art prediction models. We employ KSFinder to predict substrates of 68 "dark" kinases considered understudied by the Illuminating the Druggable Genome program and use our text-mining tool, RLIMS-P along with manual curation, to search for literature evidence for the predictions. In a case study, we performed functional enrichment analysis for two dark kinases - HIPK3 and CAMKK1 using their predicted substrates. KSFinder shows improved performance over other kinase-substrate prediction models and generalized prediction ability on different datasets. We identified literature evidence for 17 novel predictions involving an understudied kinase. All of these 17 predictions had a probability score ≥0.7 (nine at >0.9, six at 0.8-0.9, and two at 0.7-0.8). The evaluation of 93,593 negative predictions (probability ≤0.3) identified four false negatives. The top enriched biological processes of HIPK3 substrates relate to the regulation of extracellular matrix and epigenetic gene expression, while CAMKK1 substrates include lipid storage regulation and glucose homeostasis. KSFinder outperforms the current kinase-substrate prediction tools with higher kinase coverage. The strategically developed negatives provide a superior generalization ability for KSFinder. We predicted substrates of 432 kinases, 68 of which are understudied, and hypothesized the potential functions of two dark kinases using their predicted substrates.

A heterogeneous graph convolutional attention network method for classification of autism spectrum disorder

BackgroundAutism spectrum disorder (ASD) is a serious developmental disorder of the brain. Recently, various deep learning methods based on functional magnetic resonance imaging (fMRI) data have been developed for the classification of ASD. Among them, graph neural networks, which generalize deep neural network models to graph structured data, have shown great advantages. However, in graph neural methods, because the graphs constructed are homogeneous, the phenotype information of the subjects cannot be fully utilized. This affects the improvement of the classification performance.MethodsTo fully utilize the phenotype information, this paper proposes a heterogeneous graph convolutional attention network (HCAN) model to classify ASD. By combining an attention mechanism and a heterogeneous graph convolutional network, important aggregated features can be extracted in the HCAN. The model consists of a multilayer HCAN feature extractor and a multilayer perceptron (MLP) classifier. First, a heterogeneous population graph was constructed based on the fMRI and phenotypic data. Then, a multilayer HCAN is used to mine graph-based features from the heterogeneous graph. Finally, the extracted features are fed into an MLP for the final classification.ResultsThe proposed method is assessed on the autism brain imaging data exchange (ABIDE) repository. In total, 871 subjects in the ABIDE I dataset are used for the classification task. The best classification accuracy of 82.9% is achieved. Compared to the other methods using exactly the same subjects in the literature, the proposed method achieves superior performance to the best reported result.ConclusionsThe proposed method can effectively integrate heterogeneous graph convolutional networks with a semantic attention mechanism so that the phenotype features of the subjects can be fully utilized. Moreover, it shows great potential in the diagnosis of brain functional disorders with fMRI data.

Enhancing accuracy in brain stroke detection: Multi-layer perceptron with Adadelta, RMSProp and AdaMax optimizers.

The human brain is an extremely intricate and fascinating organ that is made up of the cerebrum, cerebellum, and brainstem and is protected by the skull. Brain stroke is recognized as a potentially fatal condition brought on by an unfavorable obstruction in the arteries supplying the brain. The severity of brain stroke may be reduced or controlled with its early prognosis to lessen the mortality rate and lead to good health. This paper proposed a technique to predict brain strokes with high accuracy. The model was constructed using data related to brain strokes. The aim of this work is to use Multi Layer Perceptron (MLP) as a classification technique for stroke data and used multi-optimizers that include Adaptive moment estimation with Maximum (AdaMax), Root Mean Squared Propagation (RMSProp) and Adaptive learning rate method (Adadelta). The experiment shows RMSProp optimizer is best with a data training accuracy of 95.8% and a value for data testing accuracy of 94.9%. The novelty of work is to incorporate multiple optimizers alongside the MLP classifier which offers a comprehensive approach to stroke prediction, providing a more robust and accurate solution. The obtained results underscore the effectiveness of the proposed methodology in enhancing the accuracy of brain stroke detection, thereby paving the way for potential advancements in medical diagnosis and treatment.

Effectiveness of data augmentation to predict students at risk using deep learning algorithms

The academic intervention to predict at-risk higher education (HE) students requires effective data model development. Such data modelling projects in the HE context may have common issues related to (a) adopting small-scale modelling that gives limited options for early intervention and (b) using imbalanced data that hinders capturing effective details of poorly performing students. We address the issues going beyond the distribution-based algorithm, using a multilayer perceptron classifier which shows better on confusion metric, recall, and precision measures for identifying at-risk students. Our proposed deep learning-based model, which uses data augmentation techniques to supplement the data instances and balance the dataset, aims to improve the prediction accuracy of whether the student will fail or not based on their interaction with the learning management systems to prevent struggling students from evasion.

1D convolutional neural networks-based soil fertility classification and fertilizer prescription

Sustainable agriculture is essential to meet the demands of the global population. An adequate application of fertilizers is essential for sustainable agricultural productivity. This research aims to determine soil fertility and provide precise fertilizer to improve crop yield. Many researchers have proposed soil fertility classification using deep learning-based approaches, such as extreme learning machines (ELMs) and multilayer perceptrons (MLPs). Although both ELM and MLP have the highest performance, insufficient training data prevent them from being useful. To address this limitation, this research proposes a 1D convolution neural networks (1D-CNN)-based soil fertility classification method that is straightforward, compact, and supports scalar additions and multiplications. To classify soil fertility, the classifier employs laboratory-measured soil data that encompasses electrical conductivity, pH, organic carbon, potassium, phosphorus, sulfur, boron, copper, iron, manganese, and zinc. The proposed approach employs MinMax normalization and the synthetic minority oversampling technique (SMOTE) to improve the classifier performance. The results of soil classification are used to recommend fertilizers. An experimental study using a laboratory-measured soil dataset showed that the proposed technique outperformed ELM and MLP classifiers. The proposed approach outperformed ELM and MLP with a classification accuracy of 97.9%, while ELM and MLP achieved classification accuracies of 69.80% and 87.06%, respectively. The proposed method can help farmers manage soil fertility sustainably to increase crop production.

Survival Prediction in Diabetic Foot Ulcers: A Machine Learning Approach.

Our paper proposes the first machine learning model to predict long-term mortality in patients with diabetic foot ulcers (DFUs). The study includes 635 patients with DFUs admitted from January 2007 to December 2017, with a follow-up period extending until December 2020. Two multilayer perceptron (MLP) classifiers were developed. The first MLP model was developed to predict whether the patient will die in the next 5 years after the current hospitalization. The second MLP classifier was built to estimate whether the patient will die in the following 10 years. The 5-year and 10-year mortality models were based on the following predictors: age; the University of Texas Staging System for Diabetic Foot Ulcers score; the Wagner-Meggitt classification; the Saint Elian Wound Score System; glomerular filtration rate; topographic aspects and the depth of the lesion; and the presence of foot ischemia, cardiovascular disease, diabetic nephropathy, and hypertension. The accuracy for the 5-year and 10-year models was 0.7717 and 0.7598, respectively (for the training set) and 0.7244 and 0.7087, respectively (for the test set). Our findings indicate that it is possible to predict with good accuracy the risk of death in patients with DFUs using non-invasive and low-cost predictors.

Time Series Forecasting for Energy Production in Stand-Alone and Tracking Photovoltaic Systems Based on Historical Measurement Data

This article presents a time series analysis for predicting energy production in photovoltaic (PV) power plant systems, namely fixed and solar-tracking ones, which were located in the north-east of Poland. The purpose of one-day forecasts is to determine the effectiveness of preventive actions and manage power systems effectively. The impact of climate variables affecting the production of electricity in the photovoltaic systems was analyzed. Forecasting models based on traditional machine learning (ML) techniques and multi-layer perceptron (MLP) neural networks were created without using solar irradiance as an input feature to the model. In addition, a few metrics were selected to determine the quality of the forecasts. The preparation of the dataset for constructing the forecasting models was discussed, and some ways for improving the metrics were given. Furthermore, comparative analyses were performed, which showed that the MLP neural networks used in the regression problem provided better results than the MLP classifier models. The Diebold–Mariano (DM) test was applied in this study to distinguish the significant differences in the forecasting accuracy between the individual models. Compared to KNN (k-nearest neighbors) or ARIMA models, the best results were obtained for the simple linear regression, MLPRegressor, and CatBoostRegressor models in each of the investigated photovoltaic systems. The R-squared value for the MLPRegressor model was around 0.6, and it exceeded 0.8 when the dataset was split and separated into months.