Predicting and identifying factors affecting sheet metal bending times using explainable AI

Blood pressure abnormality detection and Interpretation utilizing Explainable Artificial Intelligence

Type-2 Diabetes Mellitus (T2DM) is a rapidly escalating global health challenge, often diagnosed only after clinical symptoms appear, leading to delayed intervention and higher risk of complications. Early prediction using automated computational approaches can significantly improve disease prognosis and reduce healthcare costs. This research proposes a machine learning-driven predictive framework for early detection of Type-2 diabetes by integrating multimodal data, including clinical parameters (such as glucose level, blood pressure, insulin, BMI), demographic attributes, and lifestyle indicators (dietary habits, physical activity, stress level, and sleep patterns). The dataset underwent preprocessing techniques such as normalization, missing value imputation, correlation-based feature selection, and class imbalance handling using SMOTE. Multiple machine learning algorithms—including Logistic Regression, Random Forest, Support Vector Machine, Gradient Boosting, and Extreme Gradient Boosting (XGBoost)—were trained and evaluated to identify the best-performing model. Model performance was assessed using accuracy, precision, recall, F1-score, and ROC-AUC metrics. The XGBoost model achieved superior predictive accuracy and demonstrated strong generalization capability across test samples. Furthermore, explainable AI (XAI) techniques such as SHAP values were employed to interpret feature importance and enhance clinical transparency. Results indicate that lifestyle factors combined with clinical metrics significantly improve predictive performance compared to clinical data alone. The proposed framework shows potential for integration into digital health platforms and preventive screening systems, aiding clinicians in early risk stratification and personalized intervention. Keywords Type-2 Diabetes Prediction, Machine Learning, Multimodal Healthcare Data, XGBoost, Predictive Analytics, Lifestyle Factors, Explainable AI (XAI), Early Diagnosis.

Android is a popular target for malware attacks due to its rapid proliferation, which emphasizes the need for efficient and understandable detection methods. A novel ensemble‐based framework that combines semantic feature engineering (SFE) and explainable artificial intelligence (XAI) is presented in this paper. It addresses the generalizability and transparency issues with current models. The purpose of combining low‐level Android permissions into high‐level semantic categories is to improve interpretability and reduce feature dimensionality, making the model easier to understand. Grid search optimization is used to assess a stacked ensemble classifier that consists of k‐nearest neighbors (KNN), random forest (RF), gradient boosting (GB), and support vector machine (SVM) models with a logistic regression (LR) as a meta‐learner on the TUANDROMD dataset. The model demonstrates strong class separation and generalization with a high test accuracy of 97.54%, with an F1‐score of 0.9847 and a ROC‐AUC of 0.9929. For explainability, permutation feature importance (PFI), partial dependence plots (PDPs), and a global surrogate decision tree (GSDT) are used. Results show that network and location permissions are the most important features. This study offers a robust and interpretable solution for Android malware detection. Its combination of SFE, ensemble learning (EL), and XAI makes it suitable for real‐world cybersecurity applications, especially on mobile devices.

Student drop-out is one of the most burning issues in STEM higher education, which induces considerable social and economic costs. Using machine learning tools for the early identification of students at risk of dropping out has gained a lot of interest recently. However, there has been little discussion on dropout prediction using interpretable machine learning (IML) and explainable artificial intelligence (XAI) tools.In this work, using the data of a large public Hungarian university, we demonstrate how IML and XAI tools can support educational stakeholders in dropout prediction. We show that complex machine learning models – such as the CatBoost classifier – can efficiently identify at-risk students relying solely on pre-enrollment achievement measures, however, they lack interpretability. Applying IML tools, such as permutation importance (PI), partial dependence plot (PDP), LIME, and SHAP values, we demonstrate how the predictions can be explained both globally and locally. Explaining individual predictions opens up great opportunities for personalized intervention, for example by offering the right remedial courses or tutoring sessions. Finally, we present the results of a user study that evaluates whether higher education stakeholders find these tools interpretable and useful.

Screening for gastric intestinal metaplasia (GIM) holds significant importance for the early detection of gastric cancer. To help clinicians identify high-risk GIM patients and determine the timing of gastric mucosal biopsy, we aim to develop a predictive model for the occurrence of GIM in patients. Patients were collected from the First Affiliated Hospital of Dalian Medical University, following rigorous inclusion and exclusion criteria. Initially, the VarSelRF algorithm identified independent variables linked to GIM development. We employed eight machine learning algorithms, including Decision Trees (DT), Elastic Net (ENet), K-Nearest Neighbors (KNN), LightGBM, Random Forest (RF), eXtreme Gradient Boosting (XGBoost), Support Vector Machine (SVM), and Multi-Layer Perceptron (MLP) to construct predictive models. Their performances were benchmarked using ROC curves, calibration curves, and decision curve analysis (DCA) curves. We also applied SHAP values to interpret the RF model, quantifying the contribution of each feature to predictions. Additionally, a web-based calculator was developed based on the RF model to facilitate practical clinical applications. Among the 975 patients examined, 322 individuals were pathologically confirmed to have GIM. Eleven independent variables significantly contributed to GIM occurrence, including gastric mucosal atrophy, H. pylori infection, direct bilirubin (DBIL), creatinine (Crea), smoking and alcohol history, gender, alanine aminotransferase (ALT), age, albumin/globulin ratio (ALB/GLO), and gamma-glutamyltransferase (GGT). The RF model demonstrated strong performance among the eight machine learning algorithms tested, achieving an AUC of 0.8167 in the testing dataset, along with a specificity of 85.5% and a sensitivity of 57.0%. The model's interpretive capabilities were enhanced by SHAP values, which helped clinicians understand the decision-making process. The resulting web-based calculator serves as a practical tool for clinicians. This study highlights the innovative use of serological biomarkers to assess the risk of GIM. We found that certain markers related to liver and kidney function are strong predictors of GIM development. Additionally, the application of SHAP values improves the understanding of how features contribute to predictions, while the newly developed web-based calculator offers a practical tool for clinicians to evaluate GIM risk more easily.

The EGNOS NOTAM proposals provision is a key enabler for the publication of RNP approaches down to LPV minima in Europe, as it provides the different AIS (Aeronautical Information Service) the required information to establish a NOTAM service covering this type of procedures, in line with the ICAO standards. Currently, more than 400 APV-I and 150 LPV-200 SBAS procedures are published in Austria, Belgium, Croatia, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Lithuania, Luxembourg, Malta, Montenegro, Netherlands, Norway, Poland, Portugal, Romania, Serbia, Slovak Republic, Spain, Sweden, Switzerland, UK and the Bailiwick of Guernsey. The EURONOTAM tool is responsible for generating EGNOS NOTAM proposals, implementing an analytical service volume that propagates the orbits of the satellites (based on the GPS almanacs) and computes the satellites in view for each configured airport. Once the satellite visibility is computed, the prediction module estimates the residual errors for each Line Of Sight (LoS), taking into account the number of EGNOS Ranging Integrity Monitoring Stations (RIMS) in view (for the User Differential Range Error —UDRE— computation) and the number of Ionospheric Pierce Points surrounding the neighbouring Ionospheric Grid Points (for the Grid Ionospheric Vertical Error —GIVE— estimation). With these estimations, the tool calculates the Protection Levels for each airport and determines if a new NOTAM proposal is to be issued in case the service is not available to support APV-I or LPV-200 operations in a specific location. Thus, the accuracy of the UDRE and GIVE estimations is essential in order to precisely predict the EGNOS availability events. Taking this into account, the present study is aimed at analysing the feasibility and the potential improvement that could be obtained, by using Machine Learning (ML) or Deep Learning (DL) techniques to enhance or replace the current analytical model. More specifically, the different “classic” ML models which will be addressed are decision trees, random forests, support vector machines and extreme gradient boosting (XGBoost). On top of these ML models, DL techniques are also covered by this study. The outputs of these models will be used to estimate the Protection Levels, and its applicability to the current EURONOTAM software will be discussed. The study will start with the analysis of the input dataset used to train the model. Aimed at analysing the impact of EGNOS station outages on the system performance and the capacity of the model to predict such impact, the input dataset will be composed of a minimum of 41 scenarios, including single and multiple station outages. For the UDRE estimation, more than 1.1 million samples, including the following features, will be considered: satellite coordinates, satellite dilution of precision (DOP) (based on the number of station in view), number of RIMS-A and RIMS-C on sight, satellite family, slot number and RIMS status. Similar features will be considered for the GIVE, with more than 3.5 million samples, including as well information on the number and distribution of the IPPs surrounding each IGP. Feature importance will be evaluated using SHAP values. SHAP values are calculated for each input feature and for each sample, and they measure the contribution of that feature to the predicted value for the sample. Summing the absolute SHAP values for all samples gives a measure of how the output is driven by that feature globally. This will be used to identify which are the most important features (and eventually, consider updating EURONOTAM model to consider these features – e.g. DOP) and discard the unimportant ones from the ML model. The real UDREi and GIVEi values used for the model training and testing will be obtained from the real EGNOS messages downloaded from the EDAS FTP. After preparing the input data set and splitting it into the training and testing datasets, the model hyperparameters will be tuned through a grid search with 5 fold cross-validation (for each set of values, 4/5 of the original training set will be used for actually training the model and the remaining 1/5 will be used for assessing the performance). The parameters resulting in the best averaged model performance will be selected. The resulting model will be evaluated with the testing dataset, using the following metrics: - Binary Accuracy: ratio between number of samples correctly classified as monitored or not-monitored vs the total number of samples - Total Accuracy: ratio between the number of correctly classified samples (considering the 16 possible labels) and the total number of samples. - Matthews correlation coefficient (MCC): A coefficient of +1 represents a perfect prediction and 0 represents an average random prediction. - Mean & RMS error on UDRE or GIVE value (UDREi / GIVEi converted to UDRE /GIVE for the assessment). The results obtained so far seem to indicate that Machine learning models can significantly improve the estimates obtained from the current analytical model (geometric-based). For instance, for one of the ML models tested, the system is able to correctly classify the 86% of the UDREi labels versus the 76% obtained with the current model, meaning a 13% improvement. In terms of binary accuracy, the ML model properly estimates the monitored / not-monitored status for the 99% of the samples tested. The RMS and mean absolute error are decreased by a 45% and 55% respectively. For the GIVE estimation, the improvement in the classification problem is more remarkable, with a 30% enhancement with respect to the current value. The RMS and mean absolute errors decrease as well by a 36% and a 50% factor respectively. The study will determine if Machine learning models can significantly improve the estimates obtained from the current analytical model (geometric-based). The study will also evaluate Protection Levels predicted with the usage of ML-estimated UDRE and GIVE values, where an important improvement in terms of prediction accuracy is expected. The analysis will be completed with a brief discussion on the available software libraries enabling the embedding of ML trained models in existing code and the capability of the model to predict situations that are very different from the ones actually trained. This is particularly challenging in this study, due to the lack of a full set of samples covering all the non-nominal conditions, since potential future situations might have never happened before and cannot be therefore, trained. Future steps will focus on this topic, and aim at finding more robust machine learning models that are able to deal with non-nominal conditions while still outperforming traditional methods under nominal conditions.

Background: In this study, we aimed to establish an efficient and accurate machine -learning model for evaluating tobacco quality based on its chemical composition. To achieve this, we gathered a dataset comprising 188 tobacco samples taken from the production area of Sichuan, China in 2021. Four machine learning algorithms—Random Forest (RF), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Gradient Boosting Classifier (GBC)—were used to establish a predictive model for assessing tobacco quality. The study focused on comparing the predictive performance of these models and exploring the upper limit of prediction accuracy using genetic algorithm (GA) hyperparameter optimization. Additionally, the SHAP value model interpretation framework was introduced to provide a comprehensive global interpretation and conduct feature dependency analysis. Results: The results showed that model accuracy ranked as RF>GBC> KNN>SVM, with the GA-RF machine learning model achieving a prediction accuracy of 86.8%. SHAP values identified seven important characteristic indexes affecting Sichuan tobacco leaf quality, highlighting RF as the optimal classifier for predicting flue-cured tobacco quality in Sichuan Province. Conclusions: The GA-RF model constructed in this study effectively identifies Sichuan tobacco leaf quality. These findings offer novel insights and data support for the application of machine learning algorithms in tobacco fields and tobacco leaf quality evaluation.

Revisiting probabilistic neural networks: a comparative study with support vector machines and the microhabitat suitability for the Eastern Iberian chub (Squalius valentinus)

Accurate maize yield forecasting under climate uncertainty remains a critical challenge for global food security, yet existing studies predominantly rely on single-model frameworks, limiting generalizability and actionable insights. This study selected three regions, specifically Dali, Lijiang, and Zhaotong, and collected data on 12 agronomic traits of 114 varieties, along with eight sets of meteorological data, covering the period from 2019 to 2023. We employed three machine learning models: Random Forest (RF), Support Vector Machine (SVM), and XGBoost. The results revealed a strong correlation between yield and multiple agronomic traits, particularly grain weight per spike (GWPS) and hundred-kernel weight (HKW). Notably, the XGBoost model emerged as the top performer across all three regions. The model achieved the lowest RMSE (0.22–191.13) and a good R2 (0.98–0.99), demonstrating exceptional predictive accuracy for yield-related traits. The comparative analysis revealed that XGBoost exhibited superior accuracy and stability compared to RF and SVM. Through feature importance analysis, four critical determinants of yield were identified: GWPS, shelling percentage (SP), growth period (GP), and plant height (PH). Furthermore, partial dependence plots (PDPs) provided deeper insights into the nonlinear interactive effects between GWPS, SP, GP, PH, and yield, offering a more comprehensive understanding of their complex relationships. This study presents an innovative, data-driven methodology designed to accurately forecast corn yield across diverse locations. This approach offers valuable scientific insights that can significantly enhance precision agricultural practices by enabling the precise tailoring of fertilizer usage and irrigation strategies. The results highlight the importance of integrating agronomic and meteorological data in yield forecasting, paving the way for development of agricultural decision-support systems in the context of future climate change scenarios. This study presents an innovative, data-driven methodology designed to accurately forecast corn yield across diverse locations. This approach offers valuable scientific insights that can significantly enhance precision agricultural practices by enabling the precise tailoring of fertilizer usage and irrigation strategies.

Cesarean section (C-section) rates have been rising globally, posing potential health risks for mothers and infants. Understanding the factors that contribute to C-section delivery and leveraging machine learning (ML) techniques for predictive modeling can support targeted interventions and informed policy decisions. This study aimed to identify the determinants of C-section delivery and develop an ML-based predictive model using data from the Bangladesh Demographic and Health Survey (BDHS) 2022. A total of 2,490 complete records of ever-married women aged 15–49 years were analyzed, where the delivery mode was categorized as vaginal or C-section. Three feature selection techniques including Recursive Feature Elimination (RFE), Boruta-based selection (BFS), and Random Forest (RF) were used to identify key risk factors. Six ML algorithms, including Logistic Regression (LR), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Decision Tree (DT), Random Forest (RF), and Extreme Gradient Boosting (XGB) were employed to predict C-section. Model performance was evaluated using accuracy, precision, recall, F1-score, AUC, and ROC analysis. SHAP values were used to interpret the influence of individual features.The prevalence of C-section deliveries was 45.6%, with an average maternal age of 25.7 years and mean age at first childbirth of 19.3 years. Ten significant determinants were identified, including place of delivery, baby weight, maternal BMI, birth interval, age at first birth, partner’s education, maternal age, wealth status, ANC visits, and maternal education. The RF model achieved the highest performance with an accuracy of 81.79%, and an AUC of 0.871. SHAP analysis highlighted that place of delivery, baby weight, maternal BMI, and birth interval were the most influential predictors. These findings suggest that socio-demographic and healthcare-related factors strongly influence C-section delivery. Machine learning models particularly the RF can effectively identify women at high risk, supporting strategies to reduce unnecessary C-sections and improve maternal healthcare planning in Bangladesh.

This article examines the integration of artificial intelligence in financial evaluation and the vital role of explainability in building trustworthy decision support systems. As AI transforms traditional financial evaluation from forecasting to portfolio management, the inherent opacity of sophisticated algorithms creates tension with the financial sector's transparency requirements. The discussion explores how Explainable AI techniques—particularly SHAP values and LIME—enable financial professionals to understand AI-generated insights while maintaining regulatory compliance. Through examining real-world implementations, the article demonstrates quantifiable benefits of explainable models in reducing false positives, improving analyst confidence, and accelerating regulatory approval. The evaluation extends to comprehensive Responsible AI frameworks encompassing fairness and bias mitigation, privacy-preserving techniques, and adversarial resilience mechanisms. The discussion addresses how generative AI assistants revolutionize document evaluation by automating summarization and data extraction while confronting critical security challenges, including prompt injection attacks, data leakage, and regulatory compliance complexities. The article emphasizes human-in-the-loop paradigms and tiered governance frameworks that successfully balance innovation with appropriate oversight, while examining real-time explainability challenges and monitoring requirements. Forward-looking perspectives on regulatory harmonization and the convergence of explainable, privacy-preserving, and robust AI systems demonstrate the evolution toward trustworthy financial AI implementations.

Machine Learning Based Prediction of One Year Mortality after Allogeneic Hematopoietic Cell Transplantation (alloHCT) Highlights Importance of Pre-Transplant Immunocompetence

Effects of heavy metal exposure on hypertension: A machine learning modeling approach

Machine learning-based design of target property-oriented fuels using explainable artificial intelligence

2.25Cr-1Mo-0.25V steel is the main material of hydrogenation reactor due to its good high temperature properties and hydrogen embrittlement resistance. As we all know, there is a great performance gap between virgin material and actual product. Nevertheless, current design methods cannot evaluate the impact of manufacturing. In engineering practice, tensile testing is widely implemented on online surveillance samples or experimentally simulated samples to acquire the actual material characteristics. However, sampling from service hydrogenation reactor, as well as simulating actual manufacturing process, is time-consuming, labor-intensive, and difficult. To systematically analyze the effect of fabrication on 2.25Cr-1Mo-0.25V steel, a prediction method that can consider manufacturing residual effects needs to be proposed. Nowadays, many data-driven methods, such as back propagation neural network (BPNN), support vector machine (SVM) and random forest (RF), are employed in predicting material behavior during hot forming, whereas they are scarcely applied on 2.25Cr-1Mo-0.25V steel. In this work, the mentioned three machine learning procedures were constructed to simulate the behavior of actually manufactured 2.25Cr-1Mo-0.25V steel in uniaxial tensile test. Firstly, tensile samples were extracted from different locations of an actually manufactured hydrogenation reactor, and stretched at 454 °C and 482 °C. Subsequently, BPNN, SVM and RF models were developed on the obtained tensile curves, respectively. Their inputs were manufacturing parameter, sampling location, service temperature and tensile strain, whereas their output was tensile stress. Finally, the predictability of models was evaluated by test set data which were not involved in model training process. The results show that the predictions of BPNN, SVM and RF all agree well with experimental data. BPNN exhibits lowest prediction error and highest generalization power, SVM is somewhere in between, while RF is the worst. This work will contribute to the application of data-driven methods in material property prediction of 2.25Cr-1Mo-0.25V steel.