Bayesian Classifier Research Articles

Comparison of machine learning algorithms for predicting cognitive impairment using neuropsychological tests

Objectives Neuropsychological tests (NPTs) are standard tools for assessing cognitive function. These tools can evaluate the cognitive status of a subject, which can be time-consuming and expensive for interpretation. Therefore, this paper aimed to optimize the systematic NPTs by machine learning and develop new classification models for differentiating healthy controls (HC), mild cognitive impairment, and Alzheimer’s disease dementia (ADD) among groups of subjects. Patients and methods A total dataset of 14,926 subjects was obtained from the formal 46 NPTs based on the Seoul Neuropsychological Screening Battery (SNSB). The statistical values of the dataset included an age of 70.18 ± 7.13 with an education level of 8.18 ± 5.50 and a diagnosis group of three; HC, MCI, and ADD. The dataset was preprocessed and classified in two- and three-way machine-learning classification from scikit-learn (www.scikit-learn.org) to differentiate between HC versus MCI, HC versus ADD, HC versus Cognitive Impairment (CI) (MCI + ADD), and HC versus MCI versus ADD. We compared the performance of seven machine learning algorithms, including Naïve Bayes (NB), random forest (RF), decision tree (DT), k-nearest neighbors (KNN), support vector machine (SVM), AdaBoost, and linear discriminant analysis (LDA). The accuracy, sensitivity, specificity, positive predicted value (PPV), negative predictive value (NPV), area under the curve (AUC), confusion matrixes, and receiver operating characteristic (ROC) were obtained from each model based on the test dataset. Results The trained models based on 29 best-selected NPT features were evaluated, the model with the RF algorithm yielded the best accuracy, sensitivity, specificity, PPV, NPV, and AUC in all four models: HC versus MCI was 98%, 98%, 97%, 98%, 97%, and 99%; HC versus ADD was 98%, 99%, 96%, 97%, 98%, and 99%; HC versus CI was 97%, 99%, 92%, 97%, 97%, and 99% and HC versus MCI versus ADD was 97%, 96%, 98%, 97%, 98%, and 99%, respectively, in predicting of cognitive impairment among subjects. Conclusion According to the results, the RF algorithm was the best classification model for both two- and three-way classification among the seven algorithms trained on an imbalanced NPTs SNSB dataset. The trained models proved useful for diagnosing MCI and ADD in patients with normal NPTs. These models can optimize cognitive evaluation, enhance diagnostic accuracy, and reduce missed diagnoses.

Prediction of childbearing tendency in women on the verge of marriage using machine learning techniques

The declining fertility rate and increasing marriage age among girls pose challenges for policymakers, leading to issues such as population decline, higher social and economic costs, and reduced labor productivity. Using machine learning (ML) techniques to predict the desire to have children can offer a promising solution to address these challenges. Therefore, this study aimed to predict the childbearing tendency in women on the verge of marriage using ML techniques. Data from 252 participants (203 expressing a "desire to have children" and 49 indicating "reluctance to have children") in Abadan, and Khorramshahr cities (Khuzestan Province, Iran) was analyzed. Seven ML algorithms, including multilayer perceptron (MLP), support vector machine (SVM), logistic regression (LR), random forest (RF), J48 decision tree, Naive Bayes (NB), and K-nearest neighbors (KNN), were employed. The performance of these algorithms was assessed using metrics derived from the confusion matrix. The RF algorithm showed superior performance, with the highest sensitivity (99.5%), specificity (95.6%), and receiver operating characteristic curve (90.1%) values. Meanwhile, MLP emerged as the top-performing algorithm, showcasing the best overall performance in accuracy (77.75%) and precision (81.8%) compared to other algorithms. Factors such as age of marriage, place of residence, and strength of the family center with the birth of a child were the most effective predictors of a woman's desire to have children. Conversely, the number of daughters, the wife's ethnicity, and the spouse's ownership of assets such as cars and houses were among the least important factors in predicting this desire. ML algorithms exhibit excellent predictive capabilities for childbearing tendencies in women on the verge of marriage, highlighting their remarkable effectiveness. This capacity to offer accurate prognoses holds significant promise for advancing research in this field.

Microblog Negative Comments Data Analysis Model Based on Multi-scale Convolutional Neural Network and Weighted Naive Bayes Algorithm

As a form of public supervision, Microblog’s negative reviews allow people to share their opinions and experiences and express dissatisfaction with unfair and unreasonable phenomena. This form of supervision has the potential to promote social fairness, drive governments, businesses, and individuals to correct mistakes and enhance transparency. To characterize the sentiment trend and determine the influence of Microblog negative reviews, we propose a multi-scale convolutional neural network and weighted naive bayes algorithm (MCNN–WNB). We define the feature vector characterization index for Microblog negative review data and preprocess the data accordingly. We quantify the relationship between attributes and categories using the weighted Naive Bayes method and use the quantification value as the weighting coefficient for the attributes, addressing the issue of decreased classification performance in traditional methods. We introduce a sentiment classification model based on word vector representation and a multi-scale convolutional neural networks to filter out Microblog negative review data. We conduct simulation experiments using real data, analyzing key influencing parameters such as convergence time, training set sample size, and number of categories. By comparing with K-means, Naive Bayes algorithm, Spectral Clustering algorithm and Autoencoder algorithm, we validate the effectiveness of our proposed method. We discover that the convergence time of the MCNN–WNB algorithm increases as the number of categories increases. The average classification accuracy of the algorithm remains relatively stable with varying test iterations. The algorithm’s precision increases with the number of training set samples and eventually stabilizes.

Decoding Breast Cancer: Using Radiomics to Non-Invasively Unveil Molecular Subtypes Directly from Mammographic Images.

Breast cancer is the most commonly diagnosed cancer worldwide. The therapy used and its success depend highly on the histology of the tumor. This study aimed to explore the potential of predicting the molecular subtype of breast cancer using radiomic features extracted from screening digital mammography (DM) images. A retrospective study was performed using the OPTIMAM Mammography Image Database (OMI-DB). Four binary classification tasks were performed: luminal A vs. non-luminal A, luminal B vs. non-luminal B, TNBC vs. non-TNBC, and HER2 vs. non-HER2. Feature selection was carried out by Pearson correlation and LASSO. The support vector machine (SVM) and naive Bayes (NB) ML classifiers were used, and their performance was evaluated with the accuracy and the area under the receiver operating characteristic curve (AUC). A total of 186 patients were included in the study: 58 luminal A, 35 luminal B, 52 TNBC, and 41 HER2. The SVM classifier resulted in AUCs during testing of 0.855 for luminal A, 0.812 for luminal B, 0.789 for TNBC, and 0.755 for HER2, respectively. The NB classifier showed AUCs during testing of 0.714 for luminal A, 0.746 for luminal B, 0.593 for TNBC, and 0.714 for HER2. The SVM classifier outperformed NB with statistical significance for luminal A (p = 0.0268) and TNBC (p = 0.0073). Our study showed the potential of radiomics for non-invasive breast cancer subtype classification.

Prediction of metabolic syndrome using machine learning approaches based on genetic and nutritional factors: a 14-year prospective-based cohort study

IntroductionMetabolic syndrome is a chronic disease associated with multiple comorbidities. Over the last few years, machine learning techniques have been used to predict metabolic syndrome. However, studies incorporating demographic, clinical, laboratory, dietary, and genetic factors to predict the incidence of metabolic syndrome in Koreans are limited. In the present study, we propose a genome-wide polygenic risk score for the prediction of metabolic syndrome, along with other factors, to improve the prediction accuracy of metabolic syndrome.MethodsWe developed 7 machine learning-based models and used Cox multivariable regression, deep neural network (DNN), support vector machine (SVM), stochastic gradient descent (SGD), random forest (RAF), Naïve Bayes (NBA) classifier, and AdaBoost (ADB) to predict the incidence of metabolic syndrome at year 14 using the dataset from the Korean Genome and Epidemiology Study (KoGES) Ansan and Ansung.ResultsOf the 5440 patients, 2,120 were considered to have new-onset metabolic syndrome. The AUC values of model, which included sex, age, alcohol intake, energy intake, marital status, education status, income status, smoking status, dried laver intake, and genome-wide polygenic risk score (gPRS) Z-score based on 344,447 SNPs (p-value < 1.0), were the highest for RAF (0.994 [95% CI 0.985, 1.000]) and ADB (0.994 [95% CI 0.986, 1.000]).ConclusionsIncorporating both gPRS and demographic, clinical, laboratory, and seaweed data led to enhanced metabolic syndrome risk prediction by capturing the distinct etiologies of metabolic syndrome development. The RAF- and ADB-based models predicted metabolic syndrome more accurately than the NBA-based model for the Korean population.

PET radiomics-based lymphovascular invasion prediction in lung cancer using multiple segmentation and multi-machine learning algorithms.

The current study aimed to predict lymphovascular invasion (LVI) using multiple machine learning algorithms and multi-segmentation positron emission tomography (PET) radiomics in non-small cell lung cancer (NSCLC) patients, offering new avenues for personalized treatment strategies and improving patient outcomes. One hundred and twenty-six patients with NSCLC were enrolled in this study. Various automated and semi-automated PET image segmentation methods were applied, including Local Active Contour (LAC), Fuzzy-C-mean (FCM), K-means (KM), Watershed, Region Growing (RG), and Iterative thresholding (IT) with different percentages of the threshold. One hundred five radiomic features were extracted from each region of interest (ROI). Multiple feature selection methods, including Minimum Redundancy Maximum Relevance (MRMR), Recursive Feature Elimination (RFE), and Boruta, and multiple classifiers, including Multilayer Perceptron (MLP), Logistic Regression (LR), XGBoost (XGB), Naive Bayes (NB), and Random Forest (RF), were employed. Synthetic Minority Oversampling Technique (SMOTE) was also used to determine if it boosts the area under the ROC curve (AUC), accuracy (ACC), sensitivity (SEN), and specificity (SPE). Our results indicated that the combination of SMOTE, IT (with 45% threshold), RFE feature selection and LR classifier showed the best performance (AUC = 0.93, ACC = 0.84, SEN = 0.85, SPE = 0.84) followed by SMOTE, FCM segmentation, MRMR feature selection, and LR classifier (AUC = 0.92, ACC = 0.87, SEN = 1, SPE = 0.84). The highest ACC belonged to the IT segmentation (with 45 and 50% thresholds) alongside Boruta feature selection and the NB classifier without SMOTE (ACC = 0.9, AUC = 0.78 and 0.76, SEN = 0.7, and SPE = 0.94, respectively). Our results indicate that selection of appropriate segmentation method and machine learning algorithm may be helpful in successful prediction of LVI in patients with NSCLC with high accuracy using PET radiomics analysis.

Performance Comparison of Supervised Machine Learning Methods in Classifying Celestial Objects

In recent times, astronomy has entered a new era with rapidly growing data sources and advanced observation techniques. The construction of powerful telescopes has enabled the collection of spectral data from millions of celestial objects. However, the increasing number and variety of data have made it challenging to categorize these celestial objects. This study employs machine learning methods to address the fundamental problem of classifying stars, galaxies, and quasars in astronomy. The dataset underwent detailed preprocessing to identify effective features for classification. KNIME Analytics Platform was used for data analysis and visualization, facilitating rapid and efficient data analysis through its drag-and-drop interface. Among the machine learning methods used in our study—Decision Trees, Random Forest, and Naive Bayes—the highest accuracy rate of 97.86% was achieved with the Random Forest model. Notably, despite its lower overall performance compared to other models, the Naive Bayes classifier exhibited superior performance in distinguishing the STAR class, which is one of the study's interesting findings. Future studies aim to enhance model accuracy by using larger and more diverse datasets and exploring different machine learning algorithms. Additionally, the impact of deep learning methods on classification performance will be investigated.

Population-Based Artificial Intelligence Assessment of Relationship Between the Risk Factors for Diabetic Retinopathy in Indian Population

ABSTRACT Purpose Risk factors (RFs), like ‘body mass index (BMI),’ ‘age,’ and ‘gender’ correlate with Diabetic Retinopathy (DR) diagnosis and have been widely studied. This study examines how these three secondary RFs independently affect the predictive capacity of primary RFs. Methods The dataset consisted of four population-based studies on the prevalence of DR and associated RFs in India between 2001 and 2010. An Autoencoder was employed to categorize RFs as primary or secondary. This study evaluated six primary RFs coupled independently with each secondary RF on five machine-learning models. Results The secondary RF ‘gender’ gave a maximum increase in Area under the curve (AUC) score to predict DR when combined separately with ‘insulin treatment,’ ‘fasting plasma glucose,’ ‘hypertension history,’ and ‘glycosylated hemoglobin’ with a maximum increase in AUC for the Naive Bayes model from 0.573 to 0.646, for the Support Vector Machines (SVM) model from 0.644 to 0.691, for the SVM model from 0.487 to 0.607, and for the Decision Tree model from 0.8 to 0.848, respectively. The secondary RFs ‘age’ and ‘BMI’ gave a maximum increase in AUC score to predict DR when combined separately with ‘diabetes mellitus duration’ and ‘systolic blood pressure,’ with a maximum increase in AUC for the SVM model from 0.389 to 0.621, and for the Decision Tree model from 0.617 to 0.713, respectively. Conclusion The risk factor ‘gender’ was the best secondary RF in predicting DR compared to ‘age’ and ‘BMI,’ increasing the predictive power of four primary RFs.

A Novel Method for Targeted Identification of Essential Proteins by Integrating Chemical Reaction Optimization and Naive Bayes Model.

Targeted identification of essential proteins is of great significance for species identification, drug manufacturing, and disease treatment. It is a challenge to analyze the binding mechanism between essential proteins and improve the identification speed while ensuring the accuracy of the identification. This paper proposes a novel method called EPCRO for identifying essential proteins, which incorporates the chemical reaction optimization (CRO) algorithm and the naive Bayes model to effectively detect essential proteins. In EPCRO, the naive Bayes model is employed to analyze the homogeneity between proteins. In order to improve the identification rate and speed of essential proteins, the protein homogeneity rate is integrated into the CRO algorithm to balance between local and global searches. EPCRO is experimentally compared with 17 existing methods (including, DC, SC, IC, EC, LAC, NC, PeC, WDC, EPD-RW, RWHN, TEGS, CFMM, BSPM, AFSO-EP, CVIM, RWEP, and EPPSO-DC) based on biological datasets. The results show that EPCRO is superior to the above methods in identification accuracy and speed.

Brake fault diagnosis using a voting ensemble of machine learning classifiers

Brake fault diagnosis is a crucial aspect of enhancing driving safety, as various faults such as air in brake fluid, brake oil spill, reservoir leak, mechanical fade and distinct types of brake pad wear can compromise vehicle safety. This study presents a method for timely fault detection by analyzing vibration signals. Vibration signals for normal and faulty conditions were captured using a hydraulic brake test setup equipped with a piezoelectric transducer and data acquisition system. Feature extraction was performed using an auto-regressive moving average (ARMA) model. The performance of five different classifiers namely, random forest (RF), Naive Bayes (NB), instance-based k-nearest neighbours (IBk), logistic model trees (LMT) and J48 decision tree was evaluated. The LMT classifier achieved the highest accuracy at 95.00 % followed by IBk, RF, J48 and NB with accuracies of 92.00 %, 90.00 %, 90.00 % and 87.00 %. To further improve the diagnosis accuracy, a voting-based ensemble approach was employed by combining two, three, four and five classifiers with the application of five different voting strategies. The results obtained showcase that a combination of three classifiers LMT, IBk and NB utilizing the majority voting rule yielded an enhanced classification accuracy of 98.00 % highlighting the effectiveness of this ensemble method in brake fault diagnosis. A similar classification accuracy of 98.00 % was achieved by four (LMT-IBk-RF-J48) and five (IBk-RF-MLP-J48-NB) classifier ensembles; however, considering the computational complexity three classifier ensembles were selected. Whilst two classifier ensembles (IBk-J48) achieved a maximum classification accuracy of 96.00 %.

Interpretable Machine Learning Model for Predicting the Prognosis of Guillain-Barré Syndrome Patients.

Machine learning (ML) is increasingly used in medical predictive modeling, but there are no studies applying ML to predict prognosis in Guillain-Barré syndrome (GBS). The medical records of 223 patients with GBS were analyzed to construct predictive models that affect patient prognosis. Least Absolute Shrinkage and Selection Operator (LASSO) was used to filter the variables. Decision Trees (DT), Random Forest (RF), Extreme Gradient Boosting (XGBoost), k-nearest Neighbour (KNN), Naive Bayes (NB), Neural Network (NN). Light Gradient Boosting Machine (LGBM) and Logistic Regression (LR) were used to construct predictive models. Clinical data from 55 GBS patients were used to validate the model. SHapley additive explanation (SHAP) analysis was used to explain the model. Single sample gene set enrichment analysis (ssGSEA) was used for immune cell infiltration analysis. The AUCs (area under the curves) of the 8 ML algorithms including DT, RF, XGBoost, KNN, NB, NN, LGBM and LR were as follows: 0.75, 0.896 0.874, 0.666, 0.742, 0.765, 0.869 and 0.744. The accuracy of XGBoost (0.852) was the highest, followed by LGBM (0.803) and RF (0.758), with F1 index of 0.832, 0.794, and 0.667, respectively. The results of the validation set data analysis showed AUCs of 0.839, 0.919, and 0.733 for RF, XGBoost, and LGBM, respectively. SHAP analysis showed that the SHAP values of blood neutrophil/lymphocyte ratio (NLR), age, mechanical ventilation, hyporeflexia and abnormal glossopharyngeal vagus nerve were 0.821, 0.645, 0.517, 0.401 and 0.109, respectively. The combination of NLR, age, mechanical ventilation, hyporeflexia and abnormal glossopharyngeal vagus used to predict short-term prognosis in patients with GBS has a good predictive value.

Predicting Cardiovascular Disease Risk Using Web-Text Sentiment Analysis and Hybrid Deep Learning Models

Background: Sentiment Analysis (SA) has emerged as a key tool within Natural Language Processing (NLP) for quantifying emotions expressed in Web Text (WT). Its application in healthcare, particularly in predicting cardiovascular diseases (CD), shows promise. Sentiments related to stress, anger, or other emotional states have been linked to increased CD risk, making Web-Text Sentiment Analysis (WT-SA) a valuable tool in public health prediction. This study compares WT-SA to demographic data from the Centers for Disease Control and Prevention (CDC) in predicting CD risks. Methods: Twitter data was analyzed using SA techniques to assess sentiments related to CD. A hybrid deep learning (DL) model combining 3D-Convolutional Neural Networks (3D-CNN) and Recurrent Neural Networks (RNN) was utilized alongside traditional machine learning (ML) models: Naïve Bayes (NB), Support Vector Machine (SVM), Logistic Regression (LR), and CatBoost. The WT data, spanning three years (2020-2023), was split into training (75%) and testing (25%) datasets. Model performance was evaluated using Accuracy, Precision, Recall, and F1 score. Results: The hybrid 3D-CNN+RNN model achieved the highest test accuracy of 0.95 on Twitter data, outperforming all other models. SVM, LR, NB, and CatBoost achieved accuracies of 0.89, 0.88, 0.75, and 0.77, respectively. When applied to the CDC dataset, the hybrid model reached an accuracy of 0.678, still outperforming the other models. Overall, the WT-SA model demonstrated superior performance compared to using demographic data alone. Conclusion: WT-SA, when combined with hybrid DL models, is a more effective method for predicting CD risk than demographic-based models. This study determined the potential of NLP and DL techniques in leveraging social media data for public health monitoring, suggesting that WT-SA could be a valuable tool for predicting CD risks and enhancing early intervention efforts.

A Machine Learning Model Based on CT Imaging Metrics and Clinical Features to Predict the Risk of Hospital-Acquired Pneumonia After Traumatic Brain Injury.

To develop a validated machine learning (ML) algorithm for predicting the risk of hospital-acquired pneumonia (HAP) in patients with traumatic brain injury (TBI). We employed the Least Absolute Shrinkage and Selection Operator (LASSO) to identify critical features related to pneumonia. Five ML models-Logistic Regression (LR), Extreme Gradient Boosting (XGB), Random Forest (RF), Naive Bayes Classifier (NB), and Support Vector Machine (SVC)-were developed and assessed using the training and validation datasets. The optimal model was selected based on its performance metrics and used to create a dynamic web-based nomogram. In a cohort of 858 TBI patients, the HAP incidence was 41.02%. LR was determined to be the optimal model with superior performance metrics including AUC, accuracy, and F1-score. Key predictive factors included Age, Glasgow Coma Score, Rotterdam Score, D-dimer, and the Systemic Immune Response to Inflammation Index (SIRI). The nomogram developed based on these predictors demonstrated high predictive accuracy, with AUCs of 0.818 and 0.819 for the training and validation datasets, respectively. Decision curve analysis (DCA) and calibration curves validated the model's clinical utility and accuracy. We successfully developed and validated a high-performance ML algorithm to assess the risk of HAP in TBI patients. The dynamic nomogram provides a practical tool for real-time risk assessment, potentially improving clinical outcomes by aiding in early intervention and personalized patient management.

IoT Attacks Detection Using Supervised Machine Learning Techniques

In recent times, the growing significance of Internet of Things (IoT) devices in people's lives is undeniable, driven by their myriad benefits. However, these devices confront cybersecurity threats akin to traditional network devices, as they depend on networks for connectivity and synchronization. Artificial Intelligence (AI) techniques, specifically Machine Learning (ML) and Deep Learning (DL), have demonstrated notable reliability in the field of cyberattack detection. This study focuses on detecting Flood and Brute Force cyberattacks using Machine Learning (ML) and Deep Learning (DL) models. The primary emphasis lies in identifying traffic features that significantly detect these types of attacks. The experimental study incorporates eight models: Decision Tree (DT), K-Nearest Neighbor (KNN), Random Forest (RF), Support Vector Machines (SVM), Logistic Regression (LR), Gradient Boosting (GB), Naïve Bayes (NB), and Artificial Neural Network (ANN). Two sets of experiments were conducted, with the first set involving six features and the subsequent set, after feature selection, focusing on a reduced set of three features. The evaluation of the proposed model's efficiency and performance relied on metrics such as Accuracy, Precision, Recall, and F1-score. Remarkably, all proposed models exhibited high performance in both sets of experiments. However, the Gradient Boosting (GB) classifier suppressed others, attaining an impressive accuracy level of 95.94% and 95.28% in the sets with six features and three features, respectively. Doi: 10.28991/HIJ-2024-05-03-01 Full Text: PDF

Enhancing stroke prediction using the waikato environment for knowledge analysis

&lt;p&gt;State-of-the-art data mining tools incorporate advanced machine learning (ML) and artificial intelligence (AI) models, and it is widely used in classification, association rules, clustering, prediction, and sequential models. Data mining is important for the process of diagnosing and predicting diseases in the early stages, and this contributes greatly to the development of the health services sector. This study utilized classification to predict the stroke of a sample of the patient dataset that was taken from Kaggle. The classification model was created using the data mining program waikato environment for knowledge analysis (WEKA). This data mining tool helped identify individuals most at risk of stroke based on analysis of features extracted from the patient’s dataset. These features were used in classification processes according to the naive Bayes (NB), random forest (RF), support vector machine (SVM), and multi-layer perceptron (MLP) algorithms. Analysis of the classification results of the previous algorithms showed that the SVM outperformed other algorithms in terms of accuracy (94.4%), sensitivity (100%), and F-measure (97.1%). However, the NB algorithm had the best performance in terms of precision (95.7%).&lt;/p&gt;

Evaluation of Bayesian point-based system on the variant classification of hereditary cancer predisposition genes

PurposeTo assess the differences in variant classifications using the American College of Medical Genetics and Genomics and the Association for Molecular Pathology 2015 guidelines and the Bayesian point-based classification system (here referred to as the point system) in 115 hereditary cancer predisposition genes and explore variant subtiering by the point system. MethodsGermline variant classifications for 721 pediatric patients from an in-house panel were retrospectively evaluated using the 2 scoring systems. ResultsA total of 2376 unique variants were identified, with ∼23.5% exhibiting discordant classifications. Unique variants classified by the point system demonstrated a lower rate of variants of uncertain significance (VUS; ∼15%) compared with American College of Medical Genetics and Genomics and the Association for Molecular Pathology 2015 (∼36%). This change is attributed to unique variants with 1 benign supporting evidence (∼12%) or 1 benign strong evidence (∼4%) being classified as likely benign by the point system. Additionally, variants with conflicting/modified evidence (∼5% of 2376) are also resolved by the point system. Subtiering unique variants classified by the point system as VUS (n = 354) indicates ∼77.4% were VUS-Low (0-1 points), whereas the remaining ∼22.6% were VUS-Mid (2-3 points) and VUS-High (4-5 points). ConclusionThe point system reduces the VUS rate and facilitates their subtiering. Future large-scale studies are warranted to explore the impact of the point system on improving VUS reporting and/or VUS clinical management.

Predicting non-suicidal self-injury among Chinese adolescents: The application of ten algorithms of machine learning

Background and aimsHigh non-suicidal self-injury (NSSI) prevalence among adolescents is a global health issue. However, current prediction models for adolescent NSSI rely on a limited set of algorithms, resulting in biased predictions. Therefore, the aim of this study is to develop multiple machine learning models to enhance prediction accuracy and mitigate biases among Chinese adolescents. MethodsA total of 4487 junior and senior high school students in China were recruited. Multiple algorithms were included, such as logistic regression, decision tree, support vector machine, Naive Bayes, multi-layer perceptron, K-nearest neighbors, and ensemble learning algorithm like random forest, bagging, AdaBoost, and stacking to build predictive models. Data processing techniques, including standardization and the synthetic minority oversampling technique, were employed to optimize the predictive model. The model was trained on 70 % of the data, reserving 30 % for testing. ResultsThe ten prediction models achieved a good performance, with area under the receiver operating characteristic curve (AUC) scores above 0.700 in the test set. The stacking and random forest models achieved AUC scores of 0.904 and 0.898, respectively. The prediction performance of the Naive Bayes model was relatively poor. The top five important variables were resilience, bully, suicidal ideation, internet addiction, and depression. ConclusionsThe ensemble machine learning algorithm showed promising results predicting NSSI among adolescents. Such algorithms should be recommended for future NSSI research to enhance predictive accuracy. Identification of important features in NSSI prediction can help develop screening protocols and lay a foundation for clinical diagnosis and intervention in adolescent populations.

Employee Attrition Prediction Using Machine Learning Techniques

Effective employees are regarded as the most precious assets and the foundation of any business. Companies invest a lot of money in staff training programs because they believe they will pay off in the long run. Therefore, it is crucial to keep a long-term, promising staff; as has been seen throughout the years, this is one of HR's most challenging duties. This study aims to determine the key variables influencing employee attrition and develop a machine learning model to predict employee attrition based on the variables given. This will make it easier for management to spot workers who are likely to quit, allowing them to take preventative measures and make wise decisions about appraisal and recognition. The four most popular classification algorithms utilized in this study were KNN, Naive Bayes, Random Forest and Logistic Regression. It was shown that Naive Bayes classifier outperformed the others by 89% in terms of accuracy and produced more accurate predictions.

NEW AML TOOLS: ANALYZING ETHEREUM CRYPTOCURRENCY TRANSACTIONS USING A BAYESIAN CLASSIFIER

The emergence of cryptocurrencies as a form of digital payments has contributed to the emergence of numerous opportunities for the implementation of effective and efficient financial transactions, however, new fraud and money laundering schemes have emerged, as the anonymity and decentralization inherent in cryptocurrencies complicate the process of monitoring transactions and control by governments and law enforcement agencies. This study aims to develop a mechanism for analyzing transactions in the Ethereum cryptocurrency using a Bayesian classifier to identify potentially suspicious transactions that may be related to terrorist financing and money laundering. The Bayesian approach makes it possible to consider the probabilistic characteristics of transactions and their interrelationships to increase the accuracy of detecting anomalous and potentially illegal transactions. For the analysis, data on transactions of the Ethereum currency from June 2020 to December 2022 were taken. The developed mechanism involves determining a set of characteristics of transaction graph nodes that identify the potential for their use in illegal financial transactions and forming intervals of their permissible values. The article presents cryptocurrency transactions as an oriented graph, with the nodes being the entities conducting transactions and the arcs being the transactions between the nodes. In assessing the risks of using cryptocurrencies in money laundering, the number/amount of transactions to and from the respective node, the balance of these transactions (absolute value), and the type of node were considered. The analysis showed that among the 100 largest nodes in the network, 11 were identified as having a «critical» risk level, and the most closely connected nodes were identified. This methodology can be used not only to analyze the Ethereum cryptocurrency but also for other cryptocurrencies and similar networks.

Assessing the Impact of Urban Morphologies on Waterlogging Risk Using a Spatial Weight Naive Bayes Model and Local Climate Zones Classification

Assessing the Impact of Urban Morphologies on Waterlogging Risk Using a Spatial Weight Naive Bayes Model and Local Climate Zones Classification