Extreme Tree Research Articles

Rockburst prediction based on multi-featured drilling parameters and extreme tree algorithm for full-section excavated tunnel faces

The suddenness, uncertainty, and randomness of rockbursts directly affect the safety of tunnel construction. The prediction of rockbursts is a fundamental aspect of mitigating or even eliminating rockburst hazards. To address the shortcomings of the current rockburst prediction models, which have a limited number of samples and rely on manual test results as the majority of their input features, this paper proposes rockburst prediction models based on multi-featured drilling parameters of rock drilling jumbo. Firstly, four original drilling parameters, namely hammer pressure (Ph), feed pressure (Pf), rotation pressure (Pr), and feed speed (VP), together with the rockburst grades, were collected from 1093 rockburst cases. Then, a feature expansion investigation was performed based on the four original drilling parameters to establish a drilling parameter feature system and a rockburst prediction database containing 42 features. Furthermore, rockburst prediction models based on multi-featured drilling parameters were developed using the extreme tree (ET) algorithm and Bayesian optimization. The models take drilling parameters as input parameters and rockburst grades as output parameters. The effects of Bayesian optimization and the number of drilling parameter features on the model performance were analyzed using the accuracy, precision, recall and value of the prediction set as the model performance evaluation indices. The results show that the Bayesian optimized model with 42 drilling parameter features as inputs performs best, with an accuracy of 91.89%. Finally, the reliability of the models was validated through field tests.

Prediction of glass transition temperature of oxide glasses based on interpretable machine learning and sparse data sets

The glass transition temperature (Tg) is a crucial characteristic of oxide glasses, exerting significant influence on their properties and applications. In this study, we utilized a sparse dataset and machine learning techniques to establish a predictive model for the relationship between the composition of oxide glasses and Tg. Among four machine learning algorithms compared, the Extreme Tree Regressor (ETR) algorithm demonstrated superior performance, exhibiting robust generalization capabilities on the validation set. We performed feature selection on the original features using Pearson Correlation Coefficient (PCC) and Maximal Information Coefficient (MIC), resulting in a subset of 38 dimensions. Subsequently, we employed Sparse Principal Component Analysis (PCA) and Recursive Feature Elimination (RFE) for dimensionality reduction, yielding a final subset of 12 dimensions. Introducing SHapley Additive exPlanations (SHAP) values for interpretability analysis of the predictive model, we obtained important feature rankings and analyzed how variations in features affect the target variable. Finally, using the CaO-SiO2-Al2O3 system as an example, we demonstrated how the model predicts Tg based on composition, facilitating rational design of oxide glass compositions.

Comparative analysis of machine learning techniques for predicting drilling rate of penetration (ROP) in geothermal wells: A case study of FORGE site

Enhanced Geothermal Systems (EGS) present a compelling means to unlock the considerable yet largely untapped thermal energy within the earth's crust globally. The EGS drilling cost constitutes a substantial portion, up to 60 %, of the overall expenses. Consequently, streamlining and optimizing the drilling processes for these systems hold immense economic significance, with the potential to substantially lower costs, advance the utilization of geothermal energy, and contribute significantly to the reduction of carbon emissions. This research aims to enhance EGS drilling operations by seeking ways of reducing drilling cost.In this research, we harnessed the predictive power of 10 state-of-the-art machine learning (ML) algorithms to anticipate a crucial drilling parameter: ROP. Using the FORGE dataset, we developed a code tailored for the intensive preprocessing of drilling data. This code offers many options, including various noise-removing techniques and scaling approaches. However, the primary focus of our work extends to quantifying uncertainties intrinsic to the predictions of the 10 algorithms employed. To achieve this, a comprehensive approach involved subjecting each algorithm to 40 runs, utilizing the best model from the tuning process.The results show that unaddressed uncertainties may lead to unstable model behavior, where small changes in the input data or random initialization result in significant prediction variations. Thus, models that do not account for uncertainty may be overfit to the noise in the training data, leading to poor generalization to new, unseen data. Remarkably, the results highlight the superiority of the Extreme Tree, Light Gradient Boost, Random Forest, and Gradient Boosting algorithms, showcasing mean absolute error (MAE) and R2 values within the ranges of 3.5–4.5 and 0.9–0.95, respectively. Conversely, artificial neural networks and support vector machine algorithms demonstrate comparatively lower performance, with MAE ranging from 4.7 to 6.3 and R2 from 0.85 to 0.91.Although the results presented in this work are only based on the drilling parameters from one well of the FORGE site and do not include rock properties or geological parameters, however, the proposed nuanced understanding of algorithmic performance is valuable for refining predictive models in geothermal drilling applications, ensuring robustness and reliability in the face of diverse operational scenarios. Navigating a successful workflow, adeptly processing data, and meticulously quantifying uncertainties linked to each predictive algorithm emerge as formidable yet indispensable tasks. Our research not only sheds light on these complexities but also paves the way for a strategic optimization approach based on data-driven ROP models. Therefore, the potential for substantial savings in real-time drilling operations becomes apparent, unlocking a new realm of possibilities for the geothermal energy sector.

Prediction of digital transformation of manufacturing industry based on interpretable machine learning.

The enhancement of digital transformation is of paramount importance for business development. This study employs machine learning to establish a predictive model for digital transformation, investigates crucial factors that influence digital transformation, and proposes corresponding improvement strategies. Initially, four commonly used machine learning algorithms are compared, revealing that the Extreme tree classification (ETC) algorithm exhibits the most accurate prediction. Subsequently, through correlation analysis and recursive elimination, key features that impact digital transformation are selected resulting in the corresponding feature subset. Shapley Additive Explanation (SHAP) values are then employed to perform an interpretable analysis on the predictive model, elucidating the effects of each key feature on digital transformation and obtaining critical feature values. Lastly, informed by practical considerations, we propose a quantitative adjustment strategy to enhance the degree of digital transformation in enterprises, which provides guidance for digital development.

Performance Analysis of Tree-Based and Deep Learning Algorithms for Developing Distributed Secure Systems in IoT: A Comparative Study

Notably, IoT device utilization has experienced a substantial wave recently, and ensuring these devices' privacy and security has become a critical concern. ML-based security approaches are promising for IoT network protection against security concerns. This study provides a proximate analysis of tree-based and deep-learning algorithms for securing IoT domains. Specifically, we evaluate Decision Tree, RandomForest, XGBoost, Catboost, Extreme Tree, Light GMB, Adaptive Boosting, CNN, LSTM, MLP, GRU, and Autoencoder on four publicly available datasets - IoT23, CICID2017, EdgeIIoT, BotnetIoT and Contiki OS and Cooja simulation were used to generate a dataset featuring various RPL attacks. To assess the performance of a model, we measure its accuracy, precision, recall, and F1-score metrics. Our discoveries indicate that tree-based algorithms outperform deep learning algorithms regarding training time, memory usage, and interpretability while gaining comparable or even better detection accurateness. Conversely, deep-learning algorithms exhibit higher detection rates for rare or previously unseen attacks; their proficiency in detecting complex patterns and relationships within a given dataset has demonstrated remarkable efficacy in data analysis and classification tasks. We conclude that both tree-based and deep learning algorithms have their strengths and weaknesses, and in the IoT environment, one should base the choice of the algorithm on requirements and constraints. Our research shows hybrid approaches combining algorithm strengths can establish secure, distributed IoT systems

T2-weighted imaging-based radiomic-clinical machine learning model for predicting the differentiation of colorectal adenocarcinoma.

The study on predicting the differentiation grade of colorectal cancer (CRC) based on magnetic resonance imaging (MRI) has not been reported yet. Developing a non-invasive model to predict the differentiation grade of CRC is of great value. To develop and validate machine learning-based models for predicting the differentiation grade of CRC based on T2-weighted images (T2WI). We retrospectively collected the preoperative imaging and clinical data of 315 patients with CRC who underwent surgery from March 2018 to July 2023. Patients were randomly assigned to a training cohort (n = 220) or a validation cohort (n = 95) at a 7:3 ratio. Lesions were delineated layer by layer on high-resolution T2WI. Least absolute shrinkage and selection operator regression was applied to screen for radiomic features. Radiomics and clinical models were constructed using the multilayer perceptron (MLP) algorithm. These radiomic features and clinically relevant variables (selected based on a significance level of P < 0.05 in the training set) were used to construct radiomics-clinical models. The performance of the three models (clinical, radiomic, and radiomic-clinical model) were evaluated using the area under the curve (AUC), calibration curve and decision curve analysis (DCA). After feature selection, eight radiomic features were retained from the initial 1781 features to construct the radiomic model. Eight different classifiers, including logistic regression, support vector machine, k-nearest neighbours, random forest, extreme trees, extreme gradient boosting, light gradient boosting machine, and MLP, were used to construct the model, with MLP demonstrating the best diagnostic performance. The AUC of the radiomic-clinical model was 0.862 (95%CI: 0.796-0.927) in the training cohort and 0.761 (95%CI: 0.635-0.887) in the validation cohort. The AUC for the radiomic model was 0.796 (95%CI: 0.723-0.869) in the training cohort and 0.735 (95%CI: 0.604-0.866) in the validation cohort. The clinical model achieved an AUC of 0.751 (95%CI: 0.661-0.842) in the training cohort and 0.676 (95%CI: 0.525-0.827) in the validation cohort. All three models demonstrated good accuracy. In the training cohort, the AUC of the radiomic-clinical model was significantly greater than that of the clinical model (P = 0.005) and the radiomic model (P = 0.016). DCA confirmed the clinical practicality of incorporating radiomic features into the diagnostic process. In this study, we successfully developed and validated a T2WI-based machine learning model as an auxiliary tool for the preoperative differentiation between well/moderately and poorly differentiated CRC. This novel approach may assist clinicians in personalizing treatment strategies for patients and improving treatment efficacy.

Reconstruction of Snow Cover in Kaidu River Basin via Snow Grain Size Gap-Filling Based on Machine Learning

Fine spatiotemporal resolution snow monitoring at the watershed scale is crucial for the management of snow water resources. This research proposes a cloud removal algorithm via snow grain size (SGS) gap-filling based on a space–time extra tree, which aims to address the issue of cloud occlusion that limits the coverage and time resolution of long-time series snow products. To fully characterize the geomorphic characteristics and snow duration time of the Kaidu River Basin (KRB), we designed dimensional data that incorporate spatiotemporal information. Combining other geographic and snow phenological information as input for estimating SGS. A spatiotemporal extreme tree model was constructed and trained to simulate the nonlinear mapping relationship between multidimensional inputs and SGS. The estimation results of SGS can characterize the snow cover under clouds. This study found that when the cloud cover is less than 70%, the model’s estimation of SGS meets expectations, and snow cover reconstruction achieves good results. In specific cloud removal cases, compared to traditional spatiotemporal filtering and multi-sensor fusion, the proposed method has better detail characterization ability and exhibits better performance in snow cover reconstruction and cloud removal in complex mountainous environments. Overall, from 2000 to 2020, 66.75% of snow products successfully removed cloud coverage. This resulted in a decrease in the annual average cloud coverage rate from 52.46% to 34.41% when compared with the MOD10A1 snow product. Additionally, there was an increase in snow coverage rate from 21.52% to 33.84%. This improvement in cloud removal greatly enhanced the time resolution of snow cover data without compromising the accuracy of snow identification.

The North American Monsoon buffers forests against the ongoing megadrought in the Southwestern United States.

The US Southwest has been entrenched in a two-decade-long megadrought (MD), the most severe since 800 CE, which threatens the long-term vitality and persistence of regional montane forests. Here, we report that in the face of record low winter precipitation and increasing atmospheric aridity, seasonal activity of the North American Monsoon (NAM) climate system brings sufficient precipitation during the height of the summer to alleviate extreme tree water stress. We studied seasonally resolved, tree-ring stable carbon isotope ratios across a 57-year time series (1960-2017) in 17 Ponderosa pine forests distributed across the NAM geographic domain. Our study focused on the isotope dynamics of latewood (LW), which is produced in association with NAM rains. During the MD, populations growing within the core region of the NAM operated at lower intrinsic and higher evaporative water-use efficiencies (WUEi and WUEE , respectively), compared to populations growing in the periphery of the NAM domain, indicating less physiological water stress in those populations with access to NAM moisture. The disparities in water-use efficiencies in periphery populations are due to a higher atmospheric vapor pressure deficit (VPD) and reduced access to summer soil moisture. The buffering advantage of the NAM, however, is weakening. We observed that since the MD, the relationship between WUEi and WUEE in forests within the core NAM domain is shifting toward a drought response similar to forests on the periphery of the NAM. After correcting for past increases in the atmospheric CO2 concentration, we were able to isolate the LW time-series responses to climate alone. This showed that the shift in the relation between WUEi and WUEE was driven by the extreme increases in MD-associated VPD, with little advantageous influence on stomatal conductance from increases in atmospheric CO2 concentration.

Classification of Water Source in Coal Mine Based on PCA-GA-ET

In recent years, inrush water has hampered the regular mining of coal mines, and the proper identification of the source of inrush water is critical to the prevention and management of water hazards in mines. This paper extracts the standard water chemistry discriminating ions Na++K+, Ca2+, Mg2+, Cl−, SO42−, and HCO3− from observed water samples. An improved water source discrimination model is proposed which combines algorithms from data mining, classification models, and learning reinforcement. According to the Pearson correlation coefficient, Na++K+ has a strong correlation with HCO3−. To identify the major metrics, we performed principal component analysis (PCA), and the adaptive differential evolutionary genetic algorithm (GA) was utilized to optimize the depth of the extreme tree (ET) and the number of classifiers. Finally, the model distinguished 25 sets of studied samples from various water sources in the Pingdingshan coalfield. Comparative analysis demonstrated the efficacy of each stage of our work. PCA-GA-ET outperformed the conventional approaches, such as the support vector machine, BP artificial neural network, and random forest. The studies revealed that PCA-GA-ET can eliminate the information overlap between data and simplify the data structure and thereby improve the efficiency and accuracy of water source detection. We discovered that by utilizing the evolutionary algorithm to optimize parameters such as the depth of the extreme trees and the number of decision trees, we could get the model to converge faster and to be more stable and more accurate. The results suggest that PCA-GA-ET has good robustness and accuracy and can meet the needs of water source identification.

Estimating the effective fracture toughness of a variety of materials using several machine learning models

Since conducting laboratory tests to obtain the fracture toughness of materials is time-consuming and costly, it is necessary to provide tools to estimate this property with high accuracy quickly and without the need for such a high cost. This study considered using machine learning (ML)-based models as a suitable option to address such problems. For this purpose, twelve ML-based models were presented using 1715 datasets generated from five experimental tests to estimate the adequate fracture toughness (Keff) of 44 different materials. The behavior of the ML models compared to the practice tests was investigated, and the correct and acceptable performance of each of them in estimating the Keff of different materials was confirmed. Among the twelve ML-based models, extreme tree regressor (ETR) and Gaussian process regression (GPR) models provided the highest and lowest accuracies in estimating the Keff of different materials, respectively. To further aid in the estimation of the Keff of different materials for engineering challenges, a graphical user interface (GUI) for the ML-based models was developed.

Online evaluation of the boiler energy efficiency associated with the soot blowing in coal‐fired power plants

AbstractSoot blowing is a common practice in coal‐fired power plants to ensure high energy conversion rate. The period of the conventional soot blowing is usually fixed, without the consideration of the load change and feeding coal switching. For the purpose of the soot blowing optimization, an approach to quantitatively evaluate the performance of soot blowing operations is proposed. Different heat exchangers' thermal resistances are calculated based on dynamic mass and energy balance to monitor their ash fouling deposition, and then, the boiler energy efficiency is predicted, adopting the rolling extreme tree boosting algorithm, and finally, the efficiencies with and without soot blowing are computed separately. With the data collected from a 350 MW ultra‐supercritical unit, pseudo‐online simulations are performed. The results show that even if the energy efficiency of the unit has exceeded 94%, a full‐process soot blowing may further increase the efficiency in some extent. For those subcritical units, the benefits should be higher.

Fuels change quickly after California drought and bark beetle outbreaks with implications for potential fire behavior and emissions

BackgroundAn extreme drought from 2012–2016 and concurrent bark beetle outbreaks in California, USA resulted in widespread tree mortality. We followed changes in tree mortality, stand structure, and surface and canopy fuels over four years after the peak of mortality in Sierra mixed conifer and pinyon pine (Pinus monophylla) forests to examine patterns of mortality, needle retention after death, and snag fall across tree species. We then investigated how the tree mortality event affected surface and canopy fuel loading and potential impacts on fire hazard and emissions.ResultsDrought and beetle-related tree mortality shifted mortality patterns to be more evenly distributed across size classes and concentrated in pines. Substantial changes to surface fuel loading, stand density, canopy fuel loads, and potential wildfire emissions occurred within four years following peak levels of tree mortality, with the largest changes related to increases in coarse woody debris. Nearly complete needle fall occurred within four years of mortality for all species except red fir (Abies magnifica). Pine species and incense cedar (Calocedrus decurrens) snags fell more quickly than fir species. Potential fire behavior modelling suggested that crowning and torching hazard decreased as trees dropped dead needles and fell, but as canopy fuels were transferred to surface fuels, potential for smoldering combustion increased, causing greater emissions.ConclusionsOur study increases understanding of how extreme tree mortality events caused by concurrent disturbances alter canopy and surface fuel loading and have the potential to affect fire behavior and emissions in two compositionally different seasonally dry forest types. After a major tree mortality event, high canopy fuel flammability may only last a few years, but surface fuels can increase considerably over the same time period in these forest types. The accumulation of coarse woody surface fuels resulting from multi-year drought and concurrent bark beetle outbreaks combined with the increasing frequency of drought in the western U.S. have the potential to lead to heavy and dry fuel loads that under certain weather conditions may result in more extreme fire behavior and severe effects, particularly in forest types where decades of successful fire suppression has caused forest densification.

Research on Boruta-ET-Based Anomalous Traffic Detection Model

With the rapid development of networks, intrusion detection has received increasing attention. In order to solve the problems of large dimensionality of intrusion detection data, unbalanced data samples, and large dispersion of datasets, which seriously affect the classification performance, this study proposes an anomaly detection based on Boruta and extreme tree (Boruta-ET) model. First, the network traffic data are preprocessed, which includes data cleaning, numerical and normalization processes, as well as equalization of the attack categories for a small number of samples by random oversampling at the data level; second, the traffic features are dimensionality reduced using the Boruta-based algorithm. The goal of Boruta dimensionality reduction is to extract all the features related to the dependent variable with a global dimension and find the optimal subset of features containing the most information; finally, the optimal feature subset is used as the input parameters of the extreme tree (ET) algorithm model for training and testing. Experiments were conducted on the real network traffic dataset CICIDS2017, and by evaluating the classification performance of several different machine learning algorithms, the experimental results show that the Boruta-ET model has the best performance with an accuracy rate of 99.80%, which can effectively improve the detection rate and achieve an effective recall rate for attack types with a small number of samples.

Rangewide climatic sensitivities and non-timber values of tall Sequoia sempervirens forests

Heavily exploited for its reddish, decay-resistant heartwood, the tallest conifer, Sequoia sempervirens, is a major component of coastal forests from extreme southwestern Oregon to California’s Santa Lucia Mountains. Primary Sequoia forests are now restricted to < 5 % of their former distribution, and mature secondary forests with trees over 60 m tall are even scarcer due to repeated logging. Leveraging allometric equations recently derived from intensive work in both forest types, we climbed, measured, and core-sampled 235 trees in 45 locations distributed across the species range to examine growth trends and understand how tall Sequoia are responding to recent environmental changes. Paired samples of sapwood and heartwood collected along the height gradient were used to quantify Sequoia investment in decay resistance. During the 20th century, trees in most locations began producing more wood than expected for their size with this growth surge becoming pronounced after 1970 and ending around 2000. Radial increments—ring widths—correlate with climatic variables related to water availability, and these relationships are strengthening as temperatures rise. Sensitivity to drought increased from north to south along a 6° latitudinal gradient of decreasing precipitation and summer fog frequency. Sequoia trees north of 40° were least sensitive to drought, producing similar biomass annually during dry and wet years, whereas trees farther south produced less biomass during individual drought years. Hotter 21st century drought barely affected Sequoia growth efficiency (biomass increment per unit leaf mass) north of 40° until the fourth consecutive year (2015), when growth efficiency dropped precipitously, recovering within two years. South of 40°, Sequoia trees exhibited steadily declining growth efficiency during the multi-year drought followed by recovery, but recovery did not occur south of 37° despite ample precipitation in 2017. Sequoia growth efficiency is currently highest in secondary forests north of 40°, where trees produce relatively small amounts of heartwood with the lowest decay resistance (least fungicide) while receiving the most nocturnal summer fog. Increasing sink limitations, whereby rising temperatures, drier air at night, and extreme tree height collectively lower turgor pressure to inhibit cambial activity, may reduce Sequoia growth efficiency while contributing to more durable biomass production. Heartwood and fungicide increments are higher in primary than secondary forests across the species range. Crown structural complexity promotes development of vascular epiphytes and arboreal soil habitats in Sequoia forests with sufficient moisture availability. These habitats are lacking in secondary forests and rare in primary forests south of 40°. After logging, restoration of tall Sequoia forests can be achieved via silviculture that maximizes height increments during early stand development and then retains some dominant trees in perpetuity, allowing them to gain full stature, produce increasingly decay-resistant heartwood, and support significant arboreal biodiversity.

Spam Comments Detection on Instagram Using Machine Learning and Deep Learning Methods

The more popular a public figure on Instagram (IG), the number of followers also increase. When a public figure posts something, there are many comments from other users. In fact, from all the comments, not all of them are relevant to the post, such as advertising, links, or clickbait comments. The type of comments that are irrelevant to the post is usually called spam comments. Spam comments will interfere with information flow and may lead to misleading information. This research compares machine learning (ML) and deep learning (DL) classification methods based on our collected Indonesian IG spam comment dataset. This research was conducted in the following steps: dataset preparation, pre-processing, simple normalization, features generation using TF-IDF and word embedding, application of ML and DL classification methods, performance evaluation, and comparison. The authors compare accuracy, F-1, precision, and recall from ML and DL results. This research shows that ML and DL methods do not significantly differ. The Linear SVM, Extreme Tree (ET), Regression, and Stochastics Gradient Descent algorithms can reach the accuracy of 0.93. At the same time, the DL method has the highest accuracy of 0.94 using the SimpleTransformer BERT architecture. The difference between ML and DL methods is not significantly different.

A Fused Multidimensional EEG Classification Method Based on an Extreme Tree Feature Selection.

When a brain-computer interface (BCI) is designed, high classification accuracy is difficult to obtain for motor imagery (MI) electroencephalogram (EEG) signals in view of their relatively low signal-to-noise ratio. In this paper, a fused multidimensional classification method based on extreme tree feature selection (FMCM-ETFS) is proposed for discerning motor imagery EEG tasks. First, the EEG signal was filtered by a Butterworth filter for preprocessing. Second, C3, C4, and CZ channels were selected to extract time-frequency domain and spatial domain features using autoregressive (AR), common spatial pattern (CSP), and discrete wavelet transform (DWT). The extracted features were fused for a further feature elimination. Then, the features were selected using three feature selection methods: recursive feature elimination (RFE), principal component analysis method (PCA), and extreme trees (ET). The selected feature vectors were classified using support vector machines (SVM). Finally, a total of twelve subjects' EEG data from Inner Mongolia University of Technology (IMUT data), the 2nd BCI competition in 2003, and the 4th BCI competition in 2008 were employed to show the effectiveness of this proposed FMCM-ETFS method. The results show that the classification accuracy using the multidimensional fused feature extraction (AR + CSP + DWT) is 3%–20% higher than those using the aforementioned three single feature extractions (AR, CSP, and DWT). Extreme trees (ET), which is a sort of tree-based model method, outperforms RFE and PCA by 1%–9% in term of classification accuracies, when these three methods were applied to the procedure of feature extraction, respectively.

A Novel Tree Ensemble Model to Approximate the Generalized Extreme Value Distribution Parameters of the PM2.5 Maxima in the Mexico City Metropolitan Area

We introduce a novel spatial model based on the distribution of generalized extreme values (GEVs) and tree ensemble models to analyze the maximum concentrations levels of particulate matter with a diameter of less than 2.5 microns (PM2.5) in the Mexico City metropolitan area during the period 2003–2021. Spatial trends were modeled through a decision tree in the context of a non-stationary GEV model. We used a tree ensemble model as a predictor of GEV parameters to approximate nonlinear trends. The decision tree was built by using a greedy stagewise approach, the objective function of which was the log-likelihood. We verified the validity of our model by means of the likelihood and Akaike’s information criterion (AIC). The maps of the generalized extreme value parameters on the spatial plane show the existence of differentiated local trends in the extreme values of PM2.5 in the study area. The results indicated strong evidence of an increase in the west–east direction of the study area. A spatial map of risk with maximum concentration levels of PM2.5 in a period of 25 years was built.

XLPFE: A Simple and Effective Machine Learning Scoring Function for Protein-Ligand Scoring and Ranking.

Prediction of protein–ligand binding affinities is a central issue in structure-based computer-aided drug design. In recent years, much effort has been devoted to the prediction of the binding affinity in protein–ligand complexes using machine learning (ML). Due to the remarkable ability of ML methods in nonlinear fitting, ML-based scoring functions (SFs) can deliver much improved performance on a selected test set, such as the comparative assessment of scoring functions (CASF), when compared to the classical SFs. However, the performance of ML-based SFs heavily relies on the overall similarity of the training set and the test set. To improve the performance and transferability of an SF, we have tried to combine various features including energy terms from X-score and AutoDock Vina, the properties of ligands, and the statistical sequence-related information from either the binding site or the full protein. In conjunction with extreme trees (ET), an ML model, we have developed XLPFE, a new SF. Compared with other tested methods such as X-score, AutoDock Vina, ΔvinaXGB, PSH-ML, or CNN-score, XLPFE achieves consistently better scoring and ranking power for various types of protein–ligand complex structures beyond the CASF, suggesting that XLPFE has superior transferability. In particular, XLPFE performs better with metalloenzymes. With its faster speed, improved accuracy, and better transferability, XLPFE could be usefully applied to a diverse range of protein–ligand complexes.

A Hybrid Feature Extraction Method for Nepali COVID-19-Related Tweets Classification.

COVID-19 is one of the deadliest viruses, which has killed millions of people around the world to this date. The reason for peoples' death is not only linked to its infection but also to peoples' mental states and sentiments triggered by the fear of the virus. People's sentiments, which are predominantly available in the form of posts/tweets on social media, can be interpreted using two kinds of information: syntactical and semantic. Herein, we propose to analyze peoples' sentiment using both kinds of information (syntactical and semantic) on the COVID-19-related twitter dataset available in the Nepali language. For this, we, first, use two widely used text representation methods: TF-IDF and FastText and then combine them to achieve the hybrid features to capture the highly discriminating features. Second, we implement nine widely used machine learning classifiers (Logistic Regression, Support Vector Machine, Naive Bayes, K-Nearest Neighbor, Decision Trees, Random Forest, Extreme Tree classifier, AdaBoost, and Multilayer Perceptron), based on the three feature representation methods: TF-IDF, FastText, and Hybrid. To evaluate our methods, we use a publicly available Nepali-COVID-19 tweets dataset, NepCov19Tweets, which consists of Nepali tweets categorized into three classes (Positive, Negative, and Neutral). The evaluation results on the NepCOV19Tweets show that the hybrid feature extraction method not only outperforms the other two individual feature extraction methods while using nine different machine learning algorithms but also provides excellent performance when compared with the state-of-the-art methods.

Hybrid Model of Machine Learning Refractory Data Prediction Based on IoT Smart Cities

With the advent of the digital age in recent years, the application of artificial intelligence in urban Internet of Things (IoT) systems has become increasingly important. The concept of smart cities has gradually formed, and smart firefighting under the smart city system has also become important. The method of machine learning is now applied in various fields, but seldom to the data prediction of smart firefighting. Various types of applications including data applications of machine learning algorithms in smart firefighting have yet to be explored. In this article, we propose using machine learning algorithms to predict building fire-resistance data, aiming to provide more theoretical and technical support for IoT smart cities. This article adopts the fire-resistance data of building beam components in a real fire environment, using three integrated machine learning algorithms, Extreme random Tree (ET), AdaBoost, and Gradient Boosting Machine (GBM), and the grey wolf optimization algorithm to optimize. We improve the grey wolf algorithm and combine the grey wolf algorithm with the machine learning model. The algorithm constitutes three machine learning hybrid models: GWO-ET, GWO-AdaBoost, and GWO-GBM. Compared with traditional grid tuning, particle swarm optimization (PSO), and genetic algorithm (GA) optimization, the robustness and accuracy of the three optimization algorithms and the machine learning hybrid algorithm on the data set are compared and analyzed. Performance is measured through various performance comparisons and experimental result comparisons. For various building beam component data sets under real fires, the optimization and comparison show that the mean square error (MSE) of the proposed algorithm is extremely small. The results indicate that the GWO machine learning hybrid model is superior to other models and has a smaller prediction error.