Application Of Machine Learning Models Research Articles

Predicting housing sale prices in Germany by application of machine learning models and methods of data exploration

Predicting housing sale prices in Germany by application of machine learning models and methods of data exploration

Application of machine learning models in predicting discharge coefficient of side B-type piano key weir

Side weir is a hydraulic structure within a channel which is usually used to discharge excess water, to divert the flow, and to regulate water surface levels in rivers and irrigation and drainage networks. In general, piano key weirs (PKW) have been used as weirs perpendicular to the flow direction in straight channels. However, the use of the PKW as a side weir in the outer arch of the channels is a new approach to enhance the weir's performance. In this study, 289 tests were first performed on the B-type rectangular side piano key weir (RSPKW) at two arc angles of 30 and 120°. Then, Fuzzy Inference System (FIS), Adaptive Neuro-Fuzzy Inference System (ANFIS), ANFIS and Teaching Learning Based Optimization (TLBO), ANFIS and Grasshopper Optimization Algorithm (GOA), Extreme Learning Machine (ELM) and Outlier Robust ELM (ORELM) models were used to predict the weir discharge coefficient. The results showed that two optimization models of TLBO and GOA increased the accuracy of the ANFIS model. The results showed that the ANFIS-GOA model has accuracy of Root Mean Squared Error (RMSE) = 0.0361, Coefficient of determination (R2) = 0.9772, and Kling Gupta coefficient (KGE) = 0.9858. The ANFIS-TLBO, ANFIS, and FIS models were ranked, respectively. Also, the results showed that ELM and ORELM models have accuracy close to ANFIS-GOA and can be a suitable alternative for complex fuzzy models. According to the statistical analysis, it was found that the parameters of the ratio of weir height to flow depth at the upstream edge of weir (P/h1), arc angle (α), and the ratio of height of the foundation to the main channel width (pd/B) had the greatest role in the development of the models, respectively.

Leveraging experimental and computational tools for advancing carbon capture adsorbents research.

CO2 emissions have been steadily increasing and have been a major contributor for climate change compelling nations to take decisive action fast. The average global temperature could reach 1.5°C by 2035 which could cause a significant impact on the environment, if the emissions are left unchecked. Several strategies have been explored of which carbon capture is considered the most suitable for faster deployment. Among different carbon capture solutions, adsorption is considered both practical and sustainable for scale-up. But the development of adsorbents that can exhibit satisfactory performance is typically done through the experimental approach. This hit and trial method is costly and time consuming and often success is not guaranteed. Machine learning (ML) and other computational tools offer an alternate to this approach and is accessible to everyone. Often, the research towards materials focuses on maximizing its performance under simulated conditions. The aim of this study is to present a holistic view on progress in material research for carbon capture and the various tools available in this regard. Thus, in this review, we first present a context on the workflow for carbon capture material development before providing various machine learning and computational tools available to support researchers at each stage of the process. The most popular application of ML models is for predicting material performance and recommends that ML approaches can be utilized wherever possible so that experimentations can be focused on the later stages of the research and development.

Advancing credit risk assessment and financial decision-making: Integrating modern techniques and insights

Credit risk assessment and fraud detection are critical functions in the financial industry, necessary for ensuring the stability and integrity of financial institutions. Traditional approaches often struggle to accurately assess risk and detect fraudulent activities in a timely manner. However, the rise of machine learning has introduced powerful tools that leverage large datasets and advanced algorithms to improve these processes. This research paper investigates the application of machine learning techniques in credit risk assessment and fraud detection within financial transactions. The paper begins by emphasizing the importance of accurate risk assessment and fraud detection in the financial sector and introduces machine learning as a solution to the limitations of traditional methods. A thorough literature review examines existing methodologies, algorithms, and trends, highlighting the advancements in this field. The discussion covers data acquisition and preprocessing techniques, underscoring the necessity of clean and relevant data for effective model training. Additionally, feature engineering strategies are explored to extract valuable insights from financial transaction data, enhancing the predictive power of machine learning models. The research analyzes various machine learning algorithms, such as logistic regression, decision trees, random forests, support vector machines, and neural networks, as well as ensemble methods. Model evaluation metrics, including accuracy, precision, recall, and ROC-AUC, are employed to assess the performance of these algorithms. The paper presents case studies and experimental results to demonstrate the practical application of machine learning models in real-world scenarios, showcasing their effectiveness in improving credit risk assessment and fraud detection. Furthermore, the research explores challenges such as imbalanced datasets, model interpretability, and regulatory compliance, offering insights into potential solutions and future research directions. The integration of behavioral finance, Bayesian networks, and optimization methods is also discussed, highlighting how these modern approaches, combined with big data analytics, can enhance predictive accuracy and decision reliability. In conclusion, this research underscores the transformative potential of machine learning in credit risk assessment and fraud detection within financial transactions. By adopting advanced algorithms and data-driven approaches, financial institutions can significantly improve their risk management strategies, mitigate potential risks, and protect against fraudulent activities. This ultimately contributes to a more secure and resilient financial ecosystem, enabling institutions to maintain a competitive edge in an ever-evolving market.

Residual energy evaluation in vortex structures: On the application of machine learning models

Vortex structures are widely employed for energy dissipation in urban surface water conveyance systems. When transporting wastewater through these networks, a substantial amount of water energy is dissipated. The effectiveness of these structures is usually evaluated by their efficiency in dissipating energy. Recent literature reviews on vortex structures have emphasized that, despite numerous experimental studies aimed at assessing their hydraulic performance, a reliable mathematical model to predict the residual energy head ratio remains elusive. In this study, resilient numerical models employing Artificial Intelligence (AI) methodologies (such as non-parametric regression, decision trees, and ensemble learning) have been structured by reliable experimental tests. By analyzing the experiments, three primary factors, referred to as flow Froude number (Fr), the ratio of sump height (Hs) to shaft diameter (D), and the ratio of drop total height (L) to shaft diameter (D) were determined to estimate the residual energy head ratio. Through experimental study, the residual energy head ratio is computed as a ratio of downstream flow energy (E2) to upstream flow energy (E1) at vortex structure. During the training and testing phases of AI models, the results of statistical tests, serving as quantitative evaluations, have shown that ensemble learning models namely Adaptive Boosting (AdaBoost) and Categorical Boosting (CatBoost) models had higher level of efficiency in the E2/E1 predictions and followed by Model Tree (MT), K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost) and Multivariate Adaptive Regression Spline (MARS). Additionally, the second-order regression-based equation was obtained from Fully Factorial Method (FFM) which had lower level of precision (R = 0.8275, RMSE = 0.1156, and MAE = 0.0846) in the residual energy head ratio predictions when compared with all predictive AI models. Variations of three effective factors (i.e., Fr, L/D, Hs/D) versus the predicted E2/E1 ratios were in well agreement with observational tests. Moreover, the results of Sobol's index indicated that Fr number was determined as the most effective parameter in the evaluation of residual energy head ratio in the vortex structure.

Comparative analysis of machine learning techniques for accurate prediction of unfrozen water content in frozen soils

Unfrozen water content (UWC) plays a critical role in determining the thermal, hydraulic, and mechanical properties of frozen soils. Existing empirical, semi-empirical, and theoretical models for UWC estimation have limitations in terms of accuracy as well as generalizability. To address these challenges, the present study explored the application of six machine learning techniques to predict UWC in frozen soils: Random Forest (RF), eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), K-Nearest Neighbors (KNN), Support Vector Regression (SVR), and Backpropagation Neural Network (BPNN). Considering the UWC hysteresis phenomenon between the freezing and thawing processes, experimental UWC data collected from the literature were separated into two sub-datasets: freezing branch dataset (FBD) and thawing branch dataset (TBD). Based on that, a comprehensive framework integrating Bayesian optimization and 10-fold cross-validation was established to optimize the six models' hyperparameters and to evaluate their performance. The results highlighted significant variations in the predictive capability among the six machine learning models, with ensemble methods (i.e., RF, XGBoost, LightGBM) generally demonstrating superior accuracy. Feature importance analysis, robustness checks, and uncertainty quantification further elucidated the strengths and limitations of each model. The present study provides profound insights into the selection and application of machine learning models for accurately modeling the properties of frozen soils for cold regions science and engineering.

Application of Machine Learning Models for Short-term Drought Analysis Based on Streamflow Drought Index

Application of Machine Learning Models for Short-term Drought Analysis Based on Streamflow Drought Index

Short-Term forecasting of floating photovoltaic power generation using machine learning models

Floating photovoltaic (FPV) power generation requires accurate short-term forecasting to optimize operational efficiency and enhance grid integration. This study investigates the application of machine learning models for predicting FPV power generation using data from the Universiti Malaysia Pahang Al-Sultan Abdullah (UMPSA) solar installation, which has a capacity of 157.20 kWp. Data were collected at 15-minute intervals from January 15 to January 21, 2024, encompassing nine input features such as ambient temperature, transient horizontal irradiation, daily horizontal irradiation, AC voltages, and AC currents for phases A, B, and C, with the total active power in kW as the target variable. The dataset was divided into a training set (first five days) and a testing set (remaining two days), and five machine learning models—Neural Networks (NN), Random Forest (RF), Extreme Learning Machine (ELM), Support Vector Regression (SVR), and Long Short-Term Memory (LSTM)—were employed. The results indicate that the Neural Networks model consistently outperforms the other machine learning algorithms in terms of predictive accuracy. These findings underscore the efficacy of machine learning techniques in forecasting FPV power generation, which has significant implications for enhancing the operational efficiency and grid integration of floating solar installations.

Role of Machine Learning Assisted Biosensors in Point-of-Care-Testing For Clinical Decisions.

Point-of-Care-Testing (PoCT) has emerged as an essential component of modern healthcare, providing rapid, low-cost, and simple diagnostic options. The integration of Machine Learning (ML) into biosensors has ushered in a new era of innovation in the field of PoCT. This article investigates the numerous uses and transformational possibilities of ML in improving biosensors for PoCT. ML algorithms, which are capable of processing and interpreting complicated biological data, have transformed the accuracy, sensitivity, and speed of diagnostic procedures in a variety of healthcare contexts. This review explores the multifaceted applications of ML models, including classification and regression, displaying how they contribute to improving the diagnostic capabilities of biosensors. The roles of ML-assisted electrochemical sensors, lab-on-a-chip sensors, electrochemiluminescence/chemiluminescence sensors, colorimetric sensors, and wearable sensors in diagnosis are explained in detail. Given the increasingly important role of ML in biosensors for PoCT, this study serves as a valuable reference for researchers, clinicians, and policymakers interested in understanding the emerging landscape of ML in point-of-care diagnostics.

Application of Machine Learning in the field of Heart Disease Prediction and its Accuracy Study

The application of machine learning in the healthcare sector is a prominent area of research presently. Among these, there is significant potential in employing machine learning for predicting and treating heart disease. Hence, this study aims to explore the accuracy and application of machine learning models in predictive diagnosis and treatment of heart disease. This paper first presents commonly used machine learning algorithms within relevant fields, followed by gathering data and instances showcasing the use of machine learning in predicting and treating heart disease. Through data analysis, this study summarizes current cutting-edge development directions, trends, as well as the integration of machine learning algorithms into real medical processes. Finally, it concludes by summarizing the existing prospects and challenges faced by machine learning in predicting and treating heart disease. This paper can serve as a valuable source of inspiration and reference for researchers involved in related fields concerning heart disease prediction and treatment.

Machine learning models for predicting geomagnetic storms across five solar cycles using Dst index and heliospheric variables

This study aims to improve the understanding of geomagnetic storms by utilizing machine learning models and analyzing several heliophysical variables, such as the interplanetary magnetic field, proton density, solar wind speed, and proton temperature. Rather than relying on traditional correlation-based methods, we employ advanced machine learning techniques to examine the complex relationships between these factors and geomagnetic storms. Our analysis covers a large dataset spanning six solar cycles, including the current 25th cycle, to provide comprehensive insights into the dynamics of these storms.Our study highlights the significance of the interplanetary magnetic field as a key predictor of geomagnetic storms, challenging previous beliefs that primarily focused on sunspot activity. By using high-resolution data, we uncover new patterns and provide a more detailed analysis of the factors influencing geomagnetic storms. We emphasize the importance of considering a range of heliophysical variables, such as proton temperature and flow pressure, which offer new insights into the complex dynamics driving these storm events.The application of machine learning models, particularly Random Forest and Gradient Boosting, demonstrated superior predictive accuracy compared to traditional methods. Our results reveal that the Dst-index MIN, scalar B, and alpha/proton ratio are among the most influential factors, accounting for a significant portion of the prediction model’s accuracy. These findings underscore the utility of machine learning in identifying critical drivers of geomagnetic activity and enhancing forecast precision.Additionally, our research underscores the need for comprehensive models that can accurately predict geomagnetic storms by integrating various data sources. This machine learning approach not only improves predictive accuracy but also enhances our understanding of the underlying mechanisms of space weather. The insights gained from this study have important implications for both scientific research and practical applications, such as improving early warning systems for geomagnetic storms and mitigating their potential impacts on Earth.

A Deep Learning Approach to Predict Recanalization First-Pass Effect following Mechanical Thrombectomy in Patients with Acute Ischemic Stroke.

Following endovascular thrombectomy in patients with large-vessel occlusion stroke, successful recanalization from 1 attempt, known as the first-pass effect, has correlated favorably with long-term outcomes. Pretreatment imaging may contain information that can be used to predict the first-pass effect. Recently, applications of machine learning models have shown promising results in predicting recanalization outcomes, albeit requiring manual segmentation. In this study, we sought to construct completely automated methods using deep learning to predict the first-pass effect from pretreatment CT and MR imaging. Our models were developed and evaluated using a cohort of 326 patients who underwent endovascular thrombectomy at UCLA Ronald Reagan Medical Center from 2014 to 2021. We designed a hybrid transformer model with nonlocal and cross-attention modules to predict the first-pass effect on MR imaging and CT series. The proposed method achieved a mean 0.8506 (SD, 0.0712) for cross-validation receiver operating characteristic area under the curve (ROC-AUC) on MR imaging and 0.8719 (SD, 0.0831) for cross-validation ROC-AUC on CT. When evaluated on the prospective test sets, our proposed model achieved a mean ROC-AUC of 0.7967 (SD, 0.0335) with a mean sensitivity of 0.7286 (SD, 0.1849) and specificity of 0.8462 (SD, 0.1216) for MR imaging and a mean ROC-AUC of 0.8051 (SD, 0.0377) with a mean sensitivity of 0.8615 (SD, 0.1131) and specificity 0.7500 (SD, 0.1054) for CT, respectively, representing the first classification of the first-pass effect from MR imaging alone and the first automated first-pass effect classification method in CT. Results illustrate that both nonperfusion MR imaging and CT from admission contain signals that can predict a successful first-pass effect following endovascular thrombectomy using our deep learning methods without requiring time-intensive manual segmentation.

Using machine learning to classify the immunosuppressive activity of per- and polyfluoroalkyl substances

Per- and polyfluoroalkyl substances (PFASs), one of the persistent organic pollutants, have immunosuppressive effects. The evaluation of this effect has been the focus of regulatory toxicology. In this investigation, 146 PFASs (immunosuppressive or nonimmunosuppressive) and corresponding concentration gradients were collected from literature, and their structures were characterized by using Dragon descriptors. Feature importance analysis and stepwise feature elimination are used for feature selection. Three machine learning (ML) methods, namely Random Forest (RF), Extreme Gradient Boosting Machine (XGB), and Categorical Boosting Machine (CB), were utilized for model development. The model interpretability was explored by feature importance analysis and correlation analysis. The findings indicated that the three models developed have exhibited excellent performance. Among them, the best-performing RF model has an average AUC score of 0.9720 for the testing set. The results of the feature importance analysis demonstrated that concentration, SpPosA_X, IVDE, R2s, and SIC2 were the crucial molecular features. Applicability domain analysis was also performed to determine reliable prediction boundaries for the model. In conclusion, this study is the first application of ML models to investigate the immunosuppressive activity of PFASs. The variables used in the models can help understand the mechanism of the immunosuppressive activity of PFASs, allow researchers to more effectively assess the immunosuppressive potential of a large number of PFASs, and thus better guide environmental and health risk assessment efforts.

Forecasting infection fatality rate of COVID-19: measuring the efficiency of several hybrid models

The main goal of this paper is to delve into a crucial epidemiological metric the daily infection fatality rate in the context of the ongoing COVID-19 pandemic. The significance of understanding this metric lies in its potential to provide insights into the severity and impact of the virus on a daily basis. Methods: To achieve this overarching objective, we employ a comprehensive approach by applying various hybrid models that hybridize both machine learning and statistical techniques. In our pursuit of a deeper understanding, we leverage advanced machine learning algorithms, including Support Vector Machine and Random Forest. These techniques allow us to capture intricate patterns and relationships within the data, contributing to a more nuanced analysis of the infection fatality rate. The application of machine-learning models in epidemiological studies has gained prominence due to their ability to adapt to complex and evolving patterns inherent in infectious disease dynamics. Complementing our machine-learning arsenal, we integrate traditional statistical models such as ARIMA (AutoRegressive Integrated Moving Average), fractional ARIMA, and BATS (Bayesian Structural Time Series). Results. To assess the performance of these models, we employ key evaluation metrics, including Root Mean Squared Error (RMSE), Mean Squared Error (MSE), and Mean Absolute Error (MAE). These metrics serve as critical benchmarks, allowing us to quantify the accuracy and reliability of our models in predicting the daily infection fatality rate. A meticulous evaluation of model performance is crucial for ensuring the validity and of our findings. According to these measures, we see that hybrid models performed well especially ARIMA-RF model RMSE: 0.29, MSE: 0.084, MAE: 0.215 for the horizon 60 and for horizon 120 ARIMA-RF still the best performance, RMSE: 0.268, MSE: 0.071, MAE: 0.183, we get these results due to the capacity of this approach to handle complex patterns contrarily to other model ARIMA, BATS, RF and SVM. Conclusion. This work adopted this approach in order to build a model to predict infection fatality rate, we aspire to provide a nuanced understanding of the factors influencing the severity of the virus, ultimately contributing to the ongoing discourse on effective public health interventions and mitigation strategies.

Comparative study on the prediction of the unconfined compressive strength of the one-part geopolymer stabilized soil by using different hybrid machine learning models

With the development of green, low-carbon, and sustainable economic systems, the issues of high pollution and energy consumption in construction materials have become increasingly prominent. This study focuses on adopting one-part geopolymer (OPG) in soil stabilization for underground engineering, which exhibits environmental and low-carbon advantages. The unconfined compressive strength (UCS) serves as a crucial parameter for assessing stabilized soil’s performance. However, it is necessary to conduct a large number of experiments, inducing high costs and time consumption. In this study, one multiple linear regression model, one Decision Tree (DT) model, five ensemble machine learning (ML) models (i.e. Random Forest [RF], Extra Tree [ET], Gradient Boosting [GB], Gradient Boosting Decision Tree [GBDT], and Extreme Gradient Boosting [XGBoost]), and hybrid models of those single ensemble models with Particle Swarm Optimization (PSO) (i.e. PSO-RF, PSO-ET, PSO-GB, PSO-GBDT, and PSO-XGBoost) were adopted and compared to achieve better prediction on the UCS of the OPG-stabilized soil. Furthermore, the interpretable method including SHAP and PDP (1D and 2D), was employed to investigate the precise mechanisms by which input parameters influenced the output label. The results revealed that the multiple linear regression model delivered the lowest accuracy, and PSO-XGBoost and PSO-ET exhibited the best performance on the prediction of the UCS with R2 value of 0.9964 and 0.9928, respectively. In addition, Curing time exerted the most significant impact on the UCS, followed by FA/GGBFS, Molarity, Water/Binder, and NaOH/Precursor. Compared to the SHAP method, the PDP offered a more intuitive approach to reveal the relationship between the inputs and output. The outcome of the study shed new light on the application of ML models in the prediction of the OPG-stabilized soil’s performance in underground engineering.

An Investigation of Machine Learning Applications in the Financial Fraud Detection

Financial fraud presents itself in various forms, and it often involves intricate financial transaction networks, making it challenging to detect the perpetrators and identify the characteristics of the fraud. In recent years, machine learning has gained widespread application within the financial sector. Therefore, various financial fraud detection models have been developed based on diverse machine learning methodologies. In the method part, this paper provides an overview of the machine learning process and then discusses the application of machine learning models in various financial fraud scenarios. Financial fraud in the insurance field has been further refined into automobile and medical insurance fraud. In automobile insurance fraud detection, some studies applied implicit Naive Bayes Model to analyze observable features and estimate hidden variables, and some studies used resampler to solve data imbalance and adopted 7 kinds of machine learning models for analysis. In health insurance fraud detection, many studies train medical data with multiple models, including AdaBoost, Logistic Regression and Support Vector Machine. In credit card fraud detection, studies use five algorithms, including random forest and decision tree et al., and some construct a model combining Decision Tree (DT) and Logistic Regression (LR). In bank fraud detection, some studies introduce Value-at-Risk to combine it with machine learning algorithms, and some studies propose a decentralized model training method based on federated learning. In the discussion section, this paper addresses the current limitations of the research, such as poor interpretability, uneven distribution of data sets, and issues related to customer privacy, and proposes corresponding solutions.

Gap-filling techniques applied to the GOCI-derived daily sea surface salinity product for the Changjiang diluted water front in the East China Sea

Abstract. The spatial and temporal resolutions of contemporary microwave-based sea surface salinity (SSS) measurements are insufficient. Thus, we developed a gap-free gridded daily SSS product with higher spatial and temporal resolutions, which can provide information on short-term variability in the East China Sea (ECS), such as the front changes by Changjiang diluted water (CDW). Specifically, we conducted gap-filling for daily SSS products based on the Geostationary Ocean Color Imager (GOCI) with a spatial resolution of 1 km (0.01°), using a machine learning approach during the summer seasons from 2015 to 2019. The comparison of the Soil Moisture Active Passive (SMAP), Copernicus Marine Environment Monitoring Service (CMEMS), and Hybrid Coordinate Ocean Model (HYCOM) SSS products with the GOCI-derived SSS over the entire SSS range showed that the SMAP SSS was highly consistent, whereas the HYCOM SSS was the least consistent. In the < 31 psu range, the SMAP SSS was still the most consistent with the GOCI-derived SSS (R2=0.46; root mean squared error: RMSE = 2.41 psu); in the > 31 psu range, the CMEMS and HYCOM SSS products showed similar levels of agreement with that of the SMAP SSS. We trained and tested three machine learning models – the fine trees, boosted trees, and bagged trees models – using the daily GOCI-derived SSS as output, including the three SSS products, environmental variables, and geographical data. We combined the three SSS products to construct input datasets for machine learning. Using the test dataset, the bagged trees model showed the best results (mean R2=0.98 and RMSE = 1.31 psu), and the models that used the SMAP SSS as input had the highest level. For the dataset in the > 31 psu range, all the models exhibited similarly reasonable performances (RMSE = 1.25–1.35 psu). The comparison with in situ SSS data, time series analysis, and the spatial SSS distribution derived from models showed that all the models had proper CDW distributions with reasonable RMSE levels (0.91–1.56 psu). In addition, the CDW front derived from the model gap-free daily SSS product clearly demonstrated the daily oceanic mechanism during the summer season in the ECS at a detailed spatial scale. Notably, the CDW front in the zonal direction, as captured by the Ieodo Ocean Research Station (I-ORS), moved approximately 3.04 km d−1 in 2016, which is very fast compared with the cases in other years. Our model yielded a gap-free gridded daily SSS product with reasonable accuracy and enabled the successful recognition of daily SSS fronts at the 1 km level, which was previously not possible with ocean color data. Such successful application of machine learning models can further provide useful information on the long-term variation of daily SSS in the ECS. The gridded gap-free SSS dataset at 0.01°×0.01° spatial resolution is freely available at https://doi.org/10.22808/DATA-2023-2 (Shin et al., 2023).

Application of machine learning models for property prediction to targeted protein degraders

Machine learning (ML) systems can model quantitative structure-property relationships (QSPR) using existing experimental data and make property predictions for new molecules. With the advent of modalities such as targeted protein degraders (TPD), the applicability of QSPR models is questioned and ML usage in TPD-centric projects remains limited. Herein, ML models are developed and evaluated for TPDs’ property predictions, including passive permeability, metabolic clearance, cytochrome P450 inhibition, plasma protein binding, and lipophilicity. Interestingly, performance on TPDs is comparable to that of other modalities. Predictions for glues and heterobifunctionals often yield lower and higher errors, respectively. For permeability, CYP3A4 inhibition, and human and rat microsomal clearance, misclassification errors into high and low risk categories are lower than 4% for glues and 15% for heterobifunctionals. For all modalities, misclassification errors range from 0.8% to 8.1%. Investigated transfer learning strategies improve predictions for heterobifunctionals. This is the first comprehensive evaluation of ML for the prediction of absorption, distribution, metabolism, and excretion (ADME) and physicochemical properties of TPD molecules, including heterobifunctional and molecular glue sub-modalities. Taken together, our investigations show that ML-based QSPR models are applicable to TPDs and support ML usage for TPDs’ design, to potentially accelerate drug discovery.

Application of machine learning models to improve the prediction of pesticide photodegradation in water by ZnO-based photocatalysts

Pesticide pollution has been posing a significant risk to human and ecosystems, and photocatalysis is widely applied for the degradation of pesticides. Machine learning (ML) emerges as a powerful method for modeling complex water treatment processes. For the first time, this study developed novel ML models that improved the estimation of the photocatalytic degradation of various pesticides using ZnO-based photocatalysts. The input parameters encompassed the source of light, mass proportion of dopants to Zn, initial pesticide concentration (C0), pH of the solution, catalyst dosage and irradiation time. Additionally, physicochemical properties such as the molecular weight of the dopants and pesticides, as well as the water solubility of both dopants and pesticides, were considered. Notably, the numerical data were extracted from the literature via relevant tables (directly) or graphs (indirectly) using the web-based tool WebPlotDigitizer. Four ML models including multi-layer perceptron artificial neural network (MLP-ANN), particle swarm optimization-adaptive neuro fuzzy inference system (PSO-ANFIS), radial basis function (RBF), and coupled simulated annealing-least squares support vector machine (CSA-LSSVM) were developed. In comparison, RBF showed the best accuracy of modeling among all models, with the highest determination coefficient (R2) of 0.978 and average absolute relative deviation (AARD) of 4.80%. RBF model was effective in estimating the photocatalytic degradation of pesticides except for 2-chlorophenol, triclopyr and lambda-cyhalothrin, where CSA-LSSVM model demonstrated superior performance. Dichlorvos was completely degraded by ZnO photocatalyst under visible light. The sensitivity analysis by relevancy factor exhibited that light irradiation time and initial pesticide concentration were the most important parameters influencing photocatalytic degradation of pesticides positively and negatively, respectively. The new ML models provide a powerful tool for predicting pesticide degradation in wastewater treatment, which will reduce photochemical experiments and promote sustainable development.

Machine Learning in Cybersecurity: Advanced Detection and Classification Techniques for Network Traffic Environments

In the digital age, the integrity of business operations and the smoothness of their execution heavily depend on cybersecurity and network efficiency. The need for robust solutions to prevent cyber threats and enhance network functionality has never been more critical. This research aims to utilize machine learning (ML) techniques for the meticulous analysis of network traffic, with the dual goals of detecting anomalies and categorizing network activities to bolster security and performance. Employing a detailed methodology, this study begins with data preparation and progresses through to the deployment of advanced ML models, including logistic regression, decision trees, and ensemble learning techniques. This approach ensures the accuracy of the analysis and facilitates a nuanced understanding of network dynamics. Our findings indicate a notable enhancement in identifying network inefficiencies and in the more accurate classification of network traffic. The application of ML models significantly reduces network delays and bottlenecks by providing a strong defence strategy against cyber threats and network shortcomings, thereby improving user satisfaction, and boosting the organizational reputation as a secure and effective service layer. Conclusively, the research highlights the pivotal role of machine learning in network traffic analysis, offering innovative insights and fresh perspectives on anomaly detection and the identification of malicious activities. It lays a foundation for future explorations and acts as an evaluation benchmark in the fields of cybersecurity and network management.