Types Of Machine Learning Models Research Articles

Predictive value of machine learning on fracture risk in osteoporosis: a systematic review and meta-analysis

ObjectivesEarly identification of fracture risk in patients with osteoporosis is essential. Machine learning (ML) has emerged as a promising technique to predict the risk, whereas its predictive performance remains controversial....

Uncertainties of landslide susceptibility prediction: Influences of random errors in landslide conditioning factors and errors reduction by low pass filter method

In the existing landslide susceptibility prediction (LSP) models, the influences of random errors in landslide conditioning factors on LSP are not considered, instead the original conditioning factors are directly taken as the model inputs, which brings uncertainties to LSP results. This study aims to reveal the influence rules of the different proportional random errors in conditioning factors on the LSP uncertainties, and further explore a method which can effectively reduce the random errors in conditioning factors. The original conditioning factors are firstly used to construct original factors-based LSP models, and then different random errors of 5%, 10%, 15% and 20% are added to these original factors for constructing relevant errors-based LSP models. Secondly, low-pass filter-based LSP models are constructed by eliminating the random errors using low-pass filter method. Thirdly, the Ruijin County of China with 370 landslides and 16 conditioning factors are used as study case. Three typical machine learning models, i.e. multilayer perceptron (MLP), support vector machine (SVM) and random forest (RF), are selected as LSP models. Finally, the LSP uncertainties are discussed and results show that: (1) The low-pass filter can effectively reduce the random errors in conditioning factors to decrease the LSP uncertainties. (2) With the proportions of random errors increasing from 5% to 20%, the LSP uncertainty increases continuously. (3) The original factors-based models are feasible for LSP in the absence of more accurate conditioning factors. (4) The influence degrees of two uncertainty issues, machine learning models and different proportions of random errors, on the LSP modeling are large and basically the same. (5) The Shapley values effectively explain the internal mechanism of machine learning model predicting landslide susceptibility. In conclusion, greater proportion of random errors in conditioning factors results in higher LSP uncertainty, and low-pass filter can effectively reduce these random errors.

SVM Model Against Telecom Card Fraud Using GA Optimised Ten-Fold Cross-Testing

The increased number of payment methods also makes it easier for personal information to be stolen by criminals, and for criminals to take over financial payment accounts and steal money. With trillions of bank card transactions occurring every day, Credit Card Fraud Detection (CCFD) is a serious challenge, so this paper predicts "whether or not fraud occurs" by using six types of machine learning models. For problem 1, firstly, "mean, maximum, minimum, median, variance, standard deviation, quartile" are calculated for each indicator; secondly, data cleaning is carried out, and the data set is found to be free of missing values and outliers. Then the data preprocessing work was carried out, min_max normalisation and z-score standardisation were performed on the data. After that, correlation analysis was carried out, and the first four indicators were classified as negative indicators and the last three as positive indicators according to the characteristics of the indicators themselves. It can be found by calculating the Pearson correlation coefficient value after two data processing. Using the coefficient of variation method to calculate the weight of the seven "influence whether fraud" indicators. Finally, BP neural network model, decision tree model, random forest classification model, ELM model, SVM model, logistic regression model are established. For Problem 2, the four models constructed in Problem 1 are solved; to solve the BP neural network model: the data set is divided into training set and testing set according to the ratio of 6:4, and the sigmod function is used as the activation function. For BP neural network, "output >0.5" is recorded as 1, i.e. fraudulent behaviour; "output <0.5" is recorded as 0, i.e. non-fraudulent behaviour. Adjusting the learning rate and the number of iterations, the optimal average mean square error after optimal gradient descent is smaller. To solve the SVM model, the data set is divided into ten groups using the improved ten-fold cross-test, with one group as the training set and nine groups as the validation set, so as to obtain the model with the highest accuracy and the corresponding training data, and then the genetic algorithm is used to search for the optimisation of the kernel parameters in the SVM model on this basis. To solve the decision tree model, the training set and prediction set are divided into 7:3 and solved, and the number of leaf nodes is optimised. Solve the random forest classification model, divided into training set and prediction set according to 7:3 and solved, for similar accuracy choose the random forest classifier when the decision tree is less.

Prediction of Complications and Prognostication in Perioperative Medicine: A Systematic Review and PROBAST Assessment of Machine Learning Tools.

The utilization of artificial intelligence and machine learning as diagnostic and predictive tools in perioperative medicine holds great promise. Indeed, many studies have been performed in recent years to explore the potential. The purpose of this systematic review is to assess the current state of machine learning in perioperative medicine, its utility in prediction of complications and prognostication, and limitations related to bias and validation. A multidisciplinary team of clinicians and engineers conducted a systematic review using the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) protocol. Multiple databases were searched, including Scopus, Cumulative Index to Nursing and Allied Health Literature (CINAHL), the Cochrane Library, PubMed, Medline, Embase, and Web of Science. The systematic review focused on study design, type of machine learning model used, validation techniques applied, and reported model performance on prediction of complications and prognostication. This review further classified outcomes and machine learning applications using an ad hoc classification system. The Prediction model Risk Of Bias Assessment Tool (PROBAST) was used to assess risk of bias and applicability of the studies. A total of 103 studies were identified. The models reported in the literature were primarily based on single-center validations (75%), with only 13% being externally validated across multiple centers. Most of the mortality models demonstrated a limited ability to discriminate and classify effectively. The PROBAST assessment indicated a high risk of systematic errors in predicted outcomes and artificial intelligence or machine learning applications. The findings indicate that the development of this field is still in its early stages. This systematic review indicates that application of machine learning in perioperative medicine is still at an early stage. While many studies suggest potential utility, several key challenges must be first overcome before their introduction into clinical practice.

123 Benchmarking of Artificial Neural Network Models for Genomic Prediction of Quantitative Traits in Pigs

Abstract Artificial neural networks (ANN) are a type of machine learning model that has been applied to various genomic problems, with the ability to learn non-linear relationships and model high-dimensional data. ANNs also have the potential in genomic prediction by capturing the intricate relationship between genetic variants and phenotypes. However, there is currently a limited effort to investigate the performance and feasibility of ANNs for pig genomic predictions. In this study, we evaluated the predictive performance of TensorFlow’s ANN models with one-layer, two-layer, and three-layer structures (with zero, one, and two hidden layers, respectively), in comparison with five linear methods, including GBLUP, LDAK, BayesR, SLEMM and scikit-learn’s ridge regression using data of six quantitative traits including off-test body weight (WT), off-test back fat thickness (BF), off-test loin muscle depth (MS), number of piglets born alive (NBA), number of piglets born dead (NBD), and number of piglets weaned (NW). Furthermore, we assessed the computational efficiency of ANNs on both CPU and GPU. The benchmarking was based on cross-validations of 26,190 genotyped pigs. We employed hyperband tuning to optimize the hyper-parameters and select the best model among one-layer, two-layer, and three-layer structures. Results showed that the one-layer structure, which is equivalent to ridge regression, yielded the best performance comparable to that of GBLUP. Using the optimal hyper-parameters for two-layer and three-layer structures, ANNs underperformed GBLUP in terms of accuracy. Of the five linear methods, BayesR and SLEMM performed similarly and the best, followed by LDAK, scikit-learn’s ridge regression, and GBLUP. Moreover, SLEMM was the fastest, which completed training with 21k individuals and 30k SNPs in 2.6 minutes. Compared with CPUs, GPUs exhibited a comparable computational speed for one-layer ANN but offered significant gains in computational efficiency for multi-layer ANNs. Based on our analysis of optimal hyper-parameters for two-layer ANN with BF, we found that using a GPU can lead to a five-fold increase in processing speed compared with using a conventional CPU, but it is still slower than GBLUP. In addition, hyper-parameter tuning (particularly for L2 regularization and the number of dense units in hidden layers) is critical for improving the genomic prediction accuracy in pigs. In conclusion, we found ANN with up to three layers could not improve genomic predictions compared with routine linear methods for pig quantitative traits.

Predicting CPAP failure after less invasive surfactant administration (LISA) in preterm infants by machine learning model on vital parameter data: a pilot study

Objective. Less invasive surfactant administration (LISA) has been introduced to preterm infants with respiratory distress syndrome on continuous positive airway pressure (CPAP) support in order to avoid intubation and mechanical ventilation. However, after this LISA procedure, a significant part of infants fails CPAP treatment (CPAP-F) and requires intubation in the first 72 h of life, which is associated with worse complication free survival chances. The aim of this study was to predict CPAP-F after LISA, based on machine learning (ML) analysis of high resolution vital parameter monitoring data surrounding the LISA procedure. Approach. Patients with a gestational age (GA) <32 weeks receiving LISA were included. Vital parameter data was obtained from a data warehouse. Physiological features (HR, RR, peripheral oxygen saturation (SpO2) and body temperature) were calculated in eight 0.5 h windows throughout a period 1.5 h before to 2.5 h after LISA. First, physiological data was analyzed to investigate differences between the CPAP-F and CPAP-Success (CPAP-S) groups. Next, the performance of two types of ML models (logistic regression: LR, support vector machine: SVM) for the prediction of CPAP-F were evaluated. Main results. Of 51 included patients, 18 (35%) had CPAP-F. Univariate analysis showed lower SpO2, temperature and heart rate variability (HRV) before and after the LISA procedure. The best performing ML model showed an area under the curve of 0.90 and 0.93 for LR and SVM respectively in the 0.5 h window directly after LISA, with GA, HRV, respiration rate and SpO2 as most important features. Excluding GA decreased performance in both models. Significance. In this pilot study we were able to predict CPAP-F with a ML model of patient monitor signals, with best performance in the first 0.5 h after LISA. Using ML to predict CPAP-F based on vital signals gains insight in (possibly modifiable) factors that are associated with LISA failure and can help to guide personalized clinical decisions in early respiratory management.

Estimating Full Longwave and Shortwave Radiative Transfer with Neural Networks of Varying Complexity

Abstract Radiative transfer (RT) is a crucial but computationally expensive process in numerical weather/climate prediction. We develop neural networks (NN) to emulate a common RT parameterization called the Rapid Radiative Transfer Model (RRTM), with the goal of creating a faster parameterization for the Global Forecast System (GFS) v16. In previous work we emulated a highly simplified version of the shortwave RRTM only—excluding many predictor variables, driven by Rapid Refresh forecasts interpolated to a consistent height grid, using only 30 sites in the Northern Hemisphere. In this work we emulate the full shortwave and longwave RRTM—with all predictor variables, driven by GFSv16 forecasts on the native pressure–sigma grid, using data from around the globe. We experiment with NNs of widely varying complexity, including the U-net++ and U-net3+ architectures and deeply supervised training, designed to ensure realistic and accurate structure in gridded predictions. We evaluate the optimal shortwave NN and optimal longwave NN in great detail—as a function of geographic location, cloud regime, and other weather types. Both NNs produce extremely reliable heating rates and fluxes. The shortwave NN has an overall RMSE/MAE/bias of 0.14/0.08/−0.002 K day−1 for heating rate and 6.3/4.3/−0.1 W m−2 for net flux. Analogous numbers for the longwave NN are 0.22/0.12/−0.0006 K day−1 and 1.07/0.76/+0.01 W m−2. Both NNs perform well in nearly all situations, and the shortwave (longwave) NN is 7510 (90) times faster than the RRTM. Both will soon be tested online in the GFSv16. Significance Statement Radiative transfer is an important process for weather and climate. Accurate radiative transfer models exist, such as the RRTM, but these models are computationally slow. We develop neural networks (NNs), a type of machine learning model that is often computationally fast after training, to mimic the RRTM. We wish to accelerate the RRTM by orders of magnitude without sacrificing much accuracy. We drive both the NNs and RRTM with data from the GFSv16, an operational weather model, using locations around the globe during all seasons. We show that the NNs are highly accurate and much faster than the RRTM, which suggests that the NNs could be used to solve radiative transfer inside the GFSv16.

Identification of OSAHS patients based on ReliefF-mRMR feature selection.

Obstructive Sleep Apnea Hypopnea Syndrome (OSAHS) is a serious chronic sleep disorder. Snoring is a common and easily observable symptom of OSAHS patients. The purpose of this work is to identify OSAHS patients by analyzing the acoustic characteristics of snoring sounds throughout the entire night. Ten types of acoustic features, such as Mel-frequency cepstral coefficients (MFCC), linear prediction coefficients (LPC) and spectral entropy among others, were extracted from the snoring sounds. A fused feature selection algorithm based on ReliefF and Max-Relevance and Min-Redundancy (mRMR) was proposed for optimal feature set selection. Four types of machine learning models were then applied to validate the effectiveness of OSAHS patient identification. The results show that the proposed feature selection algorithm can effectively select features with high contribution, including MFCC and LPC. Based on the selected top-20 features and using a support vector machine model, the accuracies in identifying OSAHS patients under the thresholds of AHI = 5,15, and 30, were 100%, 100%, and 98.94%, respectively. This indicates that the proposed model can effectively identify OSAHS patients.

Advanced Design of a III‐Nitride Light‐Emitting Diode via Machine Learning

AbstractGallium nitride (GaN)‐based light‐emitting diodes (LEDs) have obtained great market success in the past 20 years. However, the traditional research paradigm, i.e., experimental trial‐and‐error method, no longer adapts to the industry development. In this work, an efficient approach is demonstrated to design and optimize GaN‐based LED structures via machine learning (ML). By using the dataset of GaN‐based LED structures over the past decade to train four typical ML models, it is found that the convolutional neural network (CNN) provides the most accurate prediction, with a root mean square error (RMSE) of 1.03% for internal quantum efficiency (IQE) and 11.98 W cm−2 for light output power density (LOPD). Based on the CNN model, 1) the feature importance analysis is adopted to reveal the critical features for LED performance; 2) the predicted trends of IQE and LOPD match well with the physical mechanism, being consistent with the experimental and simulation results; and 3) a high‐throughput screening is demonstrated to predict the properties of over 20 000 structures within seconds to obtain high efficiency LED structures. This ML‐based LED design method enables direct guiding of the LED structure optimization in terms of key parameter selection during manufacturing and greatly accelerates the development cycle of GaN‐based LEDs.

Prediction of thermal comfort indoors and cooling loads based on reasonable zoning using the improved HHO with multi-strategy fusion-FENN algorithm

A comfortable indoor thermal environment helps occupants work healthily and efficiently, and the air-conditioning system, as the most energy-consuming system in building operation, accurate energy modeling while balancing thermal comfort can provide effective control strategies for managers. Metaheuristic algorithms are widely optimized for various types of machine learning models for the study of data-driven. However, the majority of them currently use the default hyperparameter or random and grid search, which are prone to obtaining inaccurate prediction results. In this paper, a sizable commercial building in Shaanxi Province is used as a research sample. The regional per floor firstly is divided with business model distribution, the cross-validation of random forest and correlation analysis is used to extract features, and the input factors of the prediction model are decided, and then the fully elman neural network is optimized using a segmental fusion strategy combining Cauchy operator and reverse learning to promote harris hawks optimization algorithm, and the prediction model the IMFHHO-FENN (improved Harris Hawks Optimization with multi-strategy fusion-Fully Elman Neural Network) is trained and constructed, which is finally used for predicting thermal comfort and cooling load. According to verification results, it can be proved that the IMFHHO-FENN model has excellent performance compared to mainstream single and combined models and can provide accurate results of predicted models in the design phase of building energy efficiency and later operation management, reducing energy consumption waste while satisfying personnel thermal comfort as much as possible and supply marked theoretical and decision reference for building equipment management.

Techniques and Applications for Deep Learning: A Review

Deep learning is a branch of machine learning that focuses on the development and refinement of complex neural networks for data analysis, prediction, and decision-making. Deep learning models use numerous layers of artificial neurons to automatically extract important features from raw data, making them superior at many tasks to typical machine learning models. Deep learning models' success in these fields has enhanced state-of-the-art performance and created new research and application prospects. Deep learning has been popular due to its capacity to tackle complicated issues in computer vision, natural language processing, speech recognition, and decision-making. In this study, we discuss deep learning techniques and applications, including recurrent neural networks, long short-term memory, convolutional neural networks, generative adversarial networks, and autoencoders. We also demonstrate deep learning's use in various fields. Deep learning has transformed artificial intelligence by enabling computers to learn from enormous datasets and accomplish complex tasks. As a result, scientists and engineers in fields as diverse as medicine, farming, manufacturing, and transportation have increased their focus on developing deep-learning methods and software. Current research trends and potential future paths in deep learning are also highlighted.

Eucalyptus Plantation Area Extraction Based on SLPSO-RFE Feature Selection and Multi-Temporal Sentinel-1/2 Data

An accurate and efficient estimation of eucalyptus plantation areas is of paramount significance for forestry resource management and ecological environment monitoring. Currently, combining multidimensional optical and SAR images with machine learning has become an important method for eucalyptus plantation classification, but there are still some challenges in feature selection. This study proposes a feature selection method that combines multi-temporal Sentinel-1 and Sentinel-2 data with SLPSO (social learning particle swarm optimization) and RFE (Recursive Feature Elimination), which reduces the impact of information redundancy and improves classification accuracy. Specifically, this paper first fuses multi-temporal Sentinel-1 and Sentinel-2 data, and then carries out feature selection by combining SLPSO and RFE to mitigate the effects of information redundancy. Next, based on features such as the spectrum, red-edge indices, texture characteristics, vegetation indices, and backscatter coefficients, the study employs the Simple Non-Iterative Clustering (SNIC) object-oriented method and three different types of machine-learning models: Random Forest (RF), Classification and Regression Trees (CART), and Support Vector Machines (SVM) for the extraction of eucalyptus plantation areas. Each model uses a supervised-learning method, with labeled training data guiding the classification of eucalyptus plantation regions. Lastly, to validate the efficacy of selecting multi-temporal data and the performance of the SLPSO–RFE model in classification, a comparative analysis is undertaken against the classification results derived from single-temporal data and the ReliefF–RFE feature selection scheme. The findings reveal that employing SLPSO–RFE for feature selection significantly elevates the classification precision of eucalyptus plantations across all three classifiers. The overall accuracy rates were noted at 95.48% for SVM, 96% for CART, and 97.97% for RF. When contrasted with classification outcomes from multi-temporal data and ReliefF–RFE, the overall accuracy for the trio of models saw an increase of 10%, 8%, and 8.54%, respectively. The accuracy enhancement was even more pronounced when juxtaposed with results from single-temporal data and ReliefF-RFE, at increments of 15.25%, 13.58%, and 14.54% respectively. The insights from this research carry profound theoretical implications and practical applications, particularly in identifying and extracting eucalyptus plantations leveraging multi-temporal data and feature selection.

Data-driven analysis and prediction of indoor characteristic temperature in district heating systems

Many indoor temperature sensors have been installed in building rooms with district heating systems (DHSs) in China. To apply the collected indoor temperature data to district heating substation regulation, the indoor characteristic temperature of the heat substation (ICTS) should be calculated. However, previous calculation methods for ICTS cannot accurately evaluate their comprehensive characteristics. This study proposed a novel method to calculate the ICTS based on the entropy value and applied it to three actual DHSs in northern China. First, the variation trend and influencing factors of indoor temperature data were analyzed. Second, the ICTS was calculated and compared using various methods. Third, a correlation analysis of the ICTS and input features was performed. Finally, five typical machine learning models were used to predict the ICTS. The results demonstrated that the proposed method can better reflect the characteristics of the data and eliminate the impact of subjective preferences. The average mean absolute percentage errors of five models were less than 1%; the linear regression model performed best when denoising the original data, with a value of 0.12%. This study is helpful for better understanding the influencing factors on the ICTS and the strategies for adjusting feature values.

A deep learning-based interpretable decision tool for predicting high risk of chemotherapy-induced nausea and vomiting in cancer patients prescribed highly emetogenic chemotherapy.

This study aims to develop a risk prediction model for chemotherapy-induced nausea and vomiting (CINV) in cancer patients receiving highly emetogenic chemotherapy (HEC) and identify the variables that have the most significant impact on prediction. Data from Tianjin Medical University General Hospital were collected and subjected to stepwise data preprocessing. Deep learning algorithms, including deep forest, and typical machine learning algorithms such as support vector machine (SVM), categorical boosting (CatBoost), random forest, decision tree, and neural network were used to develop the prediction model. After training the model and conducting hyperparameter optimization (HPO) through cross-validation in the training set, the performance was evaluated using the test set. Shapley additive explanations (SHAP), partial dependence plot (PDP), and Local Interpretable Model-Agnostic Explanations (LIME) techniques were employed to explain the optimal model. Model performance was assessed using AUC, F1 score, accuracy, specificity, sensitivity, and Brier score. The deep forest model exhibited good discrimination, outperforming typical machine learning models, with an AUC of 0.850 (95%CI, 0.780-0.919), an F1 score of 0.757, an accuracy of 0.852, a specificity of 0.863, a sensitivity of 0.784, and a Brier score of 0.082. The top five important features in the model were creatinine clearance (Ccr), age, gender, anticipatory nausea and vomiting, and antiemetic regimen. Among these, Ccr had the most significant predictive value. The risk of CINV decreased with increased Ccr and age, while it was higher in the presence of anticipatory nausea and vomiting, female gender, and non-standard antiemetic regimen. The deep forest model demonstrated good discrimination in predicting the risk of CINV in cancer patients prescribed HEC. Kidney function, as represented by Ccr, played a crucial role in the model's prediction. The clinical application of this predictive tool can help assess individual risks and improve patient care by proactively optimizing the use of antiemetics in cancer patients receiving HEC.

Covid-19 Detection For CT-scan Images Using Transfer Learning Models

COVID-19 is a respiratory illness caused by a virus called SARS-CoV-2 which affected around 455 million people around the world. CT-scan is a medical imaging technique that uses X-rays to create detailed images of the body and which can be used to detect many respiratory diseases. Transfer learning models are a type of machine learning model that are trained on a large dataset of images and which can be used for their already trained ability to extract features from image in other tasks. They can then be used to classify new images with similar features.This paper presents a study of different transfer learning models for the task of classifying chest X-ray images into three classes: COVID-19, pneumonia, and normal. The study was implemented using Python and the dataset used was the COVID-19 Chest X-ray Dataset. The train-test split used was 0.2–0.8. The parameters used to test the models were the precision, recall, accuracy, F1 score, and Matthew’s correlation score. Other than these, different optimizers were also compared such as ADAM, SGD with different learning rates of 0.01, 0.001, and 0.0001.The models used in this study are EfficientNetB0, EfficientNetB7, VGG16, and InceptionV3. Out of these models, the most effective model was the EfficientNetB0 model, which achieved an accuracy of 98.6%. This study provides valuable insights into the use of transfer learning for medical image analysis. The results suggest that transfer learning can be used to develop accurate and efficient models that can be used as a secondary option for the diagnosis of COVID-19 using chest X-ray images.

A review of SARS-CoV-2 drug repurposing: databases and machine learning models.

The emergence of Severe Acute Respiratory Syndrome Corona Virus 2 (SARS-CoV-2) posed a serious worldwide threat and emphasized the urgency to find efficient solutions to combat the spread of the virus. Drug repurposing has attracted more attention than traditional approaches due to its potential for a time- and cost-effective discovery of new applications for the existing FDA-approved drugs. Given the reported success of machine learning (ML) in virtual drug screening, it is warranted as a promising approach to identify potential SARS-CoV-2 inhibitors. The implementation of ML in drug repurposing requires the presence of reliable digital databases for the extraction of the data of interest. Numerous databases archive research data from studies so that it can be used for different purposes. This article reviews two aspects: the frequently used databases in ML-based drug repurposing studies for SARS-CoV-2, and the recent ML models that have been developed for the prospective prediction of potential inhibitors against the new virus. Both types of ML models, Deep Learning models and conventional ML models, are reviewed in terms of introduction, methodology, and its recent applications in the prospective predictions of SARS-CoV-2 inhibitors. Furthermore, the features and limitations of the databases are provided to guide researchers in choosing suitable databases according to their research interests.

Uncertainty quantification of a deep learning model for failure rate prediction of water distribution networks

Predicting the time-dependent pipe failure rate of the water distribution networks (WDNs) is important for planning its renewal budget but also challenging due to the complex factors involved. The recent development of machine learning techniques provides a novel approach for accurate failure rate prediction based on historical data. However, the inherited randomness and uncertainty of water pipe failures and machine learning algorithms are often ignored in the training and prediction process. This article develops an uncertainty quantification framework for deep learning-based WDN system-wide failure rate prediction. The framework integrates a probabilistic long short-term memory (LSTM) model with a Monte Carlo method. The historical climate data and WDN pipe maintenance data over the past 35 years for Cuyahoga County, USA, are used to illustrate this deep-learning model-based uncertainty quantification framework in this study. A statistical time series regression model, ARIMAX, is used as a comparison benchmark. The results show that the LSTM model outperforms the ARIMAX model in prediction accuracy in most years by considering the uncertainties. Besides, the uncertainty range of the LSTM model prediction is 50% of that of the ARIMAX model. The results also identified the major contributing factors to the uncertainties of LSTM machine learning model prediction. The proposed uncertainty framework features excellent extensibility and can be adapted to quantify uncertainties with other types of machine learning models.

Optimization of thermal desorption conditions of stir bar sorptive extraction facilitated by machine learning

Stir bar sorptive extraction is an effective technique for trapping odorants, but there are limited studies on the effect of varying thermal desorption conditions on desorption efficiency of odorants. Therefore, we conducted this study to explore the relationship between desorption conditions and desorption efficiency for 18 odorants with diverse physicochemical properties using instrumental analysis and mathematical modeling. We trained four types of machine learning models using a dataset comprising 864 different combinations of four desorption conditions (each three levels) and physicochemical properties. The prediction value of the selected model was validated using a validation dataset of six new odorants. The Random Forest model had the highest performance (R = 0.910). The order of feature importance using this model was as follows: cryo-focusing temperature, molecular weight, log P, boiling point, desorption temperature, desorption time, and helium flow. For testing on new odorants, the correlations between predicted and experimental data for terpene (R = 0.99), alcohol (R = 0.98), ester (R = 0.92), sulfide (R = 0.89), phenol (R = 0.88), and aldehyde (R = 0.61) were determined.

Geological Hazard Susceptibility Analysis Based on RF, SVM, and NB Models, Using the Puge Section of the Zemu River Valley as an Example

The purpose of this study was to construct a geological hazard susceptibility evaluation and analysis model using three types of machine learning models, namely, random forest (RF), support vector machine (SVM), and naive Bayes (NB), and to evaluate the susceptibility to landslides, using the Puge section of the Zemu River valley in the Liangshan Yi Autonomous Prefecture as the study area. First, 89 shallow landslide and debris flow locations were recognized through field surveys and remote sensing interpretation. A total of eight hazard-causing factors, namely, slope, aspect, rock group, land cover, distance to road, distance to river, distance to fault, and normalized difference vegetation index (NDVI), were selected to evaluate the spatial relationship with landslide occurrence. As a result of the analysis, the results of the weighting of the hazard-causing factors indicate that the two elements of rock group and distance to river contribute most to the creation of geological hazards. After comparing all the indices of the three models, the random forest model had a higher correct area under the ROC curve (AUC) value of 0.87, root mean squared error (RMSE) of 0.118, and mean absolute error (MAE) of 0.045. The SVM model had the highest sensitivity to geological hazards. The results of geological hazard prediction susceptibility analysis matched the actual situation in the study area, and the prediction effects were good. The results of the hazard susceptibility assessment of the three models are able to provide support and help for the prevention and control of geological hazards in the same type of areas.

Rotor equivalent wind speed prediction based on mechanism analysis and residual correction using Lidar measurements

Efficient power generation and reduced operational cost are crucial to make wind energy a successful market player. To achieve satisfactory performance in wind turbine technology, it is essential to accurately predict the rotor equivalent wind speed (REWS). Since traditional wind transducers are difficult to measure multiple-point wind speed, related researches on predicting REWS are lacking. In this paper, a hybrid REWS prediction framework is proposed, which combines mechanism analysis and data-driven technique, aided by Lidar measurements. The framework employs the mechanism modeling and time delay analysis to achieve preliminary REWS prediction, but the accuracy is restricted by the quantity of Lidar measurement points. To enhance the prediction accuracy, a residual prediction model based on the data-driven technique is created to correct the mechanism prediction. Specifically, a novel combined modeling approach based on empirical mode decomposition and sample entropy algorithms is presented. Through reconstructing the input data of the combined model into high, middle, and low frequency components and exploring the prediction laws of different frequency components obtained by typical machine learning models, the combined residual prediction model is established. Additionally, an enhanced whale optimization algorithm named WOAmM is presented to optimize the parameters of the combined model. The proposed method is validated through a four-beam Lidar measurement scenario simulated by BLADED software. The experimental results indicate that the combined model outperforms each single model and the mechanism prediction, and improves the REWS prediction accuracy by up to75.04%, 60.92%, 51.85%, and 57.52% in the four datasets, respectively. The proposed method could facilitate the utilization of Lidar in improving the performance of wind turbines.