Proposing a Design Theory for a Human-Learning-Guided Virtual Negotiator for Online Trading PlatformsSupplementary Materials for Proposing a Design Theory for a Human-Learning-Guided Virtual Negotiator for Online Trading Platforms

The Unmanned Aerial Vehicle (UAV) has become more and more important in both civil use and military operations. The overall reconnaissance capability of the UAV swarm is often affected by multiple signals. A new approach is proposed by recognizing data credibility (DC) using multiple machine learning (ML) techniques, i.e., a novel DCML approach. There are two major components (and major theoretical contributions) of the proposed approach. The first component is the initial identification of less-credible data using a single ML technique. The second component is the cross-identification of less-credible data using multiple ML techniques based on the initial identification results. A practical case is studied for validating the proposed DRML approach. Case study results show that (1) The proposed approach in this paper demonstrates a proficient ability to identify less credible data, (2) The validation with various machine learning methods proves effective, but the efficacy of the method is not necessarily proportional to the quantity of methods employed, (3) The combination of BPNN and GPR yields the most favorable outcomes.

Background Most infectious diseases are caused by viruses, fungi, bacteria and parasites. Their ability to easily infect humans and trigger large-scale epidemics makes them a public health concern. Methods for early detection of these diseases have been developed; however, they are hindered by the absence of a unified, interoperable and reusable model. This study seeks to create a holistic and real-time model for swift, preliminary detection of infectious diseases using symptoms and additional clinical data. Materials and methods In this study, we present a medical knowledge graph (MKG) that leverages multiple data sources to analyse connections between different nodes. Medical ontologies were used to enhance the MKG. We applied various graph algorithms to extract key features. The performance of multiple machine-learning (ML) techniques for influenza and hepatitis detection was assessed, selecting multi-layer perceptron (MLP) and random forest (RF) models due to their superior outcomes. The hyperparameters of both graph-based ML models were automatically fine-tuned. Results Both the graph-based MLP and RF models showcased the least loss and error rates, along with the most specific, accurate recall, precision and F1 scores. Their Matthews correlation coefficients were also optimal. When compared with existing ML techniques and findings from the literature, these graph-based ML models manifested superior detection accuracy. Conclusions The graph-based MLP and RF models effectively diagnosed influenza and hepatitis, respectively. This underlines the potential of graph data science in enhancing ML model performance and uncovering concealed relationships in the MKG.

This paper describes experiments using and comparing multiple machine learning (ML) algorithms for entity resolution (ER). In these experiments, person references were classified as "linked" or "not linked" by the four different methods. The objective of the experiments was to compare the linking performance of each method to evaluate the effectiveness of various ML techniques as an extension or augmentation to existing ER Systems. Each experiment used synthetic data. In this paper, some promising empirical results are reported that demonstrate favorable performance of the ML techniques.

BackgroundPragmatic language refers to using spoken language to convey messages effectively across diverse social communication contexts. However, minimal longitudinal research has focused on defining early predictors of pragmatic development in children with autism spectrum disorder (ASD).MethodsIn the present study, 71 children with ASD and 38 age- and gender- matched 24- to 30-month-old typically developing (TD) children were enrolled. Social-communication, language, and parent‒child interaction measures were collected for the ASD group at baseline. Three years later, all subjects were assessed for pragmatic ability via the Chinese version of the Language Use Inventory (LUI-Mandarin). First, the differences in pragmatic performance between the ASD and TD groups at follow-up were analyzed. Second, pragmatic performance was correlated with autism symptomatology at follow-up, as well as the structural language difficulties and joint engagement (JE) levels at baseline in the ASD group. Furthermore, hierarchical multiple regression analyses and machine learning techniques were performed to explore the effects of early potential predictors of pragmatic development in the ASD group.ResultsFirst, our results revealed that performance was significantly lower in the ASD than in the TD group with respect to the LUI-Mandarin total scores and subscale scores (except for subscale C). Second, correlation analysis revealed that more severe symptoms of ASD at follow-up were associated with lower LUI-Mandarin total scores and better language performance on the Gesell Developmental Schedules (GDS). Additionally, increased proportions of supported JE (SJE) states were associated with higher LUI-Mandarin total scores. In contrast, increased proportions of unengaged (UE) states were associated with lower LUI-Mandarin total scores in the ASD group. Third, hierarchical multiple regression analyses and machine learning techniques indicated that the proportions of SJE during parent‒child interactions, as well as the degree of social symptoms and structural language impairments, were significant contributors to pragmatic development for the ASD group in the prediction models.ConclusionIn summary, our findings suggest that pragmatic language difficulties are present in children with ASD as early as preschool age. Additionally, given the close correlation between the LUI-Mandarin score and symptom severity on the ADOS/ADI-R, the LUI-Mandarin might be a good way to triage children who need to wait a long time for a more extensive evaluation. Furthermore, this study provides new insights into potential targets for pragmatic interventions, and interventions can be designed to promote SJE between caregivers and children in future work.

At the core of enterprise networks lies routing protocols that make forwarding decisions based on a set of rules and metrics. One of the most popular and widely used routing protocols is the Open Shortest Path First (OSPF). Traditional OSPF calculates the cost of the route primarily based on interface bandwidth, without considering real-time factors such as latency, congestion, or link stability. These calculations are static and can lead to deficiencies in adapting to unstable network conditions. This study proposes the integration of multiple machine learning (ML) models and techniques to enhance OSPF routing decisions. Four important ML functions namely traffic forecast, anomaly detection, failure prediction, and dynamic cost optimization—have been used to improve OSPF performance. ML methods such as Random Forest and XGBoost are used to predict and assign costs in traffic utilization and real-time performance assessments. AutoRegressive Integrated Moving Average models and Long Short-Term Memory are applied to enable traffic predictions and route adjustments before potential congestions. Furthermore, link and node failure are common in network routing. Random Forest and logistic regression models are employed to predict these. The simulation took place in Graphical Network Simulator-3 using Cisco routers and Linux servers to allow thorough testing before and after applying the ML models. The results and findings have shown that the integration of ML models reroutes the traffic to enhance latency and throughput by approximately 30%. The findings demonstrate the upside of ML-enhanced OSPF routing as a versatile and scalable solution for high-demand networks.

Autism spectrum disorder (ASD) is a multifaceted neurodevelopmental condition that affects millions of individuals globally. This study leverages the publicly available UCI Machine Learning ASD Screening Dataset, which contains behavioral questionnaire responses from children aged 4-11 years, to develop and compare multiple machine learning (ML) models for early ASD diagnosis. This study applies SHAP (SHapley Additive exPlanations) to interpret and analyze the feature importance of various attributes in the dataset. Advanced ML models like Extreme Gradient Boosting (XGBoost), Random Forest (RF), Extra Trees Classifier (ETC), Support Vector Machines (SVM), and Multilayer Perceptrons (MLPs) were implemented to evaluate and compare their performance in predicting early ASD. The developed ASD model underwent a rigorous evaluation through comprehensive performance metrics, including accuracy, precision, recall, F1-score, specificity, and Matthews correlation coefficient. Ranking-based metrics, such as area under the receiver operating characteristic curve (AUROC), provided additional insights into classification robustness, while 10-fold cross-validation ensured reliability and generalizability. Results indicate that XGBoost achieved the highest diagnostic performance (accuracy = 97.8%, AUROC = 0.99), followed closely by MLP, with RF and ETC offering balanced performance and interpretability. SHAP analysis identified key behavioral features influencing model predictions, enabling explainable AI-driven insights. These findings suggest that ML-based tools can support clinicians in identifying ASD earlier, allowing for timelier interventions in disability services, particularly in underserved populations. While results are promising, the use of secondary data and the lack of clinical validation remain limitations, highlighting the need for real-world pilot testing. These findings underscore the transformative potential of ML techniques in enhancing ASD diagnostics. Despite their promise, the results are constrained by reliance on secondary data and the absence of clinical validation, underscoring the need for real-world pilot testing. Nonetheless, incorporating these models into clinical practice could accelerate early detection, facilitate personalized interventions, and optimize resource allocation for improved ASD management outcomes.

This paper presents an investigation into the modeling of Gallium Nitride (GaN) High Electron Mobility Transistors (HEMTs) using multiple Machine Learning (ML) algorithms. Despite the documented use of various ML techniques, a thorough comparison and performance analysis under different operating conditions were lacking. This study fills this gap by conducting a rigorous evaluation of nine ML models using TCAD-generated data of Pseudomorphic AlGaN/InGaN/GaN HEMT. The research focuses on Small Signal Behavioral Modeling and examines regression techniques such as Multiple Linear Regression (MLR), Multivariate Linear Regression (MVLR), Ridge Regression (L2), Lasso Regression (L1), Elastic Net Regression (ENR), Decision Trees (DT), Random Forest (RF), Gradient Boosting Regression(GBR), and Support Vector Regression (SVR). These methods use biases, frequency, and device geometry as independent variables, with S-parameters being the dependent variables. K-fold cross-validation was employed to ensure model reliability and accuracy across diverse operating conditions. Results reveal that the RF, coupled with 10-fold cross-validation, exhibits superior performance giving 99.7% accurate results, with a Mean Squared Error (MSE) of 4.6375×10-5, and a coefficient of determination (R2) of 0.9977. Conversely, SVR, L1, and ENR fall short of expectations. This study underscores the significance of methodological advancements in ML-based GaN HEMT modeling and provides valuable insights for future research in this domain.

The number of Internet of Things (IoT) equipment and the data those devices produce have both increased significantly over the past few years. Due to their resource limitations, IoT network participants can be problematic, hence it’s important to integrate security on these devices. Attackers now have more reasons to target IoT devices due to this. It is crucial to develop strategies to counter such attacks and shield IoT devices from malfunction. In order to accurately detect attacks and abnormalities in the IoT environment, the performances of multiple machine learning (ML) approaches have been compared in this paper. Here Support Vector Machine (SVM), K-Nearest Neighbour (KNN) and Random Forest (RF) have been applied as ML approaches utilizing the most recent LUFlow dataset. The performance measures such as accuracy, recall, precision and f1-score have been calculated for the comparison of the above-mentioned ML techniques. The system found that RF had the highest accuracy, at 97.7%, compared to SVM and KNN, which had accuracy rates of 92.80% and 94.7%, respectively.

Predicting intraoperative bleeding in patients undergoing a hepatectomy using multiple machine learning and deep learning techniques

Background Children’s immune systems are particularly vulnerable to infections. In developing countries, malnutrition, diarrheal diseases, and acute respiratory infections (ARI) remain leading causes of illness and mortality among children. This study applied multiple machine learning (ML) techniques to identify key risk factors associated with ARI symptoms in Bangladesh. Methods Secondary data from the Bangladesh Demographic and Health Survey (BDHS) 2022 were analyzed to assess ARI risk factors. Feature selection was conducted using the SHAP algorithm, and six ML classifiers were trained and evaluated with 10-fold cross-validation. Model performance was measured using accuracy, sensitivity, specificity, precision, F1-score, G-mean, and ROC AUC. Results Among the classifiers, the Decision Tree (DT) model achieved the highest performance across several metrics, with accuracy, sensitivity, specificity, precision, F1-score, and G-mean all at 0.81, and an ROC AUC of 0.88. Random Forest (RF) and AdaBoost (AdaB) also demonstrated strong performance, with RF showing an accuracy of 0.79 and ROC AUC of 0.89, and AdaB achieving an accuracy of 0.78 and ROC AUC of 0.86. Key predictive features included fever, age groups, geographic division, and area of residence. Conclusions The Decision Tree classifier outperformed other ML models in predicting ARI risk among children under five in Bangladesh, closely followed by Random Forest and AdaBoost. These findings highlight the potential of ML approaches to support targeted interventions by identifying critical risk factors. Government strategies should focus on early detection, improved treatment of fever, and enhanced household conditions to mitigate ARI risk in young children.

Crop seed yield modeling and prediction can act as a key approach in the precision agriculture industry, enabling the reliable assessment of the effectiveness of agro-traits. Here, multiple machine learning (ML) techniques are employed to predict sesame (Sesamum indicum L.) seed yields (SSY) using agro-morphological features. Various ML models were applied, coupled with the PCA (principal component analysis) method to compare them with the original ML models, in order to evaluate the prediction efficiency. The Gaussian process regression (GPR) and radial basis function neural network (RBF-NN) models exhibited the most accurate SSY predictions, with determination coefficients, or R2 values, of 0.99 and 0.91, respectfully. The root-mean-square error (RMSE) obtained using the ML models ranged between 0 and 0.30 t/ha (metric tons/hectare) for the varied modeling process phases. The estimation of the sesame seed yield with the coupled PCA-ML models improved the performance accuracy. According to the k-fold process, we utilized the datasets with the lowest error rates to ensure the continued accuracy of the GPR and RBF models. The sensitivity analysis revealed that the capsule number per plant (CPP), seed number per capsule (SPC), and 1000-seed weight (TSW) were the most significant seed yield determinants.

High penetration of Distributed Energy Resources (DERs), fundamental load behavior changes, controllable loads, and significant increase in Electrical Vehicles (EVs) lead to complex dynamic behavior of the electric distribution system. Increasing number of components also means more measurements, more data and more data anomalies. Detecting, classifying and localizing these anomalies are important for situational awareness, and at the same time, very challenging given increasing complexity of the system. Highly accurate and high-resolution analytical techniques are needed to support anomaly detection, classification and localization (AD-C-L) for monitoring, root cause analysis and decision making. This paper provides comprehensive review and analysis of the existing spatio-temporal AD-C-L techniques within the distribution system. Challenges for specific problems in AD-C- L have been also discussed in this paper. Existing AD-C- L techniques have been categorized and synthesized for specific merits and limitations of multiple Machine Learning (ML) methodologies using common developed metrics of performance. The comparative analysis is summarized and presented with the open research challenges and path forward for future research needs.

Current irrigation water use (IWU) estimation methods confront uncertainties warranting further attention, primarily stemming from constraints within model structure and data quality. This study proposes a hybrid framework that integrates multiple machine learning (ML) methods with theory‐guided strategies to calculate IWU for three principal cereal crops within the Chinese agricultural landscape. We generated high resolution time series data sets of evapotranspiration and surface soil moisture (SM) using remote sensing resources. ML techniques, along with the Bayesian three‐cornered hat ensemble, were employed to drive multiple remote sensing‐derived data sets in IWU calculation. We applied two theory‐guided mechanisms to quantify irrigation signals: first, converting original SM values into logarithmic terms, and second, extracting process‐based SM residuals. Proposed framework has been validated at 12 field stations across China, yielding coefficient of determination (R2) ranging from 0.54 to 0.70, and root mean square error (RMSE) spanning 278–335 mm/yr. Our framework demonstrates considerable strength in IWU estimation when compared to reported IWU values form 341 cities across China. Specifically, for rice, wheat, and maize, the R2 values range from 0.78 to 0.83, 0.68 to 0.76, and 0.53 to 0.64, respectively, with corresponding RMSE measuring 0.22–0.25, 0.10–0.12, and 0.11–0.13 km3/yr, respectively. These findings highlight the effectiveness of theory‐guided strategies in discerning irrigation‐related information, thereby improving overall model performance. Attention should be directed toward the uncertainties in evapotranspiration and precipitation products on model performance, which remained modest, with a relative change of less than 5%.

The Integrated Multisatellite Retrievals for Global Precipitation Measurement (GPM) (IMERG) Level 3 estimates rainfall from passive microwave sensors onboard satellites that are associated with several uncertainty sources such as sensor calibration, retrieval errors, and orographic effects. This study aims to provide a comprehensive investigation of multiple machine learning (ML) techniques (Random Forest, and Neural Networks), to stochastically generate an error-corrected improved IMERG precipitation product at a daily time scale and 0.1°-degree spatial resolution over the Brahmaputra river basin. In this study, we used the operational IMERG-Late Run version 06 product along with several meteorological and land surface parameters (elevation, soil type, land type, soil moisture, and daily maximum and minimum temperature) to produce an improved precipitation product in the Brahmaputra basin. We trained, tested, and optimized ML algorithms using 4 years (from 2015 through 2019) of reference rainfall data derived from the rain gauge. The ML generated precipitation product exhibited improved systematic and random error statistics for the study area, which is a strong indication for using the proposed algorithms in retrieving precipitation across the globe. We conclude that the proposed ML-based ensemble framework has the potential to quantify and correct the error sources for improving and promoting the use of satellite-based precipitation estimates for water resources applications.

This article uses a nation-wide household survey data from India and identifies important predictors of neonatal and infant mortality using multiple machine learning (ML) techniques. The consensus on the leading predictors from the interpretable ML algorithms (that we use) serve as early warning signs of neonatal and infant mortality. This enables us to identify a ‘high-mortality risk’ group of mothers and infants – an important goal of India’s ‘India Newborn Action Plan’. This high-risk group comprises firstborns, mothers with prior deaths or several previous births, newborns suffering from complicated deliveries, small size at birth and unvaccinated infants. We identify early newborn care, folic acid supplements and conditional cash transfer (Janani Suraksha Yojana) as the most effective policy interventions. Given the imbalanced nature of the dependent variable (‘events’ being rarer than ‘non-events’) we use additional ML methods (along with the commonly used ones) that are tailor-made for ‘rare-event’ prediction for robustness checks. We also use an evaluation measure called Area under Precision Recall Curve that is tailored for measuring prediction accuracy with imbalanced data. Our analysis sheds light on policy relevance and suggests some new policy prescriptions such as close monitoring of at-risks babies including females and those with small birth-size.