Input Predictors Research Articles

Prediction of Knee Joint Compartmental Loading Maxima Utilizing Simple Subject Characteristics and Neural Networks.

Joint loading may affect the development of osteoarthritis, but patient-specific load estimation requires cumbersome motion laboratory equipment. This reliance could be eliminated using artificial neural networks (ANNs) to predict loading from simple input predictors. We used subject-specific musculoskeletal simulations to estimate knee joint contact forces for 290 subjects during over 5000 stance phases of walking and then extracted compartmental and total joint loading maxima from the first and second peaks of the stance phase. We then trained ANN models to predict the loading maxima from predictors that can be measured without motion laboratory equipment (subject mass, height, age, gender, knee abduction-adduction angle, and walking speed). When compared to the target data, our trained models had NRMSEs (RMSEs normalized to the mean of the response variable) between 0.14 and 0.42 and Pearson correlation coefficients between 0.42 and 0.84. The loading maxima were predicted most accurately using the models trained with all predictors. We demonstrated that prediction of knee joint loading maxima may be possible without laboratory-measured motion capture data. This is a promising step in facilitating knee joint loading predictions in simple environments, such as a physician's appointment. In future, the rapid measurement and analysis setup could be utilized to guide patients in rehabilitation to slow development of joint disorders, such as osteoarthritis.

A Transformer‐Based Deep Learning Model for Successful Predictions of the 2021 Second‐Year La Niña Condition

AbstractA purely data‐driven and transformer‐based model with a novel self‐attention mechanism (3D‐Geoformer) is used to make predictions by adopting a rolling predictive manner similar to that in dynamical coupled models. The 3D‐Geoformer yields a successful prediction of the 2021 second‐year cooling conditions that followed the 2020 La Niña event, including covarying anomalies of surface wind stress and three‐dimensional (3D) upper‐ocean temperature, the reoccurrence of negative subsurface temperature anomalies in the eastern equatorial Pacific and a corresponding turning point of sea surface temperature (SST) evolution in mid‐2021. The reasons for the successful prediction with interpretability are explored comprehensively by performing sensitivity experiments with modulating effects on SST due to wind and subsurface thermal forcings being separately considered in the input predictors for prediction. A comparison is also conducted with physics‐based modeling, illustrating the suitability and effectiveness of 3D‐Geoformer as a new platform for El Niño and Southern Oscillation studies.

Estimating fall‐harvested alfalfa (Medicago sativa L.) yield using unmanned aerial vehicle–based multispectral and thermal images in southern California

AbstractThis study aims to evaluate the efficacy of simple linear, multiple, and robust regression methods to predict fall‐harvested alfalfa (Medicago sativa L.) yield using unmanned aerial vehicle (UAV)‐acquired multispectral and thermal images. Four alfalfa fields in southern California were selected, and a composite dataset containing 180 ground truth sampling points was formed to build and test the performance of the regression models. The UAV was flown in September 2020, 5–29 days before the ground truth data collection. A total of nine crop indices, canopy temperature, and the difference between canopy temperature and air temperature were used as input predictors. Among the simple linear models, the model with normalized difference vegetation index as input showed a strong performance (coefficient of determination [R2] = 0.76; root mean square error [RMSE] = 170.29 kg ha−1; and mean absolute error [MAE] = 132.18 kg ha−1). A multiple linear regression model with three input predictors showed the highest accuracy with R2 = 0.83, RMSE = 142.99 kg ha−1, and MAE = 109.30 kg ha−1. The top‐performing models accurately estimated mean yield at the field level and differentiated fields with low and high alfalfa productivity. Including canopy temperature‐related inputs did not improve the yield prediction power of the models. The error in the yield prediction increased as the days between UAV flights and field harvest increased. Results here suggested that UAV‐based remote sensing has the potential to estimate fall‐harvested alfalfa yield in southern California.

National-scale spatial prediction of soil organic carbon and total nitrogen using long-term optical and microwave satellite observations in Google Earth Engine

Modeling accurate and detailed soil spatial information is essential for environmental modeling, precision soil management and decision-making. In this study, we integrated long-term optical (Sentinel-2) and radar (Sentinel-1) satellite observations via the Google Earth Engine (GEE) platform for high-resolution national-scale digital mapping of soil organic carbon (SOC) and total soil nitrogen (TSN) in Austria. Our soil predictive models based on boosted regression tree (BRT) and regression kriging (RK) methods were constructed from 449 soil samples (0–20 cm) covering the study area in the LUCAS soil database and Sentinel observations synthesized with different time intervals. The different input predictors of these soil predictive models resulted in seven modeling scenarios, and their prediction performance was evaluated by a cross-validation technique. Comparative analysis indicated that satellite sensors, modeling techniques, and SAR data acquisition configurations greatly affected the model outputs. Cross-polarization and co-polarization had similar performance in TSN and SOC predictions, and their combination improved the prediction accuracy. Predictive models based on Sentinel-1 with the “ASCENDING” orbits outperformed the models involving the “DESCENDING” orbits; the prediction accuracy of the former was comparable to models involving two orbital data. The models built by Sentinel-1 and Sentinel-2 performed similarly in predicting SOC (R2 = 0.51 vs. R2 = 0.52, respectively) and TSN (their R2 were both 0.42); their synergistic utilization improved the prediction results. Models involving more years of Sentinel observations on the GEE platform provided more accurate modeling results. The best soil predictive models explained 55% and 45% of soil variability for SOC and TSN, respectively, both constructed from long-term Sentinel-1/2 observations using the RK method. The overall trends of the mapping results of the models constructed by Sentinel-1 and Sentinel-2 and their combinations were consistent. The predicted digital soil maps displayed high spatial heterogeneity: SOC and TSN—shared similar spatial patterns—were greater in high-altitude central and western regions than other regions. This study provides valuable information for revealing the effects of satellite sensors, modeling techniques and SAR configurations on mapping SOC and TSN.

An integrated approach based on the correction of imbalanced small datasets and the application of machine learning algorithms to predict total phosphorus concentration in rivers

Increased concentrations of Total Phosphorus (TP) in freshwater systems lead to eutrophication and can contribute to a wide range of environmental effects. In the modern era, water quality models have increasingly been used globally for the development of management scenarios with the aim of reducing the eutrophication risk. However, the accuracy of these models is limited by the quality of the boundary conditions forcing data, namely TP concentration datasets. In this study, a novel methodology is proposed to improve machine learning prediction accuracy in the modeling of river TP concentration forced with small input training datasets. These models can then be used to increase the quality and consistency of the TP concentration datasets required to force water quality models. This new methodology relies on the generation of 100 new training datasets from the raw training datasets of input predictors through the implementation of an over/undersampling technique. The modeling approach used in this study was supported by the application of ten machine learning algorithms to estimate the TP concentration values in 22 rivers located in Portugal. The modeling approach also included an input feature importance evaluation, as well as model hyperparameter optimization. In general terms, the Extreme Gradient Boosting (XGBoost) and Support Vector Regressor (SVR) models performed best overall, with the ensemble results recorded for both models working to increase the mean Nash-Sutcliffe efficiency (NSE) across all the areas being studied by 96% (0.01 ± 0.22 to 0.31 ± 0.32) and reduce the mean percentage bias (PBIAS) by 43% (18.47 ± 17.31 to 10.60 ± 17.40). The results of this study suggest that the solution proposed has the potential to significantly improve the modeling of TP concentration in rivers with machine learning methods, as well as providing increased scope for its application to larger training datasets and the prediction of other types of dependent variables. Hopefully, the results of this study will further add to the body of information available in this area of research and aid the development of the water management process.

A Fracture Risk Assessment Tool for High Resolution Peripheral Quantitative Computed Tomography.

Most fracture risk assessment tools use clinical risk factors combined with bone mineral density (BMD) to improve assessment of osteoporosis; however, stratifying fracture risk remains challenging. This study developed a fracture risk assessment tool that uses information about volumetric bone density and three-dimensional structure, obtained using high-resolution peripheral quantitative compute tomography (HR-pQCT), to provide an alternative approach for patient-specific assessment of fracture risk. Using an international prospective cohort of older adults (n = 6802) we developed a tool to predict osteoporotic fracture risk, called μFRAC. The model was constructed using random survival forests, and input predictors included HR-pQCT parameters summarizing BMD and microarchitecture alongside clinical risk factors (sex, age, height, weight, and prior adulthood fracture) and femoral neck areal BMD (FN aBMD). The performance of μFRAC was compared to the Fracture Risk Assessment Tool (FRAX) and a reference model built using FN aBMD and clinical covariates. μFRAC was predictive of osteoporotic fracture (c-index = 0.673, p < 0.001), modestly outperforming FRAX and FN aBMD models (c-index = 0.617 and 0.636, respectively). Removal of FN aBMD and all clinical risk factors, except age, from μFRAC did not significantly impact its performance when estimating 5-year and 10-year fracture risk. The performance of μFRAC improved when only major osteoporotic fractures were considered (c-index = 0.733, p < 0.001). We developed a personalized fracture risk assessment tool based on HR-pQCT that may provide an alternative approach to current clinical methods by leveraging direct measures of bone density and structure. © 2023 The Authors. Journal of Bone and Mineral Research published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research (ASBMR).

Analysis of the Factors Affecting Student Performance Using a Neuro-Fuzzy Approach

Predicting students’ academic performance and the factors that significantly influence it can improve students’ completion and graduation rates, as well as reduce attrition rates. In this study, we examine the factors influencing student academic achievement. A fuzzy-neural approach is adopted to build a model that predicts and explains variations in course grades among students, based on course category, student course attendance rate, gender, high-school grade, school type, grade point average (GPA), and course delivery mode as input predictors. The neuro-fuzzy system was used because of its ability to implicitly capture the functional form between the dependent variable and input predictors. Our results indicate that the most significant predictors of course grades are student GPA, followed by course category. Using sensitivity analysis, student attendance was determined to be the most significant factor explaining the variations in course grades, followed by GPA, with course delivery mode ranked third. Our findings also indicate that a hybrid course delivery mode has positively impacted course grades as opposed to online or face-to-face course delivery alone.

Ensemble robust local mean decomposition integrated with random forest for short-term significant wave height forecasting

A robust short-term significant wave height (Hs) modelling framework based on an ensemble local mean decomposition method integrated with random forest (i.e., En-RLMD-RF) is developed. The robust local mean decomposition (RLMD) decomposed the Hs data series into three subseries; amplitude modulation, frequency modulation and the low-frequency product function (PFs). The partial autocorrelation function was employed to determine the correlation-based significant predictor signals between the PFs at t0 and t1. Then the statistically significant PFs were incorporated into the random forest (RF) to construct the RLMD-RF model. The RLMD-RF based forecasted PFs were used again in the RF model as input predictors resulting in an ensemble-based RLMD-RF (i.e., En-RLMD-RF) model for forecasting short-term Hs. The En-RLMD-RF model is validated and compared with RF, extreme learning machine (ELM) and multiple linear regression (MLR) models and their hybrids RLMD-RF, RLMD-ELM, RLMD-MLR, En-RLMD-ELM and En-RLMD-MLR counterparts using a set of performance metrics. The results demonstrated that the En-RLMD-RF model generates better forecasting accuracy against the benchmarking models. This study is beneficial for the application and optimization of more clean energy resources worldwide for sustained energy generation.

Retrieval of Volcanic Ash Cloud Base Height Using Machine Learning Algorithms

There are distinct differences between radiation characteristics of volcanic ash and meteorological clouds, and conventional retrieval methods for cloud base height (CBH) of the latter are difficult to apply to volcanic ash without substantial parameterisation and model correction. Furthermore, existing CBH inversion methods have limitations, including the involvement of many empirical formulae and a dependence on the accuracy of upstream cloud products. A machine learning (ML) method was developed for the retrieval of volcanic ash cloud base height (VBH) to reduce uncertainties in physical CBH retrieval methods. This new methodology takes advantage of polar-orbit active remote-sensing data from the Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP), from vertical profile information and from geostationary passive remote-sensing measurements from the Spinning Enhanced Visible and Infrared Imager (SEVIRI) and the Advanced Geostationary Radiation Imager (AGRI) aboard the Meteosat Second Generation (MSG) and FengYun-4B (FY-4B) satellites, respectively. The methodology involves a statistics-based algorithm with hybrid use of principal component analysis (PCA) and one of four ML algorithms including the k-nearest neighbour (KNN), extreme gradient boosting (XGBoost), random forest (RF), and gradient boosting decision tree (GBDT) methods. Eruptions of the Eyjafjallajökull volcano (Iceland) during April-May 2010, the Puyehue-Cordón Caulle volcanic complex (Chilean Andes) in June 2011, and the Hunga Tonga-Hunga Ha’apai volcano (Tonga) in January 2022 were selected as typical cases for the construction of the training and validation sample sets. We demonstrate that a combination of PCA and GBDT performs more accurately than other combinations, with a mean absolute error (MAE) of 1.152 km, a root mean square error (RMSE) of 1.529 km, and a Pearson’s correlation coefficient (r) of 0.724. Use of PCA as an additional process before training reduces feature relevance between input predictors and improves algorithm accuracy. Although the ML algorithm performs well under relatively simple single-layer volcanic ash cloud conditions, it tends to overestimate VBH in multi-layer conditions, which is an unresolved problem in meteorological CBH retrieval.

Machine Learning Weather Analogs for Near-Surface Variables

Numerical weather prediction models and high-performance computing have significantly improved our ability to model near-surface variables, but their uncertainty quantification still remains a challenging task. Ensembles are usually produced to depict a series of possible future states of the atmosphere, as a means to quantify the prediction uncertainty, but this requires multiple instantiation of the model, leading to an increased computational cost. Weather analogs, alternatively, can be used to generate ensembles without repeated model runs. The analog ensemble (AnEn) is a technique to identify similar weather patterns for near-surface variables and quantify forecast uncertainty. Analogs are chosen based on a similarity metric that calculates the weighted multivariate Euclidean distance. However, identifying optimal weights for similarity metric becomes a bottleneck because it involves performing a constrained exhaustive search. As a result, only a few predictors were selected and optimized in previous AnEn studies. A new machine learning similarity metric is proposed to improve the theoretical framework on how weather analogs are identified. First, a deep learning network is trained to generate latent features using all the temporal multivariate input predictors. Analogs are then selected in this latent space, rather than the original predictor space. The proposed method does not require prior predictor selection and an exhaustive search, thus presenting a significant computational benefit and scalability. It is tested for surface wind speed and solar irradiance forecasts in Pennsylvania from 2017 to 2019. Results show that the proposed method is capable of handling a large number of predictors, and it outperforms the original similarity metric in RMSE, bias, and CRPS. Since the data-driven transformation network is trained using the historical record, the proposed method has been found to be more flexible for searching through a longer record.

Pre-surgical features of intrinsic brain networks predict single and joint epilepsy surgery outcomes

Despite the effectiveness of surgical interventions for the treatment of intractable focal temporal lobe epilepsy (TLE), the substrates that support good outcomes are poorly understood. While algorithms have been developed for the prediction of either seizure or cognitive/psychiatric outcomes alone, no study has reported on the functional and structural architecture that supports joint outcomes. We measured key aspects of pre-surgical whole brain functional/structural network architecture and evaluated their ability to predict post-operative seizure control in combination with cognitive/psychiatric outcomes. Pre-surgically, we identified the intrinsic connectivity networks (ICNs) unique to each person through independent component analysis (ICA), and computed: (1) the spatial–temporal match between each person’s ICA components and established, canonical ICNs, (2) the connectivity strength within each identified person-specific ICN, (3) the gray matter (GM) volume underlying the person-specific ICNs, and (4) the amount of variance not explained by the canonical ICNs for each person. Post-surgical seizure control and reliable change indices of change (for language [naming, phonemic fluency], verbal episodic memory, and depression) served as binary outcome responses in random forest (RF) models. The above functional and structural measures served as input predictors. Our empirically derived ICN-based measures customized to the individual showed that good joint seizure and cognitive/psychiatric outcomes depended upon higher levels of brain reserve (GM volume) in specific networks. In contrast, singular outcomes relied on systematic, idiosyncratic variance in the case of seizure control, and the weakened pre-surgical presence of functional ICNs that encompassed the ictal temporal lobe in the case of cognitive/psychiatric outcomes. Our data made clear that the ICNs differed in their propensity to provide reserve for adaptive outcomes, with some providing structural (brain), and others functional (cognitive) reserve. Our customized methodology demonstrated that when substantial unique, patient-specific ICNs are present prior to surgery there is a reliable association with poor post-surgical seizure control. These ICNs are idiosyncratic in that they did not match the canonical, normative ICNs and, therefore, could not be defined functionally, with their location likely varying by patient. This important finding suggested the level of highly individualized ICN’s in the epileptic brain may signal the emergence of epileptogenic activity after surgery.

Monthly Solar Prediction Using Machine Learning: Diyala Governorate, Iraq as a Case Study

Solar radiation constitutes the Earth’s primary energy source and is critical in regulating surface radiation equilibrium, vegetation photosynthesis, hydrological cycles, and extreme atmospheric. On the other hand, the depletion of global fossil fuel reserves mandates the power sector to adopt renewable energy-based sources, including photovoltaic and wind energy conversion systems. Therefore, the precise solar radiation prediction is imperative for climate research and the solar industry. This paper illustrates the use of two machine-learning approaches: random forest (RF) and support vector machine (SVM), to predict surface solar radiation in the Diyala governorate of Iraq for one step ahead, utilizing only lagged monthly time series data of the factor as input predictors. The findings were evaluated using three performance measures: coefficient of determination (R2), root mean square error (RMSE), and mean absolute error (MAE). The results showed that using 10 monthly lags time series as input predictors leads to the best prediction performance. Furthermore, in terms of the RMSE, the prediction performance of the RF algorithm was better than that of the SVM algorithm (RF's RMSE, MAE, and R2 were 181.398, 129.522, and 0.979, while for SVM were 240.149, 184.802, and 0.978, respectively).

Machine-learning approaches in geo-environmental engineering: Exploring smart solid waste management

Over the past few decades, increased attention has been paid to domestic waste (DW) generation. DW comprises a large percentage of municipal solid waste (MSW), and its handling and processing involves serious technical issues while also consuming a major portion of municipal budgets. The accurate estimation, prediction, and characterization of DW is an ongoing challenge for many cities, municipalities, and local governments as they strive to implement sustainable strategies for MSW. The main objective of the present study is to estimate and correctly predict DW quantities using machine-learning (ML) algorithms. Several different ML algorithms are used in the research, including linear regression, regression trees, Gaussian process regression, support vector machine, and autoregressive integrated moving average methods for time series analysis. Two case studies are presented in this paper. In the first, domestic waste data covering the period from 2010 to 2021 were collected from the Saudi and Bahrain authorities, and in the second, the domestic waste-generating behavior of a family of eleven members was followed for one month. The results show that the biodegradable and non-biodegradable wastes generated by the family were in the range of 1.7–7.9 kg and 0.0–2.0 kg, respectively, and promising outcomes were obtained using an appropriate selection of input predictors in conjunction with time series analysis. The trained models are validated and tested using several types of evaluation metrics, including calculated residuals, mean square error, root mean square error, and coefficient determination (R2-Score). The latter values are in the range of 0.67–0.85 for the training and testing datasets for many of the predicted waste quantities. The results obtained from the study show that these algorithms can be used to reduce the environmental, economic, and societal impacts of waste by designing a smart waste management engineering system.

Machine learning-based predictions of gamma passing rates for virtual specific-plan verification based on modulation maps, monitor unit profiles, and composite dose images

Machine learning (ML) methods have been implemented in radiotherapy to aid virtual specific-plan verification protocols, predicting gamma passing rates (GPR) based on calculated modulation complexity metrics because of their direct relation to dose deliverability. Nevertheless, these metrics might not comprehensively represent the modulation complexity, and automatically extracted features from alternative predictors associated with modulation complexity are needed. For this reason, three convolutional neural networks (CNN) based models were trained to predict GPR values (regression and classification), using respectively three predictors: (1) the modulation maps (MM) from the multi-leaf collimator, (2) the relative monitor units per control point profile (MUcp), and (3) the composite dose image (CDI) used for portal dosimetry, from 1024 anonymized prostate plans. The models’ performance was assessed for classification and regression by the area under the receiver operator characteristic curve (AUC_ROC) and Spearman’s correlation coefficient (r). Finally, four hybrid models were designed using all possible combinations of the three predictors. The prediction performance for the CNN-models using single predictors (MM, MUcp, and CDI) were AUC_ROC = 0.84 ± 0.03, 0.77 ± 0.07, 0.75 ± 0.04, and r = 0.6, 0.5, 0.7. Contrastingly, the hybrid models (MM + MUcp, MM + CDI, MUcp+CDI, MM + MUcp+CDI) performance were AUC_ROC = 0.94 ± 0.03, 0.85 ± 0.06, 0.89 ± 0.06, 0.91 ± 0.03, and r = 0.7, 0.5, 0.6, 0.7. The MP, MUcp, and CDI are suitable predictors for dose deliverability models implementing ML methods. Additionally, hybrid models are susceptible to improving their prediction performance, including two or more input predictors.

Detailed Geogenic Radon Potential Mapping Using Geospatial Analysis of Multiple Geo-Variables—A Case Study from a High-Risk Area in SE Ireland

A detailed investigation of geogenic radon potential (GRP) was carried out near Graiguenamanagh town (County Kilkenny, Ireland) by performing a spatial regression analysis on radon-related variables to evaluate the exposure of people to natural radiation (i.e., radon, thoron and gamma radiation). The study area includes an offshoot of the Caledonian Leinster Granite, which is locally intruded into Ordovician metasediments. To model radon release potential at different points, an ordinary least squared (OLS) regression model was developed in which soil gas radon (SGR) concentrations were considered as the response value. Proxy variables such as radionuclide concentrations obtained from airborne radiometric surveys, soil gas permeability, distance from major faults and a digital terrain model were used as the input predictors. ArcGIS and QGIS software together with XLSTAT statistical software were used to visualise, analyse and validate the data and models. The proposed GRP models were validated through diagnostic tests. Empirical Bayesian kriging (EBK) was used to produce the map of the spatial distribution of predicted GRP values and to estimate the prediction uncertainty. The methodology described here can be extended for larger areas and the models could be utilised to estimate the GRPs of other areas where radon-related proxy values are available.

Cross-track infrared sounder cloud fraction retrieval using a deep neural network

The retrieval of cloud fraction in satellite hyperspectral sounder field of view (FOV) is crucial for numerical weather prediction. This study proposes an innovative cloud fraction retrieval model for the hyperspectral infrared sounder - Cross-track Infrared Sounder (CrIS). The model is trained with a deep neural network (DNN), using the CrIS radiation spectra as the predictors and Visible Infrared Imaging Radiometer Suite (VIIRS) cloud mask as the learning target. An ensemble of randomly selected CrIS and VIIRS data are collocated and used as the training dataset. An optimized 5-layer neural network is built to establish the relationship between the CrIS spectra and the cloud fraction calculated from the VIIRS cloud mask within the CrIS FOV. In order to reduce the number of input predictors to enhance the efficiency of the model, a principal component transformation is performed on the original CrIS spectra and only the top 77 principal component scores are adopted as the final predictors. In general, the cloud fraction retrieved from the proposed DNN model are consistent with truth values calculated from the VIIRS cloud mask product. Further analysis on use cases demonstrates a slightly better cloud retrieval result during the daytime than that of the nighttime, and ocean retrievals are more accurate than land retrievals. However, since the relationship between CrIS spectrum and the cloud fraction is nonlinear, the model tends to slightly overestimate the cloud fractions over low cloud coverage regions and underestimate the values over high cloud fraction areas. Even so, the proposed model can still be a useful tool for obtaining cloud fraction information from hyperspectral infrared sounders and has the potential to be used for the numerical weather prediction and climate models, as well as other cloud studies.

Introductory Engineering Mathematics Students’ Weighted Score Predictions Utilising a Novel Multivariate Adaptive Regression Spline Model

Introductory Engineering Mathematics (a skill builder for engineers) involves developing problem-solving attributes throughout the teaching period. Therefore, the prediction of students’ final course grades with continuous assessment marks is a useful toolkit for degree program educators. Predictive models are practical tools used to evaluate the effectiveness of teaching as well as assessing the students’ progression and implementing interventions for the best learning outcomes. This study develops a novel multivariate adaptive regression spline (MARS) model to predict the weighted score WS (i.e., the course grade). To construct the proposed MARS model, Introductory Engineering Mathematics performance data over five years from the University of Southern Queensland, Australia, were used to design predictive models using input predictors of online quizzes, written assignments, and examination scores. About 60% of randomised predictor grade data were applied to train the model (with 25% of the training set used for validation) and 40% to test the model. Based on the cross-correlation of inputs vs. the WS, 12 distinct combinations with single (i.e., M1–M5) and multiple (M6–M12) features were created to assess the influence of each on the WS with results bench-marked via a decision tree regression (DTR), kernel ridge regression (KRR), and a k-nearest neighbour (KNN) model. The influence of each predictor on WS clearly showed that online quizzes provide the least contribution. However, the MARS model improved dramatically by including written assignments and examination scores. The research demonstrates the merits of the proposed MARS model in uncovering relationships among continuous learning variables, which also provides a distinct advantage to educators in developing early intervention and moderating their teaching by predicting the performance of students ahead of final outcome for a course. The findings and future application have significant practical implications in teaching and learning interventions or planning aimed to improve graduate outcomes in undergraduate engineering program cohorts.

Investigating the Use of Machine Learning Methods in Direct Ridership Models for Bus Transit

This test paper develops and tests 13 direct ridership models (DRMs) for transit sketch planning the Dallas–Fort Worth region. We explore both, machine learning modeling approaches (e.g., ridge regression and random forest) and traditional statistical models (e.g., linear regression and multiplicative regression). This effort provides a detailed description of modeling workflows and of the preprocessing of input data including general transit feed specification (GTFS), employment, socio-demographic, and ridership data. We also describe metrics to compare model performance; in our experiments the ridge regression framework using a Yeo-Johnson power transformation led to the most accurate predictions with an [Formula: see text] of 0.88. The sensitivity of the DRM model to errors in the service-related predictor variables is within acceptable limits with the root mean squared error (RMSE) increasing by less than 20% for a 25% error in any one of the input predictors. Our findings suggest that DRMs can be a powerful complement to the four-step planning process, providing an alternative that is easier to maintain and run, and which may lead to more accurate ridership estimates given the limitations of transit modeling in traditional regional models. To illustrate the benefits of DRMs, this effort describes the deployment of trained models using a web-based framework which allows practitioners to obtain ridership estimates by drawing prospective routes on a map and providing a small number of service attributes as input.

Spatial mapping of landslide susceptibility in Jerash governorate of Jordan using genetic algorithm-based wrapper feature selection and bagging-based ensemble model

This study employs five genetic algorithm (GA)-based machine learning (ML) models, namely the Decision Tree (DT), k-Nearest Neighbors (kNN), NaïveBayes (NB), Support Vector Machine (SVM), and Extreme Learning Machine (ELM), to build a novel ensemble algorithm that is founded on the Bagging method for landslide susceptibility mapping (LSM) in Jerash Governorate, north of Jordan. The GA-based wrapper feature selection (FS) was done based on the five individual models and in the initial stages of modeling, an inquiry for the best feature for each of the five models was made. Finally, five hybrid models, namely DT-GA, kNN-GA, NB-GA, SVM-GA, and ELM-GA were constructed and combined to create Bagging-based ensemble model. The FS outcomes uncovered that rainfall depth, distance to roads, the Stream Power Index, the Normalized Difference Vegetation Index, slope, geology, and aspect are the most influential determinants of landslides. After the significant variables were identified, they were selected as input predictors and entered into the models. GA-based Bagging ensemble model with the area under the receiver operating characteristic curve (AUROC) of 0.85 achieved the highest accuracy in the validation run, followed by SVM-GA (AUROC = 0.80), NB-GA (AUROC = 0.76), DT-GA (AUROC = 0.72), kNN-GA (AUROC = 0.70), and ELM-GA (AUROC = 0.48).

A photovoltaic power prediction approach enhanced by feature engineering and stacked machine learning model

The rapid depletion of world reserves of fossil fuels escalates energy costs, raises concerns regarding energy supplies, and increases climate impacts. The deployment of renewable energy has therefore become a worldwide trend. Among the other known forms of energy, solar energy is unquestionably the cleanest. In an era of widespread deployment of solar photovoltaic plants, research and development activities have made significant progress in assessing and investigating the behavior of solar installations in order to increase their reliability; however, there is still a local component that has not been thoroughly considered in many climate areas, including the semi-arid one. Understanding the distinctive features of each environment as well as how they affect photovoltaic power is the first step towards developing intelligent and data-driven maintenance algorithms suited to the region’s environmental context. This study proposes a stacked machine learning model making hourly predictions of two PV systems varying in size and age. Three machine learning algorithms were compared with a baseline linear regression model and a reference physical model for the PV prediction. The algorithms were also evaluated in terms of the input predictors. The findings demonstrate that the stacked ensemble model outperforms the individual models chosen for comparison in both systems and offers new perspectives for improving PV prediction in this area.