Random Forest Machine Learning Algorithm Research Articles

A Machine Learning Approach to Understanding the Physical Properties of Magnetic Flux Ropes in the Solar Wind at 1 au

Interplanetary magnetic flux ropes (MFRs) are commonly observed structures in the solar wind, categorized as magnetic clouds (MCs) and small-scale MFRs (SMFRs) depending on whether they are associated with coronal mass ejections. We apply machine learning to systematically compare SMFRs, MCs, and ambient solar wind plasma properties. We construct a data set of 3-minute averaged sequential data points of the solar wind’s instantaneous bulk fluid plasma properties using about 20 years of measurements from Wind. We label samples by the presence and type of MFRs containing them using a catalog based on Grad–Shafranov (GS) automated detection for SMFRs and NASA's catalog for MCs (with samples in neither labeled non-MFRs). We apply the random forest machine learning algorithm to find which categories can be more easily distinguished and by what features. MCs were distinguished from non-MFRs with an area under the receiver-operator curve (AUC) of 94% and SMFRs with an AUC of 89%, and had distinctive plasma properties. In contrast, while SMFRs were distinguished from non-MFRs with an AUC of 86%, this appears to rely solely on the 〈B〉 > 5 nT threshold applied by the GS catalog. The results indicate that SMFRs have virtually the same plasma properties as the ambient solar wind, unlike the distinct plasma regimes of MCs. We interpret our findings as additional evidence that most SMFRs at 1 au are generated within the solar wind. We also suggest that they should be considered a salient feature of the solar wind’s magnetic structure rather than transient events.

Boreal to temperate forests transition within Kunashir Island (the Kuril Archipelago): Insights from the novel vegetation map

The vegetation of the Kuril Islands remains insufficiently explored due to their remote location and harsh environment. Understanding the factors that influence vegetation distribution in this extensive volcanic chain of islands, spanning over a thousand kilometers, is important for global biogeography. Our paper presents the first vegetation map of Kunashir Island from the Kuril archipelago, compiled using Sentinel-2 medium-resolution multispectral satellite imagery, incorporating original field research data and the Random Forest machine-learning algorithm. The temperate broad-leaved forests are mainly composed of Acer mono subsp. mayrii and Quercus mongolica, predominantly limited to the southwestern coast, and true-boreal dark-coniferous Abies sachalinensis–Picea jezoensis forests located in the northeast part of Kunashir. The vegetation distribution pattern is associated with the warming effect of the Soja current on the east coast of Kunashir Island and the cooling effect of the Kuril current on the west coast. Kunashir's vegetation displays features of an ecotone between temperate and boreal zones. Still, the boundary between these zones should be drawn within the island, rather than between Kunashir and Hokkaido or Kunashir and Iturup, as suggested in some previous biogeographic generalizations.

Power PV Forecasting using Machine Learning Algorithms Based on Weather Data in Semi-Arid Climate

As the energy demand continues to rise, renewable energy sources such as photovoltaic (PV) systems are becoming increasingly popular. PV systems convert solar radiation into electricity, making them an attractive option for reducing reliance on traditional electricity sources and decreasing carbon emissions. To optimize the usage of PV systems, intelligent forecasting algorithms are essential. They enable better decisionmaking regarding cost and energy efficiency, reliability, power optimization, and economic smart grid operations. Machine learning algorithms have proven to be effective in estimating the power of PV systems, improving accuracy by allowing models to understand complex relationships between parameters and evaluate the output power performance of photovoltaic cells. This work presents a study on the use of machine learning algorithms Catboost, LightGBM, XGboost and Random Forest to improve prediction. The study results indicate that using machine learning algorithms LightGBM can improve the accuracy of PV power prediction, which can have significant implications for optimizing energy usage. In addition to reducing uncertainty, machine learning algorithms improve PV systems’ efficiency, reliability, and economic viability, making them more attractive as renewable energy sources.

Predicting hotspots for disease-causing single nucleotide variants using sequences-based coevolution, network analysis, and machine learning.

To enable personalized medicine, it is important yet highly challenging to accurately predict disease-causing mutations in target proteins at high throughput. Previous computational methods have been developed using evolutionary information in combination with various biochemical and structural features of protein residues to discriminate neutral vs. deleterious mutations. However, the power of these methods is often limited because they either assume known protein structures or treat residues independently without fully considering their interactions. To address the above limitations, we build upon recent progress in machine learning, network analysis, and protein language models, and develop a sequences-based variant site prediction workflow based on the protein residue contact networks: 1. We employ and integrate various methods of building protein residue networks using state-of-the-art coevolution analysis tools (RaptorX, DeepMetaPSICOV, and SPOT-Contact) powered by deep learning. 2. We use machine learning algorithms (Random Forest, Gradient Boosting, and Extreme Gradient Boosting) to optimally combine 20 network centrality scores to jointly predict key residues as hot spots for disease mutations. 3. Using a dataset of 107 proteins rich in disease mutations, we rigorously evaluate the network scores individually and collectively (via machine learning). This work supports a promising strategy of combining an ensemble of network scores based on different coevolution analysis methods (and optionally predictive scores from other methods) via machine learning to predict hotspot sites of disease mutations, which will inform downstream applications of disease diagnosis and targeted drug design.

Emotion, Cultural Valuation of Being Human and AI Services

This study alerts that successful AI services depend not only on the improvement of the ability of AI to understand customers’ emotions but also on whether the utility function of the customer allows for the perfect substitution between an AI-delivered service and a human-delivered service. We argue that the utility function of customers includes the subjective valuation of being human, which AI services cannot meet by nature. A culture-based development (CBD) model for the transformation of emotions into feelings and order of preferences is offered to explain the cultural valuation of being human as a component of customers’ utility function. Using primary data from an experimental survey and employing two alternative machine learning algorithms (Lasso and Random Forest), we operationalize our CBD model to show that the estimation efficiency of the AI algorithms cannot compensate for the omission of the cultural valuation of being human from the modeling of the utility function of the customer.

Inversion model of soil salinity in alfalfa covered farmland based on sensitive variable selection and machine learning algorithms.

Timely and accurate monitoring of soil salinity content (SSC) is essential for precise irrigation management of large-scale farmland. Uncrewed aerial vehicle (UAV) low-altitude remote sensing with high spatial and temporal resolution provides a scientific and effective technical means for SSC monitoring. Many existing soil salinity inversion models have only been tested by a single variable selection method or machine learning algorithm, and the influence of variable selection method combined with machine learning algorithm on the accuracy of soil salinity inversion remain further studied. Firstly, based on UAV multispectral remote sensing data, by extracting the spectral reflectance of each sampling point to construct 30 spectral indexes, and using the pearson correlation coefficient (PCC), gray relational analysis (GRA), variable projection importance (VIP), and support vector machine-recursive feature elimination (SVM-RFE) to screen spectral index and realize the selection of sensitive variables. Subsequently, screened and unscreened variables as model input independent variables, constructed 20 soil salinity inversion models based on the support vector machine regression (SVM), back propagation neural network (BPNN), extreme learning machine (ELM), and random forest (RF) machine learning algorithms, the aim is to explore the feasibility of different variable selection methods combined with machine learning algorithms in SSC inversion of crop-covered farmland. To evaluate the performance of the soil salinity inversion model, the determination coefficient (R2), root mean square error (RMSE) and performance deviation ratio (RPD) were used to evaluate the model performance, and determined the best variable selection method and soil salinity inversion model by taking alfalfa covered farmland in arid oasis irrigation areas of China as the research object. The variable selection combined with machine learning algorithm can significantly improve the accuracy of remote sensing inversion of soil salinity. The performance of the models has been improved markedly using the four variable selection methods, and the applicability varied among the four methods, the GRA variable selection method is suitable for SVM, BPNN, and ELM modeling, while the PCC method is suitable for RF modeling. The GRA-SVM is the best soil salinity inversion model in alfalfa cover farmland, with Rv 2 of 0.8888, RMSEv of 0.1780, and RPD of 1.8115 based on the model verification dataset, and the spatial distribution map of soil salinity can truly reflect the degree of soil salinization in the study area. Based on our findings, the variable selection combined with machine learning algorithm is an effective method to improve the accuracy of soil salinity remote sensing inversion, which provides a new approach for timely and accurate acquisition of crops covered farmland soil salinity information.

Development and validation of a random survival forest model for predicting long-term survival of early-stage young breast cancer patients based on the SEER database and an external validation cohort.

Young breast cancer (YBC) patients often face a poor prognosis, hence it's necessary to construct a model that can accurately predict their long-term survival in early stage. To realize this goal, we utilized data from the Surveillance, Epidemiology, and End Results (SEER) databases between January 2010 and December 2020, and meanwhile, enrolled an independent external cohort from Tianjin Medical University Cancer Institute and Hospital. The study aimed to develop and validate a prediction model constructed using the Random Survival Forest (RSF) machine learning algorithm. By applying the Least Absolute Shrinkage and Selection Operator (LASSO) regression analysis, we pinpointed key prognostic factors for YBC patients, which were used to create a prediction model capable of forecasting the 3-year, 5-year, 7-year, and 10-year survival rates of YBC patients. The RSF model constructed in the study demonstrated exceptional performance, achieving C-index values of 0.920 in the training set, 0.789 in the internal validation set, and 0.701 in the external validation set, outperforming the Cox regression model. The model's calibration was confirmed by Brier scores at various time points, showcasing its excellent accuracy in prediction. Decision curve analysis (DCA) underscored the model's importance in clinical application, and the Shapley Additive Explanations (SHAP) plots highlighted the importance of key variables. The RSF model also proved valuable in risk stratification, which has effectively categorized patients based on their survival risks. In summary, this study has constructed a well-performed prediction model for the evaluation of prognostic factors influencing the long-term survival of early-stage YBC patients, which is significant in risk stratification when physicians handle YBC patients in clinical settings.

Incorporating landscape ecological approach in machine learning classification for agricultural land-use mapping based on a single date imagery

Land-use maps containing crop rotational information are very important in land management and physical planning. Such maps are usually generated using multitemporal data. Although recent technology allows analysts to process multitemporal information effectively, the use of single date imagery for such purpose is more efficient. This study aimed to map detailed agricultural land-use with crop rotational information based on a single date Landsat 8 imagery and SRTM-derived terrain attributes. A landscape ecological approach assuming the influence of terrain characteristics on the existence of crop and land-use types was implemented in multisource classification using random decision forest (RDF) machine learning algorithm. The use of seven optical bands and five terrain attributes could provide a land-use map at 88.03% accuracy, compared to seven optical bands only that generate 82.45% accuracy. These results are also better than those of maximum likelihood. The most influential variables in the achieved accuracy are elevation and thermal band.

Analisis Static Malware Menggunakan Algoritma Random Forest Machine Learning

Malware is a malicious software program that aims to damage or steal data from a victim's system. Malware analysis is needed to determine the performance of the malware. Static analysis was used for feature extraction from a total of 10 PEfiles, 5 malware files and 5 benign files. The random forest machine learning algorithm obtained an accuracy of 99.99%.

잠재전이분석과 머신러닝 기법을 적용한 초-중 학교급 전환에 따른 자기조절학습 변화양상과 예측변인 탐색

The purpose of this study is to examine the transition patterns of self-regulated learning (SRL) profiles among students transitioning from elementary to middle school and to identify factors influencing these changes. To do this, the「Korea Education Longitudinal Study 2013」data was utilized. Latent transition analysis and six machine learning algorithms (Random Forest, Decision Tree, Support Vector Machine, Logistic Regression, K-Nearest Neighbors, and Artificial Neural Network) were approached for analysis. As results, first, the SRL profiles of elementary to middle school transition students were categorized into three levels: high, medium, and low. Second, Random Forest(RF) demonstrated the highest performance in classifying the transition groups of SRL patterns. Third, examining RF analysis to investigate factors influencing the transition of SRL profiles, academic achievement emerged as the most significant predictor, followed by academic self-concept, self-management, creativity, and comprehension of class materials, among others. Finally, based on the study’s results, educational implications for fostering positive development of students’ SRL were suggested.

Investigation of Laser Ablation Quality Based on Data Science and Machine Learning XGBoost Classifier

This work proposes a matching data science approach for the laser ablation quality, reb, the study of Si3N4 film based on supervised machine learning classifiers in the CMOS-MEMS process. The study demonstrates that there exists an energy threshold, Eth, for laser ablation. If the laser energy surpasses this threshold, increasing the interval time will not contribute significantly to the recovery of pulse laser energy. Thus, reb enhancement is limited. When the energy is greater than 0.258 mJ, there exists a critical value of interval time at which the reb value is relatively low for each energy level, respectively. In addition, the variation of reb, Δreb, is independent of the interval time at the invariant point of energy between 0.32 mJ and 0.36 mJ. Energy and interval time exhibit a Pearson correlation of 0.82 and 0.53 with reb, respectively. To maintain Δreb below 0.15, green laser ablation of Si3N4 at operating energies of 0.258–0.378 mJ can adopt a baseline interval time of the initial baseline multiplied by 1/∜2. Additionally, for operating energies of 0.288–0.378 mJ during Si3N4 laser ablation, Δreb can be kept below 0.1. With the forced partition methods, namely, the k-means method and percentile method, the XGBoost (v 2.0.3) classifier maintains a competitive accuracy across test sizes of 0.20–0.40, outperforming the machine learning algorithms Random Forest and Logistic Regression, with the highest accuracy of 0.78 at a test size of 0.20.

Bullet ricochet mark plan-view morphology in concrete: an experimental assessment of five bullet types and two distances using machine learning.

Bullet ricochet impact sites can help with shooting incident reconstruction.A random forest machine learning algorithm classified bullet type from ricochet morphology.Results suggest that unknown bullets can potentially be determined from ricochet impact site morphology.Human identification of bullet types from ricochet sites may be erroneous.

Machine Learning-Based Rice Field Mapping in Kulon Progo using a Fusion of Multispectral and SAR Imageries

The land-conversion of rice fields can reduce rice production and negatively impact food security. Consequently, monitoring is essential to prevent the loss of productive agricultural land. This study uses a combination of Sentinel-2 MSI, Sentinel-1 SAR, along with SRTM (elevation and slope data) to monitor rice fields land-conversion. NDVI, NDBI and NDWI indices are transformed from the annual median composite Sentinel-2 MSI images used to identify different rice fields with another object. A monthly median composite of SAR images from Sentinel-1 data are used to identify cropping patterns of rice fields in the inundation phase. The classification is performed by using the Random Forest machine learning algorithm in the Google Earth Engine (GEE) platform. Random Forest classification is run using 1000 trees, with a 70:30 ratio of training and testing data from sample features extracted by visual interpretation of high resolution Google Earth imagery. In this study, Random Forest classification is effective in computing a high amount of multi-temporal and multi-sensory data to map rice-field land conversion with an accuracy rate of 96.16% (2021) and 95.95% (2017) for mapping paddy fields. From the multitemporal rice field maps in 2017—2021, a conversion of 826.66 hectares of rice-fields to non-rice fields was identified. Based on the spatial distribution, the conversion from rice-field to non-rice field is higher at the area near the roads, built area and Yogyakarta International Airport. Therefore, it is important to assess and ensure that National Strategic Projects are managed with due regard to environmental impacts and food security.

Application of Satellite Images and Artificial Intelligence to Monitor Land Cover Changes in Hanoi Area During 2013-2023 Period

Artificial intelligence (AI) and remote sensing technology have now increasingly improved their efficiency and reliability in monitoring the changes in land cover. With the amendment of the Vietnamese Law on Land in 2013 and the administrative boundary expansion of Hanoi, Hanoi experiences significant changes in land use and land cover for the last ten years. To monitor the actual land use changes in the area, this study used the Random Forest (RF) machine learning algorithm to classify the basic land covers, monitor, and analyze the spatial variation of land use and land cover in the 2013 to 2023 period. The study findings indicate a relatively high rate of expansion of construction zone area and a decrease in land cover related to water bodies and vegetated area. Water bodies decrease by an average of 0.8% annually, whereas the construction zone area increased by 7% of the total area.

Using ensemble machine learning algorithm to predict forest fire occurrence probability in Madhya Pradesh and Chhattisgarh, India

The escalating global threat of forest fires, driven by global warming, requires the development of effective prediction systems to mitigate damages. This research focuses on Madhya Pradesh (MP) and Chhattisgarh (CG) states in central India, where forest fire risk has become particularly pronounced. The primary objectives of the study are to quantify and map the spatial and temporal dynamics of forest fires over the period 2001 to 2020, and to predict future fire risks using satellite derived datasets and machine learning techniques. Through a long-term analysis, the study revealed an alarming increase in the number of forest fire incidents in MP and CG. From an average of 1200 and 1000 during 2001 to 2005, the incidents increased to 2800 and 2100 during 2016 to 2020, in MP and CG respectively. To predict forest fire risk, Random Forest machine learning algorithm was adopted utilizing various satellite derived climatic, topographical, and ecological parameters such as temperature, precipitation, solar radiation, NDVI, soil moisture, litter availability, evapotranspiration and terrain parameters (at monthly scale for 20 years). While forecasting fire probability for 2018–2020, the model achieves high accuracy rate of 86.46 % in MP and 93.78 % in CG. The results highlight significant forest fire likelihood regions in the central MP and the Southern CG, identifying areas requiring enhanced fire management strategies. This study has revealed that NDVI and rainfall have played a positive role in restricting the forest fire, and their negative anomaly amplified the fire risk. The study would help forest planners and administrators to characterise vulnerable areas and prioritise their conservation provisions.

A new perspective in radon risk assessment: Mapping the geological hazard as a first step to define the collective radon risk exposure

Radon is a radioactive gas and a major source of ionizing radiation exposure for humans. Consequently, it can pose serious health threats when it accumulates in confined environments. In Europe, recent legislation has been adopted to address radon exposure in dwellings; this law establishes national reference levels and guidelines for defining Radon Priority Areas (RPAs). This study focuses on mapping the Geogenic Radon Potential (GRP) as a foundation for identifying RPAs and, consequently, assessing radon risk in indoor environments. Here, GRP is proposed as a hazard indicator, indicating the potential for radon to enter buildings from geological sources. Various approaches, including multivariate geospatial analysis and the application of artificial intelligence algorithms, have been utilised to generate continuous spatial maps of GRP based on point measurements. In this study, we employed a robust multivariate machine learning algorithm (Random Forest) to create the GRP map of the central sector of the Pusteria Valley, incorporating other variables from census tracts such as land use as a vulnerability factor, and population as an exposure factor to create the risk map. The Pusteria Valley in northern Italy was chosen as the pilot site due to its well-known geological, structural, and geochemical features. The results indicate that high Rn risk areas are associated with high GRP values, as well as residential areas and high population density. Starting with the GRP map (e.g., Rn hazard), a new geological-based definition of the RPAs is proposed as fundamental tool for mapping Collective Radon Risk Areas in line with the main objective of European regulations, which is to differentiate them from Individual Risk Areas.

Machine learning prediction of mafic–ultramafic rock-related Cr-spinel formation environments and its application to the tectonic settings of magmatic sulfide deposits

This study uses the random forest machine learning algorithm to classify and predict Cr-spinel formation environments in mafic–ultramafic rocks. Cr-spinel is an early-crystallized oxide phase in these rocks and exhibits significant compositional variations that provide valuable insights into their formation environments. Traditionally, empirical diagrams have been used to classify the origins of Cr-spinel based on major and trace element compositions; however, overlap exists among spinel from different formation environments. To overcome this limitation, we develop a random forest model using a compiled dataset of Cr-spinel compositions from various tectonic settings, including small intrusions from continental collisional zones, layered intrusions and small intrusions from intra-continental plates (including large igneous provinces and intra-continental rifts), Alaskan-type complexes from oceanic subduction zones, and ophiolites from oceanic spreading centers. The model is trained, validated, and optimized using cross-validation and hyperparameter tuning. The results show high accuracy in predicting the Cr-spinel formation environments, as demonstrated by high precision, recall, and F1 scores. Feature importance analyses reveal that elements, such as Ti, Cr, and Al, play a more significant role in the prediction. Visualization techniques, i.e., the t-distributed stochastic neighbor embedding (TSNE) algorithm, are applied to reduce the dimensionality of the dataset and create decision boundaries for better interpretation. The model is validated using an independent Cr-spinel composition dataset obtained from an area with known geological settings, to confirm the model’s reliability. The model is applied successfully to reveal the heterogeneous sources of the rifting-related Daxueshan magmatic Ni-Cu sulfide deposit in the Baoshan block in the eastern part of the Tethyan orogenic belt. The machine learning algorithm can accurately distinguish different Cr-spinel types using large and diverse datasets to provide valuable insights into geological formation conditions. This approach is a powerful tool for understanding Earth’s geological processes, improving exploration models of magmatic sulfide deposit and reducing mineral exploration costs.

ASSOCIATION AND RISK PREDICTION OF PAIN AND FALL-RELATED INJURIES AMONG COMMUNITY-DWELLING OLDER ADULTS IN CHINA

Abstract Background Pain is common in older adults and has also been identified as a fall risk factor, whereas the impact and mechanism of fall-related injuries are still unknown. This study aims to examine the association and risk prediction of pain and fall-related injuries among Chinese older adults. Method This study enrolled 5,721 older adults from the China Health and Retirement Longitudinal Study (wave 2011-2015). Logistic regression was used to analyze the associations between pain characteristics and fall-related injuries. Furthermore, 2022 older adults with pain at baseline were further used to build a fall-related injury prediction model, using five machine learning algorithms (logistic regression, naive bayesian, random forest, adaptive boosting and artificial neural network) with 118 input variables as candidate features, and SHapley Additive exPlanations was used for the prediction model explanation. Results Logistic regression models showed that whether pain (OR=1.35) or pain severity (ORmoderate=1.37, ORsevere =1.48) and multisite pain (OR=1.41) independently predicted fall-related injuries. The Random forest model achieved the best performance among the prediction model with an area under the receiver operating characteristic curve of 0.764. Interpretable results showed that the top 10 significant predictors were: experience of falling, frailty, sex, short physical performance battery, received inpatient care, estimated glomerular filtration rate, lung function, pain quantity, White Blood Cell and Age. Conclusions Pain is associated with fall-related injuries among Chinese older adults. Fall-related injury prevention strategies for older adults with pain should specifically focus on fall history, frailty, and women.

Machine learning models to predict ligand binding affinity for the orexin 1 receptor

The orexin 1 receptor (OX1R) is a G-protein coupled receptor that regulates a variety of physiological processes through interactions with the neuropeptides orexin A and B. Selective OX1R antagonists exhibit therapeutic effects in preclinical models of several behavioral disorders, including drug seeking and overeating. However, currently there are no selective OX1R antagonists approved for clinical use, fueling demand for novel compounds that act at this target. In this study, we meticulously curated a dataset comprising over 1300 OX1R ligands using a stringent filter and criteria cascade. Subsequently, we developed highly predictive quantitative structure-activity relationship (QSAR) models employing the optimized hyper-parameters for the random forest machine learning algorithm and twelve 2D molecular descriptors selected by recursive feature elimination with a 5-fold cross-validation process. The predictive capacity of the QSAR model was further assessed using an external test set and enrichment study, confirming its high predictivity. The practical applicability of our final QSAR model was demonstrated through virtual screening of the DrugBank database. This revealed two FDA-approved drugs (isavuconazole and cabozantinib) as potential OX1R ligands, confirmed by radiolabeled OX1R binding assays. To our best knowledge, this study represents the first report of highly predictive QSAR models on a large comprehensive dataset of diverse OX1R ligands, which should prove useful for the discovery and design of new compounds targeting this receptor.

Understanding Active Transportation to School Behavior in Socioeconomically Disadvantaged Communities: A Machine Learning and SHAP Analysis Approach

Active Transportation to School (ATS) offers numerous health benefits and is considered an affordable option, especially in disadvantaged neighborhoods. The US Centers for Disease Control and Prevention (CDC) advises 60 min of daily physical exercise for children aged 6 to 17, making ATS a compelling approach to promote a healthier lifestyle among students. Initiated in 2005 by the US Department of Transportation (DOT), the Safe Routes to School (SRTS) program aims to foster safe and regular walking and biking to school for students. This paper examines students’ travel behavior using SRTS survey data and assesses the program’s effectiveness in promoting ATS in Chula Vista, California. Employing machine learning algorithms (random forest, logistic regression, and support vector machines) to predict students’ likelihood to walk to school, it utilizes SHAP (SHapley Additive exPlanations) to pinpoint significant variables influencing ATS across all models. SHAP underscores critical factors affecting transportation choices to school, highlighting the importance of home-to-school distance, with shorter distances positively impacting active transportation. However, only half of students within schools’ walking distance opted to walk to school, underscoring the necessity of addressing parental safety concerns, including factors such as crime rates and traffic speed along the route.