Prediction Model Analysis Research Articles

A Study on the Application of Explainable AI on Ensemble Models for Predictive Analysis of Chronic Kidney Disease

A Study on the Application of Explainable AI on Ensemble Models for Predictive Analysis of Chronic Kidney Disease

Training Model of Students’ English Creative Thinking Ability Based on Big Data Analysis

In order to optimize the evaluation of students’ English creative thinking ability under the multimedia assisted teaching mode, a training model of students’ English Creative Thinking Ability Based on big data analysis is proposed. Establish a big data block fusion structure analysis model, analyze the fuzzy parameters of constraint indicators, obtain a quantitative table for the state allocation of evaluation constraint indicators, formulate an evaluation quantitative analysis scoring table, and use hierarchical structural reconstruction to construct the clustering model parameters for the evaluation of students’ English creative thinking ability training. Through the quantitative index feature analysis, the optimization design of teaching activities and teaching methods for the evaluation of the cultivation process of students’ English creative thinking ability are realized. In the multi-dimensional hierarchical structure parameter model, the parameter configuration of students’ English creative thinking ability training indicators is realized, and the evaluation characteristic index fusion clustering processing is carried out on the parameter configuration results to form a pair of classification prediction and index analysis model. According to the hierarchical distribution density and grid clustering of indicators, the evaluation of students’ English creative thinking ability under the multimedia-assisted teaching mode is realized. The results of empirical analysis show that the fuzzy parameter analysis ability of the constraint index of process evaluation using this method is strong, the evaluation results are accurate and reliable, and the reliability and confidence level of evaluation are improved.

Physics-Informed Neural Network Model for Predictive Risk Assessment and Safety Analysis

This paper presents a physics-informed neural network (PINN) designed to predict the future locations of both vehicles and pedestrians, providing critical insights into road safety risks. By forecasting potential trajectories of road users, the proposed model informs preemptive strategies to avoid accidents. The physics model incorporates the intelligent driver model for vehicles and the social force model for pedestrians. The stochastic nature of risk evaluation is addressed by probabilistically predicting future locations based on the expected distribution in a two-dimensional open space. The framework accurately assesses the risk by predicting the future locations of vehicles and pedestrians within a 2- to 4-s time frame with approximately 2% error rates. The risk evaluation performance of the proposed framework was tested by calculating the time to collision (TTC) between vehicles and pedestrians and analyzing traffic conflicts. Leveraging the probabilistic predictions, the TTC was evaluated stochastically using Monte Carlo simulations and the Kolmogorov–Smirnov test, enabling a more granular and effective traffic conflict analysis. The developed method demonstrated over 95% accuracy when evaluating potentially dangerous events occurring within 3 s or less, providing actionable insights for improving road safety. The framework was deployed in a real-world setting, demonstrating reliable and robust test results. This comprehensive approach is expected to pave the way for more effective risk evaluation and mitigation at intersections and on roads.

A Scalable Fog Computing Solution for Industrial Predictive Maintenance and Customization

This study presents a predictive maintenance system designed for industrial Internet of Things (IoT) environments, focusing on resource efficiency and adaptability. The system utilizes Nicla Sense ME sensors, a Raspberry Pi-based concentrator for real-time monitoring, and a Long Short-Term Memory (LSTM) machine-learning model for predictive analysis. Notably, the LSTM algorithm is an example of how the system’s sandbox environment can be used, allowing external users to easily integrate custom models without altering the core platform. In the laboratory, the system achieved a Root Mean Squared Error (RMSE) of 0.0156, with high accuracy across all sensors, detecting intentional anomalies with a 99.81% accuracy rate. In the real-world phase, the system maintained robust performance, with sensors recording a maximum Mean Absolute Error (MAE) of 0.1821, an R-squared value of 0.8898, and a Mean Absolute Percentage Error (MAPE) of 0.72%, demonstrating precision even in the presence of environmental interferences. Additionally, the architecture supports scalability, accommodating up to 64 sensor nodes without compromising performance. The sandbox environment enhances the platform’s versatility, enabling customization for diverse industrial applications. The results highlight the significant benefits of predictive maintenance in industrial contexts, including reduced downtime, optimized resource use, and improved operational efficiency. These findings underscore the potential of integrating Artificial Intelligence (AI) driven predictive maintenance into constrained environments, offering a reliable solution for dynamic, real-time industrial operations.

Multi-objective optimization of solar air heater having centerline perforated sine wave baffles using experimental and machine learning based approaches

ABSTRACT The efficiency of smooth surface solar air heaters (SAHs) is often limited by viscous boundary layer formation and aerodynamic constraints, resulting in suboptimal thermal performance. This study introduces a novel enhancement technique using centerline perforated sine wave baffles as turbulators on the SAH surface to improve heat transfer and reduce flow resistance. The current work deals with experimental analysis of solar air heater performance using centerline perforated sine wave baffles placed as turbulators over the surface of solar air heater. Experimental analyses are conducted by varying Reynolds number, relative baffle pitch, relative baffle height, angle of attack, and relative perforation diameter, ranging from 3000 to 18,000, 8 to 14, 0.4 to 0.55, 10° to 45°, and 0.1 to 0.2 respectively. Collected experimental data are utilized to develop predictive machine learning models for Nusselt number, friction factor, and thermo-hydraulic analysis. Several models including Linear Regression, Random Forest, K-Nearest Neighbors, Adaptive Boosting, and Extreme Gradient Boosting are employed, with their performance evaluated using Mean Absolute Error, Mean Squared Error, Root Mean Squared Error, and R2. Extreme Gradient Boosting emerges as the most effective predictive model with an R2 value of 98.2%, alongside Mean Absolute Error and Root Mean Squared Error values of 8.025 and 11.3 respectively. Multi-objective optimization techniques such as Particle Swarm Optimization, Simulated Annealing, Genetic Algorithm, and Hybrid Genetic Algorithm are employed for this purpose. The optimized parameters obtained from Genetic Algorithm include Reynolds number, and values of 17,447, 8.37, 0.48, 25.045, and 0.11 respectively, aiming at maximizing and while minimizing. The results of genetic algorithms are being taken as optimal parameters which resulted in Reynolds number, relative baffle pitch, relative baffle height, angle of attack, and relative perforation diameter as 17,447, 8.37, 0.48, 25.045 and 0.11.

Development and Validation of a Machine Learning-Based Early Warning Model for Lichenoid Vulvar Disease: Prediction Model Development Study.

Given the complexity and diversity of lichenoid vulvar disease (LVD) risk factors, it is crucial to actively explore these factors and construct personalized warning models using relevant clinical variables to assess disease risk in patients. Yet, to date, there has been insufficient research, both nationwide and internationally, on risk factors and warning models for LVD. In light of these gaps, this study represents the first systematic exploration of the risk factors associated with LVD. The risk factors of LVD in women were explored and a medically evidence-based warning model was constructed to provide an early alert tool for the high-risk target population. The model can be applied in the clinic to identify high-risk patients and evaluate its accuracy and practicality in predicting LVD in women. Simultaneously, it can also enhance the diagnostic and treatment proficiency of medical personnel in primary community health service centers, which is of great significance in reducing overall health care spending and disease burden. A total of 2990 patients who attended West China Second Hospital of Sichuan University from January 2013 to December 2017 were selected as the study candidates and were divided into 1218 cases in the normal vulvovagina group (group 0) and 1772 cases in the lichenoid vulvar disease group (group 1) according to the results of the case examination. We investigated and collected routine examination data from patients for intergroup comparisons, included factors with significant differences in multifactorial analysis, and constructed logistic regression, random forests, gradient boosting machine (GBM), adaboost, eXtreme Gradient Boosting, and Categorical Boosting analysis models. The predictive efficacy of these six models was evaluated using receiver operating characteristic curve and area under the curve. Univariate analysis revealed that vaginitis, urinary incontinence, humidity of the long-term residential environment, spicy dietary habits, regular intake of coffee or caffeinated beverages, daily sleep duration, diabetes mellitus, smoking history, presence of autoimmune diseases, menopausal status, and hypertension were all significant risk factors affecting female LVD. Furthermore, the area under the receiver operating characteristic curve, accuracy, sensitivity, and F1-score of the GBM warning model were notably higher than the other 5 predictive analysis models. The GBM analysis model indicated that menopausal status had the strongest impact on female LVD, showing a positive correlation, followed by the presence of autoimmune diseases, which also displayed a positive dependency. In accordance with evidence-based medicine, the construction of a predictive warning model for female LVD can be used to identify high-risk populations at an early stage, aiding in the formulation of effective preventive measures, which is of paramount importance for reducing the incidence of LVD in women.

Machine Learning in Financial Fraud Detection: New Models for Predictive Analysis and Mitigating Business Risks

Abstract This research investigates how machine learning (ML) algorithms can be applied to financial fraud detection as a way to improve predictive analysis and reduce business risk of fraudulent activities. Financial fraud is a major threat according to Association of Certified Fraud Examiners (2023) which may cost corporations and consumers more than $5 trillion. Many traditional fraud detection systems have relied on rule-based methods which can be restricted with predefined criteria and not able to adapt to evolving fraud patterns. For this research, we use advanced ML models such as Random Forest, Gradient Boosting, Neural Networks which are trained to analyze the large datasets for anomaly detection and to measure the predictive accuracy with real time analysis. In achieving such rates and with the objectives of accuracy, precision and recall met, we employ a dataset of over 1 million financial transactions from verified sources and apply algorithms which we then evaluate against the metrics. Most notably, we show that ML driven models integrate with conventional methods to reduce false positives by 30%, leading to operational efficiency and cost savings. In addition to providing academic knowledge by validating the robustness of ML techniques, this study provides actionable insights for financial institutions which are looking to implement scalable, data driven fraud prevention system. Through addressing technical and operation challenges, our research demonstrates the practicality of applying ML in lowering financial risk.

Understanding the effect of experimental variables on the recovery of 231Pa from siliceous cake using factorial design

AbstractFour-factor (oxalic acid concentration, contact time, temperature, ultra-sound), two-level experimental design was applied during the recovery of 231Pa from siliceous cake—a solid radioactive-waste, produced in monazite processing. Oxalic acid (0.5 to 1.1 M) had the highest positive (F = 1018.6) and ultrasound had the least (F = 33.0) impact on recovery. All inter-factor interactions were significant, as Fcalculated < Fcritical and P < 0.05. Enthalpy of protactinium-oxalate complexation was found to be − 24.8 kJ mol−1. The regression coefficient (0.9853), randomness and magnitude (< ± 2) of fitted residuals explicated the suitability of present statistical model for predictive analysis.

Identifying the Urban Development Pattern of Subway Cities Using Predictive Analysis

Starting from the hypothesis that there are structural differences in urban development between cities that have implemented underground transport and those that have not, the paper aims to provide a predictive analysis model to identify cities that show the pattern for the development of the subway network. In order to identify the predictive analysis model capable of classifying urban settlements, the predictive analysis part consisted of two stages. In the first stage, a Random Forest classification model will be estimated on a sample of the 27 capitals of the European Member States. Subsequently, based on this model, 9 cities in Romania will be classified from the perspective of the existence of the urban development pattern for the development of underground public transport. In order for the classification error of the predictive model to be as low as possible, variables from all areas specific to urban development (economic development, population density, circular economy, green city, digitalization, urban mobility) were used. The element of originality of the paper consists in calculating variables on the area of urban mobility, urban infrastructure and circular economy, using GIS coordinates. Thus, in addition to official international data sources, variables calculated by using geospatial technology (average distances, densities of green spaces, etc.) were included in this research. The main results of the research show that there are cities in Romania that present the same pattern of development similar to the capitals of the Member States of the European Union that have implemented underground public transport. Therefore, it can be said that for these cities it is necessary to implement this type of transport, the depreciation of the investment representing an easy process, which can be achieved in a medium or short term.

Optimization of Marginal Price Forecasting in Mexico through applying Machine Learning Models

The Local Marginal Price (LMP) represents the value of energy at a specific moment and location, and its proper management is crucial for the development of the country's strategic sectors. This study compares the ADR, RPSG, SARIMA, and LSTM-H models for predicting the LMP, achieving an approximate effectiveness of 88%. By implementing it in 28 nodes of the three interconnection systems (SIN, BCA, and BCS) in Mexico, the results of the enhanced LSTM network analysis are presented through sensitivity analysis and an ensemble with Prophet, yielding the following metrics: MAE: 0.0189, MSE: 0.0101, RMSE: 0.1007, and MAPE: 12.18, at node 05PAR-115 in Hidalgo del Parral, Chihuahua. This model can construct tree diagrams (ADR) that identify the critical variables for predicting the LMP of any node, significantly contributing to the accuracy of predictive analysis models.

A Short-term PV Power Prediction and Uncertainty Analysis Model Based on CEEMDAN and AHA-BP

The stochastic and intermittent nature of photovoltaic (PV) generation brings a series of scheduling problems to the power system. An effective prediction of PV power is essential to minimize the impact of uncertainty. Therefore, this paper presents an integrated prediction model with complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), the artificial hummingbird algorithm (AHA), and the BP neural network (BPNN) for predicting power generation from PV power plants, and a methodology for uncertainty analysis by using the nonparametric kernel density estimation (NPKDE). First, one month of PV power is decomposed into an array of components using CEEMDAN. Then, the weights and thresholds of the BPNN are optimized by using AHA. These components are trained using the BPNN. Finally, the final prediction results are obtained by superimposing the components, and NPKDE is employed to compute the probability density and confidence interval of the prediction error. The proposed prediction method demonstrates superior predictive performance in comparison with other models. Also, the NPKDE approach better describes the accuracy of the probability density distribution.

Sentiment analysis of Algerian Arabic dialect on social media Using Bi-LSTM recurrent neural networks

This paper presents a sentiment analysis approach using Bidirectional Long Short-Term Memory (Bi-LSTM) Recurrent Neural Networks to train predictive models for sentiment analysis on social media, particularly focusing on Algerian Arabic Dialect. The method leverages word-to-vector embedding for word representation and incorporates natural language understanding of emojis to improve semantic interpretation. The model achieves a high accuracy of 94%, demonstrating its effectiveness in analyzing sentiments in online discussions. The originality lies in applying Bi-LSTM to handle multilingual challenges on social platforms. The findings have practical implications for business, policymaking, and public sentiment evaluation, while also contributing positively to fostering informed online discourse.

Numerical analysis on performance evaluation of photovoltaic thermal systems using ternary hybrid nanofluids and forced convection around copper cylinders

This study evaluates the performance of a photovoltaic thermal (PV/T) system using forced convection and ternary hybrid nanofluids, comprising water with cobalt, zinc, and silver. The PV/T system includes a glass absorber, polycrystalline silicon, and a flow channel with eight copper cylinders, positioned at 20 %, 40 %, 60 %, and 80 % along the channel, both above and below the flow path for optimal thermal conductivity. The working fluid is modeled under turbulent, steady-state conditions using the κ−ε turbulence model.Numerical simulations via COMSOL Multiphysics 6.0 evaluate the system's thermal and fluid dynamics, focusing on local Nusselt numbers and photovoltaic cell thermal efficiency compared to a baseline. The investigation spans Reynolds numbers from 10,000 to 100,000, nanoparticle volume fractions of 1 %–20 %, and aspect ratios of 0.1–0.3. The aspect ratio in this study is defined as the ratio of the height of a square obstacle within the flow channel to the total height of the flow channel. A supervised machine learning framework develops predictive regression models for system analysis. Results show fluid force at the outlet increases by about 1200 % and 1300 % for aspect ratios of 0.1 and 0.3 as the Reynolds number rises. The Nusselt number enhances by 420 % and 466 % at a 20 % nanoparticle volume fraction for aspect ratios of 0.1 and 0.3, respectively, with a 9 % improvement in photovoltaic cell efficiency. This work uniquely applies machine learning to derive regression equations for fluid force, Nusselt number, and electrical efficiency, an approach not previously explored in laminar flow studies.

A Method for Predicting the Result of University Employment Education Transformation Based on Multi-Source Information Technology

In the data on the outcomes of higher education employment transformation, there is high correlation among multiple independent variables, leading to multicollinearity. If multicollinearity exists, it can complicate the establishment of predictive analysis models and the interpretation of relationships between variables for higher education employment transformation outcomes. The employment issues of college graduates have become a focal point of social concern. These issues not only reflect the employment status of college graduates but also indirectly indicate the quality of higher education and the level of talent cultivation. Therefore, a prediction method for the effectiveness of higher education employment transformation based on multi-source information technology is proposed. According to the theory of multi-source information technology, an employment data classification framework is constructed to integrate multi-source information. The matching operator function output values are calculated, weights are determined, and the final college student information collection results are output. The standard error between employment unit types and salary ranges is analyzed, clustering training samples are trained, and the distance from each employment information sample to the centroid is calculated to obtain preprocessing results of the transformation outcomes, thus avoiding the impact of multicollinearity on the results of higher education employment transformation. A model structure for analyzing factors influencing higher education employment is established, allowing for an understanding of the characteristics of different types of educational resources and enabling the prediction of the effectiveness of higher education employment transformation. Experimental results show that this method can accurately monitor the dynamic employment situation of university students. After applying this method, the standard error and probability error values are significantly reduced, proving to be significantly effective in predicting the outcomes of higher education employment transformation.

Machine learning-assisted risk evaluation of heavy metals in the Hainan gold mining region, China.

This study employed machine learning (ML) to thoroughly investigate the impact of informal mining activities on the distribution and pollution status of heavy metals in soils near private gold mines in Hainan Province, southern China, a region known for its ecological sensitivity and economic importance. By systematically collecting surface soil samples and samples at depths of 0.5-1m from 175 drilling sites, a comprehensive quantitative analysis was conducted on major heavy metal elements, including lead (Pb), copper (Cu), cadmium (Cd), nickel (Ni), mercury (Hg), chromium (Cr), arsenic (As), and zinc (Zn). Combined with evaluation methods such as the Pollution Load Index (PLI), Normalized Pollution Index (NIPI), and Ecological Risk Index (ERI), the study revealed a high level of soil pollution at informal mining sites. The findings indicated that the average concentrations of Pb, Cd, Hg, As, and Zn in surface soils significantly exceeded the background values for soils in China, with a pronounced positive correlation observed between these heavy metal elements in both surface and deep soil profiles (r > 0.5). Furthermore, leveraging the heavy metal content in surface soils and the constructed environmental indicators, the predictive accuracy for metal content in deep soils was found to range from R2 = 0.27 to 0.68, suggesting that informal mining activities have led to substantial variations in metal content across different soil profiles. Through the application of a random forest model for predictive analysis of the PLI, NIPI, and ERI, high prediction accuracy was achieved (R2 = 0.78, 0.86, and 0.60, respectively). The study demonstrates that informal mining activities not only elevate the risk of soil pollution but also alter the distribution patterns of heavy metals. Also, this study provides a crucial foundation for the scientific assessment of soil quality and potential environmental hazards, while also affirming the efficacy of ML techniques in forecasting soil quality parameters.

Energy-efficient and location-aware IoT and WSN-based precision agricultural frameworks

Precision agriculture has emerged as a promising approach to enhance crop yield, reduce environmental impact, and optimize resource utilization through advanced sensing and automation technologies. This paper proposes an energy-efficient and location-aware framework for Internet of Things (IoT) and Wireless Sensor Networks (WSN)-based precision agriculture systems. The framework leverages low-power wireless communication protocols, adaptive sensor scheduling, and location-based clustering algorithms to minimize energy consumption and prolong the network lifetime. Key features include real-time monitoring of soil moisture, temperature, humidity, and crop health through geographically distributed sensors, with automated decision-making for irrigation, fertilization, and pest control. The proposed framework also integrates machine learning models for predictive analysis and anomaly detection, enabling early identification of potential issues that could adversely affect crop productivity. Simulation results demonstrate a significant reduction in energy consumption and communication overhead, while maintaining high accuracy in environmental parameter monitoring and resource allocation. This framework offers a scalable and robust solution for implementing sustainable precision agriculture practices, particularly in remote and resource-constrained areas

401.7: Utilizing large language models for predictive analysis of intraoperative hemorrhage risks in lung transplant recipients.

401.7: Utilizing large language models for predictive analysis of intraoperative hemorrhage risks in lung transplant recipients.

Integrated machine learning models for predictive analysis of thermal and electrical power generation of a photo-thermal system at Catania, Italy

The Photovoltaic-Thermal (PV/T) system is designed for producing both electrical and thermal energy. Its efficiency is improved when temperature-sensitive PV cells are cooled using nanofluid coolants. However, the current models for predicting PV/T system performance with nanofluid cooling are limited. In order to fill this gap, this study aims to develop machine learning models to predict the electrical and thermal efficiencies of a water-based PV/T system. Three types of machine learning algorithms from the supervised learning method were selected in this study because of their ability to handle complex data and provide accurate predictions: the Multi-layer Perceptron (MLP), the Gradient Boosting Regressor (GBR) and the Light Gradient-Boosting Machine (LightGBM). Initially, the models are trained and validated using data consisting of 15,540 samples from a water-based PV/T system. On the basis of the results obtained, the MLP algorithm proved to be the best for the prediction of electrical energy, with an R2 value = 0.9906, against 0.9709 and 0.99 respectively for the other GBR and LGB algorithms, while the LGB algorithm proved effective for the prediction of thermal energy, with an R2 value = 0.983, against 0.983, 0.97 and 0.988 respectively for the GBR and LGB algorithms. Secondly, the best MLP and LGB models previously identified for the water-based PV/T system were adapted to predict, this time, the energy performance of an Ag/water nanofluid-based PV/T system. The results obtained show that these models are highly effective in predicting the electrical and thermal performance of the Ag/water nanofluid PV/T system, even in the absence of real data, making them very useful for real applications.

Development, Calibration and Validation of Time Series Analysis and Artificial Neural Network Joint Model for Urban Noise Prediction

Noise in large urban areas, which is mainly generated by road traffic and by the human activities carried out nearby and inside the area under study, is a relevant problem. The continuous exposure to high noise levels, in fact, can lead to several problems, largely documented in the scientific literature. The analysis and forecasting of the noise level in a given area are, then, fundamental for control and prevention, especially when field measurements present peculiar trends and slopes, which can be modeled with a Time Series Analysis approach. In this paper, a hybrid model is presented for the analysis and the forecasting of noise time series in urban areas: this technique is based on the application of a deterministic decomposition model followed in cascade by a predictor of the forecasting errors based on an artificial neural network. Two variants of the hybrid model have been implemented and presented. The time series used to calibrate and validate the model is composed of sound pressure level measurements detected on a busy road near the commercial port of an Italian city. The proposed hybrid model has been calibrated on a part of the entire time series and validated on the remaining part. Residuals and error analysis, together with a detailed statistical description of the simulated noise levels and error metrics describe in detail the method’s performances and its limitations.

Predictive Analytics in Consumer Behavior Forecasting: A Literature Review

This paper introduces non-experts to the process of using predictive analytics models to solve real-world classification and prediction problems. There are two types of models: decision trees and regression models. This article summarizes the previous understanding and application of predictive models and consumer behavior, and also introduces the better use of predictive analysis models, including the accuracy and efficiency of each model. Predictive analytics models are widely used in our lives. Anyone interested in either business or medical science can learn this technology to better explore different fields. The results of this paper suggest that people can further innovate such models, so as to make the use of methods simpler and more efficient.