Random Forest Machine Learning Model Research Articles

Quantitative microbiome profiling of honey bee (Apis mellifera) guts is predictive of winter colony loss in northern Virginia (USA)

For the past 15 years, the proportion of honey bee hives that fail to survive winter has averaged ~ 30% in the United States. Winter hive loss has significant negative impacts on agriculture, the economy, and ecosystems. Compared to other factors, the role of honey bee gut microbial communities in driving winter hive loss has received little attention. We investigate the relationship between winter survival and honey bee gut microbiome composition of 168 honey bees from 23 hives, nine of which failed to survive through winter 2022. We found that there was a substantial difference in the abundance and community composition of honey bee gut microbiomes based on hive condition, i.e., winter survival or failure. The overall microbial abundance, as assessed using Quantitative Microbiome Profiling (QMP), was significantly greater in hives that survived winter 2022 than in those that failed, and the average overall abundance of each of ten bacterial genera was also greater in surviving hives. There were no significant differences in alpha diversity based on hive condition, but there was a highly significant difference in beta diversity. The bacterial genera Commensalibacter and Snodgrassella were positively associated with winter hive survival. Logistic regression and random forest machine learning models on pooled ASV counts for the genus data were highly predictive of winter outcome, although model performance decreased when samples from the location with no hive failures were excluded from analysis. As a whole, our results show that the abundance and community composition of honey bee gut microbiota is associated with winter hive loss, and can potentially be used as a diagnostic tool in evaluating hive health prior to the onset of winter. Future work on the functional characterization of the honey bee gut microbiome’s role in winter survival is warranted.

Preoperative prediction model for risk of readmission after total joint replacement surgery: a random forest approach leveraging NLP and unfairness mitigation for improved patient care and cost-effectiveness

BackgroundThe Center for Medicare and Medicaid Services (CMS) imposes payment penalties for readmissions following total joint replacement surgeries. This study focuses on total hip, knee, and shoulder arthroplasty procedures as they account for most joint replacement surgeries. Apart from being a burden to healthcare systems, readmissions are also troublesome for patients. There are several studies which only utilized structured data from Electronic Health Records (EHR) without considering any gender and payor bias adjustments.MethodsFor this study, dataset of 38,581 total knee, hip, and shoulder replacement surgeries performed from 2015 to 2021 at Novant Health was gathered. This data was used to train a random forest machine learning model to predict the combined endpoint of emergency department (ED) visit or unplanned readmissions within 30 days of discharge or discharge to Skilled Nursing Facility (SNF) following the surgery. 98 features of laboratory results, diagnoses, vitals, medications, and utilization history were extracted. A natural language processing (NLP) model finetuned from Clinical BERT was used to generate an NLP risk score feature for each patient based on their clinical notes. To address societal biases, a feature bias analysis was performed in conjunction with propensity score matching. A threshold optimization algorithm from the Fairlearn toolkit was used to mitigate gender and payor biases to promote fairness in predictions.ResultsThe model achieved an Area Under the Receiver Operating characteristic Curve (AUROC) of 0.738 (95% confidence interval, 0.724 to 0.754) and an Area Under the Precision-Recall Curve (AUPRC) of 0.406 (95% confidence interval, 0.384 to 0.433). Considering an outcome prevalence of 16%, these metrics indicate the model’s ability to accurately discriminate between readmission and non-readmission cases within the context of total arthroplasty surgeries while adjusting patient scores in the model to mitigate bias based on patient gender and payor.ConclusionThis work culminated in a model that identifies the most predictive and protective features associated with the combined endpoint. This model serves as a tool to empower healthcare providers to proactively intervene based on these influential factors without introducing bias towards protected patient classes, effectively mitigating the risk of negative outcomes and ultimately improving quality of care regardless of socioeconomic factors.

Determining target levels of power distribution system reliability indices using machine learning

Electrical distribution system (EDS) reliability has always been a critical subject due to the considerable share of distribution systems in power interruptions to end-users. Meeting regulatory service quality standards requires distribution companies to achieve defined reliability levels for performance improvement. Conventionally, specialists determine target reliability based on experience, experimental knowledge, and rules of thumb. However, these methods are error-prone and may result in suboptimal investments, economic inefficiency, and inadequate resource allocation in EDS operation and planning. Accordingly, this paper presents a machine learning-based method to determine the target level for EDS's reliability indices based on its historical performance. The approach involves comprehensive analysis of historical EDS reliability data, followed by a data envelopment analysis (DEA) to assess performance across subdivisions like substations or feeders, identifying the most effective units. Subsequently, machine learning models—Random Forest (RF), Support Vector Machine (SVM), and Radial Basis Function (RBF) network—are trained on optimal subdivisions from the DEA stage. These models accurately estimate target reliability for all subdivisions, contributing to the overall determination of the EDS target reliability level. A three-scenario case study validates the proposed method, suggesting its potential applicability as a valuable tool for regulatory agencies in setting EDS reliability targets. While admitting the potential challenges, notably the method's reliance on the quality and completeness of the historical data collected by distribution companies, the paper emphasizes the importance of collaborative efforts to ensure data accuracy and sufficiency. The proposed method leads to a fair mechanism to determine the target level for a real EDS's reliability indices in comparison with a business-as-usual scenario. Moreover, it provides a reliability performance ranking for the EDS's subdivisions.

Comparison and analysis of prediction accuracy between traditional machine learning algorithms and XGBoost algorithm in music emotion classification

Music emotion classification refers to determining the emotional types of music, such as happiness, sadness, anger, and passion, based on aspects like rhythm, melody, and tone. The development of artificial intelligence can be applied to music recommendation by employing machine learning algorithms to ascertain emotional attributes of music, enabling precise music suggestions for users to enhance their musical experience. This paper compares and analyzes the classification performance of traditional Random Forest machine learning algorithms and the XGBoost model on a Turkish music emotion dataset. We selected a comprehensive dataset from Kaggle, containing a vast array of music samples and their corresponding emotional labels, making it an extensive music emotion classification dataset. From the experimental results, it's evident that the traditional Random Forest machine learning model outperforms the XGBoost model in terms of accuracy, precision, and recall. The accuracy of the traditional Random Forest machine learning model stands at 80.8%, whereas the XGBoost model's accuracy is 75%. The recall rate for the traditional Random Forest machine learning model is 80.8%, while for XGBoost, it's 77.2%. The F1 score for the traditional Random Forest machine learning model is 80.5%, whereas for XGBoost, it's 75.3%. These parameters indicate that the traditional Random Forest machine learning model exhibits superior predictive performance in music emotion classification. However, the XGBoost model possesses its own advantages, such as faster learning and prediction speeds, high accuracy, and strong generalization capabilities. In summary, the traditional Random Forest machine learning model demonstrates better robustness and interpretability, effectively handling samples with noise and missing data, thus finding widespread practical application. On the other hand, the XGBoost model excels in rapid training and prediction, coupled with higher accuracy and versatility, making it advantageous in dealing with large datasets. The research outcomes of this paper hold significant importance for the study and application of music emotion classification. The experimental results presented herein offer valuable insights for researchers and practitioners, aiding them in selecting appropriate machine learning models, optimizing, and adjusting them to achieve the best classification results.

Relatedness and product complexity meet gravity models of international trade

Researchers have long used gravity models to analyze international trade patterns, identify export opportunities, and negotiate trade agreements. Recent research has emphasized the significance of relatedness and product complexity research in developing robust economic development strategies. This paper presents a novel approach, incorporating relatedness and product complexity as integral elements for interpreting export potential within gravity models powered by machine learning. Our approach stands out for its proficiency in accurately predicting bilateral trade values at a detailed product group level, providing valuable insights for policymakers and other stakeholders. The research leverages random forest machine learning models for predictions and incorporates relatedness and complexity to reveal new dimensions in international trade analysis.

Role of environmental and stand factors on forest dieback: An approach using structural equation modelling and machine learning

The Zagros forest region in western Iran is facing dieback processes, impacting both the environment and the local economy. Understanding the cause of forest dieback remains a challenge for forest experts due to the wide set of factors usually involved in the dieback phenomenon. To achieve this goal, we measured woody species dieback intensity (DI) using 124 plots distributed in oak forests across a wide range of environmental conditions. On these plots, we measured stand and vegetation factors (e.g. woody species richness, WSR), various soil properties and topographic indices (in particular the SAGA wetness index) and climatic factors. We then modeled dieback intensity using structural equation modeling (SEM) and machine learning (ML). We found a significant positive correlation between WSR and DI (r=0.48), and very low to low but still significant negative correlations with the SAGA wetness index (r=-0.10) and soil clay content (r=-0.16), respectively. Other soil properties and climate factors showed no significant influence on DI. After six rounds of respecification, the SEM fitting gave satisfactory results based on the validation map of DI indices (Goodness-of-fit Index = 0.948 and Comparative fit index =0.958). The DI map was generated using a random forest (RF) machine learning model to achieve an accuracy of 75% and a kappa index of 0.45, and then categorized into four classes. The map showed that the intensity of dieback was contrasting in our area: 41.1% of plots had no dieback, 58.4% had low DI, and less than 1% had moderate or high DI. We also found that soil clay content and woody species richness were crucial factors, explaining around 50% of the variance in DI. In conclusion, combining the methods from SEM and RF models provides an efficient approach to understand and identify key factors in the complex process of dieback.

Prediction of soil organic carbon using machine learning techniques and geospatial data for sustainable agriculture

Sustainable agriculture revolves around soil organic carbon (SOC), which is essential for numerous soil functions and ecological attributes. Farmers are interested in conserving and adding additional soil organic carbon to certain fields in order to improve soil health and productivity. The relationship between soil and environment that has been discovered and standardized throughout time has enhanced the progress of digital soil-mapping techniques; therefore, a variety of machine learning techniques are used to predict soil properties. Studies are thriving at how effectively each machine learning method maps and predicts SOC, especially at high spatial resolutions. To predict SOC of soil at a 30 m resolution, four machine learning models—Random Forest, Support Vector Machine, Adaptive Boosting, and k-Nearest Neighbour were used. For model evaluation, two error metrics, namely R2 and RMSE have been used. The findings demonstrated that the calibration and validation sets’ descriptive statistics sufficiently resembled the entire set of data. The range of the calculated SOC content was 0.06 to 1.76 %. According to the findings of the study, Random Forest showed good results for both cases, i.e. evaluation using cross validation and without cross validation. Using cross validation, RF confirmed highest R2 as 0.5278 and lowest RMSE as 0.1683 for calibration dataset while without cross validation it showed R2 as 0.8612 and lowest RMSE as 0.0912 for calibration dataset. The generated soil maps will help farmers adopt precise knowledge for decisions that will increase farm productivity and provide food security through the sustainable use of nutrients and the agricultural environment.

Characterizing land use/land cover change dynamics by an enhanced random forest machine learning model: a Google Earth Engine implementation

Land use and land cover (LULC) analysis is crucial for understanding societal development and assessing changes during the Anthropocene era. Conventional LULC mapping faces challenges in capturing changes under cloud cover and limited ground truth data. To enhance the accuracy and comprehensiveness of the descriptions of LULC changes, this investigation employed a combination of advanced techniques. Specifically, multitemporal 30 m resolution Landsat-8 satellite imagery was utilized, in addition to the cloud computing capabilities of the Google Earth Engine (GEE) platform. Additionally, the study incorporated the random forest (RF) algorithm. This study aimed to generate continuous LULC maps for 2014 and 2020 for the Shrirampur area of Maharashtra, India. A novel multiple composite RF approach based on LULC classification was utilized to generate the final LULC classification maps utilizing the RF-50 and RF-100 tree models. Both RF models utilized seven input bands (B1 to B7) as the dataset for LULC classification. By incorporating these bands, the models were able to influence the spectral information captured by each band to classify the LULC categories accurately. The inclusion of multiple bands enhanced the discrimination capabilities of the classifiers, increasing the comprehensiveness of the assessment of the LULC classes. The analysis indicated that RF-100 exhibited higher training and validation/testing accuracy for 2014 and 2020 (0.99 and 0.79/0.80, respectively). The study further revealed that agricultural land, built-up land, and water bodies have changed adequately and have undergone substantial variation among the LULC classes in the study area. Overall, this research provides novel insights into the application of machine learning (ML) models for LULC mapping and emphasizes the importance of selecting the optimal tree combination for enhancing the accuracy and reliability of LULC maps based on the GEE and different RF tree models. The present investigation further enabled the interpretation of pixel-level LULC interactions while improving image classification accuracy and suggested the best models for the classification of LULC maps through the identification of changes in LULC classes.

Community coalescence and plant host filtering determine the spread of tetracycline resistance genes from pig manure into the microbiome continuum of the soil–plant system

The spread of livestock manure-borne antibiotic resistance genes (ARGs) into agroecosystems through manure application poses a potential threat to human health. However, there is still a knowledge gap concerning ARG dissemination in coalescing manure, soil and plant microbiomes. Here, we examined the fate of tetracycline resistance genes (TRGs) originating from pig manure microbiomes and spread in the soil-A thaliana system and explored the effects of microbial functions on TRGs spread at different interfaces. Our results indicate that the TRGs abundances in all microbiome continuum of the soil-A. thaliana system were significantly increased with the application of a living manure microbiome, although the addition of manure with both an active and inactive microbiome caused a shift in the microbial community composition. This was attributed to the increasing relative abundances of tetA, tetL, tetM, tetO, tetW and tolC in the system. The application of living manure with DOX residues resulted in the highest relative abundance of total TRGs (3.30×10−3 copies/16S rRNA gene copies) in the rhizosphere soil samples. Community coalescence of the manure and soil microbiomes increased the abundance of Firmicutes in the soil and root microbiome, which directly explains the increase in TRG abundance observed in these interfaces. In contrast, the leaf microbiome differed markedly from that of the remaining samples, indicating strong plant host filtering effects on Firmicutes and TRGs from pig manure. The random forest machine learning model revealed microbial functions and their significant positive correlation with TRG abundance in the microbiome continuum of the system. Our findings revealed that community coalescence is the main driver of TRG spread from manure to the soil and root microbiomes. Plant host filtering effects play a crucial role in allowing certain microbial groups to occupy ecological niches in the leaves, thereby limiting the establishment of manure-borne TRGs in aboveground plant tissues.

Predicting in-hospital mortality among patients admitted with a diagnosis of heart failure: a machine learning approach.

Existing risk prediction models for hospitalized heart failure patients are limited. We identified patients hospitalized with a diagnosis of heart failure between 7 May 2013 and 26 April 2022 from a large academic, quaternary care medical centre (training cohort). Demographics, medical comorbidities, vitals, and labs were collected and were used to construct random forest machine learning models to predict in-hospital mortality. Models were compared with logistic regression, and to commonly used heart failure risk scores. The models were subsequently validated in patients hospitalized with a diagnosis of heart failure from a second academic, community medical centre (validation cohort). The entire cohort comprised 21802 patients, of which 14539 were in the training cohort and 7263 were in the validation cohort. The median age (25th-75th percentile) was 70 (58-82) for the entire cohort, 43.2% were female, and 6.7% experienced inpatient mortality. In the overall cohort, 7621 (35.0%) patients had heart failure with reduced ejection fraction (EF≤40%), 1271 (5.8%) had heart failure with mildly reduced EF (EF 41-49%), and 12910 (59.2%) had heart failure with preserved EF (EF≥50%). Random forest models in the validation cohort demonstrated a c-statistic (95% confidence interval) of 0.96 (0.95-0.97), sensitivity (SN) of 87.3%, and specificity (SP) of 90.6% for the prediction of in-hospital mortality. Models for those with HFrEF demonstrated a c-statistic of 0.96 (0.94-0.98), SN 88.2%, and SP 91.0%, and those for patients with HFpEF showed a c-statistic of 0.95 (0.93-0.97), SN 87.4%, and SP 89.5% for predicting in-hospital mortality. The random forest model significantly outperformed logistic regression (c-statistic 0.87, SN 75.9%, and SP 86.9%), and current existing risk scores including the Acute Decompensated Heart Failure National Registry risk score (c-statistic of 0.70, SN 69%, and SP 62%), and the Get With the Guidelines-Heart Failure risk score (c-statistic 0.69, SN 67%, and SP 63%); P<0.001 for comparison. Machine learning models built from commonly recorded patient information can accurately predict in-hospital mortality among patients hospitalized with a diagnosis of heart failure.

Intelligent Energy Management: Leveraging an Effective Machine Learning for Predictive Appliance Energy Optimization in Smart Homes

In the context of escalating energy demands and the proliferation of smart home technologies, this study introduces a novel approach to energy management using the Random Forest machine learning model. Our research focuses on optimizing household appliance energy use, harmonizing efficiency with user comfort. By analyzing data on appliance usage patterns, environmental conditions, and user preferences, the Random Forest model predicts future energy needs, enabling the intelligent scheduling of appliances to reduce unnecessary consumption. The model’s strength lies in its capacity to unravel complex, non-linear relationships in high-dimensional data typical of household energy usage scenarios. Initial results demonstrate a notable decrease in energy consumption, affirming the model’s efficacy in enhancing energy efficiency without diminishing user convenience. This research not only highlights the potential of machine learning in energy management but also sets a foundation for future exploration into adaptive, real-time energy optimization strategies in smart homes. In this research, we compared our model with other machine learning models, and our model got a good accuracy of 95.71%, while for time series data, we got 99.71%.

Enhancing Reliability of Advanced Driver-Assistance Systems through Predictive Maintenance and Data-Driven Insights

The advancement of Advanced Driver Assistance Systems (ADAS) marks a pivotal evolution in automotive technology, aiming to enhance road safety and driving efficiency through a wide array of functionalities like blind spot detection, emergency braking, and adaptive cruise control. This research paper delves into the operational integrity, performance metrics, and maintenance strategies of ADAS components, underpinned by a comprehensive methodology involving data collection, pre-processing, feature engineering, machine learning model development, and rigorous validation processes. Systematic inspection of ADAS components indicates their importance in vehicle safety and reliability. The visibility, distance, speed, and steering angle of front cameras, LiDAR, radar, and ultrasonic sensors are carefully evaluated. Maintenance logs show proactive error code management, boosting efficiency. SVM, Gradient Boosting, and Random Forest machine learning models predicted ADAS component failures during validation and testing. Random Forest scored 90% accuracy, 92% precision, 88% recall, and 90% F1. Gradient Boosting was the most accurate, with 93% accuracy, 94% precision, 91% recall, and 92% F1. SVM predicted ADAS component failures with 88% accuracy, 90% precision, 85% recall, and 87% F1 score. Machine learning helps shift from reactive to proactive maintenance. Modelling sensor signal quality, actuator reaction times, error code frequencies, and maintenance intervals enables predictive maintenance and failure detection. Feature engineering builds predictive models using maintenance logs and operational KPIs. The models predict ADAS component failures, boosting vehicle safety and dependability. Using external data improves predictive maintenance models. The maintenance model's adaptability and forecast accuracy are proved by ADAS operation after traffic, accident, and manufacturer upgrades. Predictive maintenance and machine learning improve ADAS dependability and safety, the study found. Advanced analytics and data-driven insights can reduce automotive system failures, improving safety and reliability.

Identification of KRAS mutation-associated gut microbiota in colorectal cancer and construction of predictive machine learning model.

Gut microbiota has demonstrated an increasingly important role in the onset and development of colorectal cancer (CRC). Nonetheless, the association between gut microbiota and KRAS mutation in CRC remains enigmatic. We conducted 16S rRNA sequencing on stool samples from 94 CRC patients and employed the linear discriminant analysis effect size algorithm to identify distinct gut microbiota between KRAS mutant and KRAS wild-type CRC patients. Transcriptome sequencing data from nine CRC patients were transformed into a matrix of immune infiltrating cells, which was then utilized to explore KRAS mutation-associated biological functions, including Gene Ontology items and Kyoto Encyclopedia of Genes and Genomes pathways. Subsequently, we analyzed the correlations among these KRAS mutation-associated gut microbiota, host immunity, and KRAS mutation-associated biological functions. At last, we developed a predictive random forest (RF) machine learning model to predict the KRAS mutation status in CRC patients, based on the gut microbiota associated with KRAS mutation. We identified a total of 26 differential gut microbiota between both groups. Intriguingly, a significant positive correlation was observed between Bifidobacterium spp. and mast cells, as well as between Bifidobacterium longum and chemokine receptor CX3CR1. Additionally, we also observed a notable negative correlation between Bifidobacterium and GOMF:proteasome binding. The RF model constructed using the KRAS mutation-associated gut microbiota demonstrated qualified efficacy in predicting the KRAS phenotype in CRC. Our study ascertained the presence of 26 KRAS mutation-associated gut microbiota in CRC and speculated that Bifidobacterium may exert an essential role in preventing CRC progression, which appeared to correlate with the upregulation of mast cells and CX3CR1 expression, as well as the downregulation of GOMF:proteasome binding. Furthermore, the RF model constructed on the basis of KRAS mutation-associated gut microbiota exhibited substantial potential in predicting KRAS mutation status in CRC patients.IMPORTANCEGut microbiota has emerged as an essential player in the onset and development of colorectal cancer (CRC). However, the relationship between gut microbiota and KRAS mutation in CRC remains elusive. Our study not only identified a total of 26 gut microbiota associated with KRAS mutation in CRC but also unveiled their significant correlations with tumor-infiltrating immune cells, immune-related genes, and biological pathways (Gene Ontology items and Kyoto Encyclopedia of Genes and Genomes pathways). We speculated that Bifidobacterium may play a crucial role in impeding CRC progression, potentially linked to the upregulation of mast cells and CX3CR1 expression, as well as the downregulation of GOMF:Proteasome binding. Furthermore, based on the KRAS mutation-associated gut microbiota, the RF model exhibited promising potential in the prediction of KRAS mutation status for CRC patients. Overall, the findings of our study offered fresh insights into microbiological research and clinical prediction of KRAS mutation status for CRC patients.

Phishing URL Detection using Machine Learning

Abstract: Phishing attacks pose a significant threat in cyberspace, Cybercrimes such as Phishing means accessing personal data and violating security through the Internet and its main aim is to steal the information from the users using different techniques in that the primary one is Phishing which demands effective detection mechanisms. This study evaluates the performance of Gradient Boosting Classifier, Random Forest, and Decision Tree machine learning models in conjunction with feature selection techniques namely SelectKBest and Chi-Square. Initially, a comprehensive feature set of 30 attributes achieved a baseline accuracy of 97.4%. Through SelectKBest and Chi-Square feature selection methods, 13 key features were identified, leading to a slightly decreased accuracy of 95.6% upon model retraining. This research highlights the importance of feature selection in enhancing phishing detection accuracy while maintaining model interpretability. Technologies such as NumPy, Pandas, Matplotlib, Scikit-learn, and Flask drive. The project, emphasising the exploration of ML models, EDA (Exploratory Data Analysis) on phishing data, and understanding feature importance. The Machine Learning Models like Gradient Boosting Classifier, Random Forest and Decision tree are used to detect whether the given URL is Malicious URL or Legitimate URL using Comprehensive Feature set and Key Feature set. This Models and Feature sets are compared based on the Performance Metrics namely Accuracy, Precision, Recall and F1- score. This study contributes to advancing cyber defence mechanism through the fusion of sophisticated Machine Learning algorithms and meticulous Feature Selection methodologies.

EXPRESS: Amplifying Off-Site Purchases with On-Site Retail Media Advertising

Retail media is of growing interest to brands, and online retailers offer a large collection of product information that is freely available 24/7 for consumers' online and offline purchase decisions. In this study, the authors map webrooming across ten product categories by combining internal data from consumer browsing behavior at an online retailer and survey responses from the same customers about their offsite behavior. Moreover, they build a random forest machine-learning model to predict the magnitude of webrooming across categories. They find that webrooming is economically substantial. On average, for every 10 customers who research the product category on-site and buy on-site, 17 other customers research the product category on-site and buy off-site. As to retail media's impact, upper-funnel and always-on online ad forms are associated with a higher number of online searchers in the laptop category that buy the brand offline. Beyond finding support for their propositions, the authors provide directions for future research on the cross-channel effects of retail media and how they generalize across brands, categories and retailers.

A machine learning approach feature to forecast the future performance of the universities in Canada

University ranking is a technique of measuring the performance of Higher Education Institutions (HEIs) by evaluating them on various criteria like student satisfaction, expenditure, research and teaching quality, citation count, grants, and enrolment. Ranking has been determined as a vital factor that helps students decide which institution to attend. Hence, universities seek to increase their overall rank and use these measures of success in their marketing communications and prominently place their ranked status on their institution's websites. Despite decades of research on ranking methods, a limited number of studies have leveraged predictive analytics and machine learning to rank universities. In this article, we collected 49 Canadian universities’ data for 2017–2021 and divided them based on Maclean's categories into Primarily Undergraduate, Comprehensive, and Medical/Doctoral Universities. After identifying the input and output components, we leveraged various feature engineering and machine learning techniques to predict the universities’ ranks. We used Pearson Correlation, Feature Importance, and Chi-Square as the feature engineering methods, and the results show that “student to faculty ratio,” “total number of citations”, and “total number of Grants” are the most important factors in ranking Canadian universities. Also, the Random Forest machine learning model for the “primarily undergraduate category,” the Voting classifier model for the “comprehensive category” and the Gradient Boosting model for the “medical/doctoral category” performed the best. The selected machine learning models were evaluated based on accuracy, precision, F1 score, and recall.

Applying Multi-Sensor Satellite Data to Identify Key Natural Factors in Annual Livestock Change and Winter Livestock Disaster (Dzud) in Mongolian Nomadic Pasturelands

In the present study, we tested the applicability of multi-sensor satellite data to account for key natural factors of annual livestock number changes in county-level soum districts of Mongolia. A schematic model of nomadic landscapes was developed and used to select potential drivers retrievable from multi-sensor satellite data. Three alternative methods (principal component analysis, PCA; stepwise multiple regression, SMR; and random forest machine learning model, RF) were used to determine the key drivers for livestock changes and Dzud outbreaks. The countrywide Dzud in 2010 was well-characterized by the PCA as cold with a snowy winter and low summer foraging biomass. The RF estimated the annual livestock change with high accuracy (R2 &gt; 0.9 in most soums). The SMR was less accurate but provided better intuitive insights on the regionality of the key factors and its relationships with local climate and Dzud characteristics. Summer and winter variables appeared to be almost equally important in both models. The primary factors of livestock change and Dzud showed regional patterns: dryness in the south, temperature in the north, and foraging resource in the central and western regions. This study demonstrates a synergistic potential of models and satellite data to understand climate–vegetation–livestock interactions in Mongolian nomadic pastures.

Design and Selection of High Entropy Alloys for Hardmetal Matrix Applications Using a Coupled Machine Learning and Calculation of Phase Diagrams Methodology

This study aims to utilize a combined machine learning (ML) and CALculation of PHAse Diagrams (CALPHAD) methodology to design hardmetal matrix phases for metal‐forming applications that can serve as the basis for carbide reinforcement. The vast compositional space that high entropy alloys (HEAs) occupy offers a promising avenue to satisfy the application design criteria of wear resistance and ductility. To efficiently explore this space, random forest ML models are constructed and trained from publicly available experimental HEA databases to make phase constitution and hardness predictions. Interrogation of the ML models constructed reveals accuracies &gt;78.7% and a mean absolute error of 66.1 HV for phase and hardness predictions respectively. Six promising alloy compositions, extracted from the ML predictions and CALPHAD calculations, are experimentally fabricated and tested. The hardness predictions are found to be systematically under‐ and overpredicted depending on the alloy microstructure. In parallel, the phase classification models are found to lack sensitivity toward additional intermetallic phase formation. Despite the discrepancies identified between ML and experimental results, the fabricated compositions show promise for further experimental evaluation. These discrepancies are believed to be directly associated with the available databases but, importantly, have highlighted several avenues for both ML and database development.

Machine Learning-based Classification of Variable Stars

The classification of variable stars, essential for revealing information on stellar properties and cosmic distances, traditionally relied on statistical methods and limited data. With the emergence of transformative machine learning methodologies and large stellar surveys, we are able to perform more efficient, accurate, and robust handling of expansive databases. We deploy two different machine learning models—random forest and XGBoost—that effectively classify variable stars identified in the most recent phase of the Optical Gravitational Lensing Experiment using four features: (1) time of minimum brightness, (2) I-band amplitude, (3) mean or maximum I-band magnitude, and (4) period. The two models achieve a cross-correlation score of 99.00%, indicating largely shared classifications. The results produced by the models uncover new insights into the precision of supervised learning models in predicting variable stars with the aid of the most up-to-date survey data.

Multisite development and validation of machine learning models to predict severe outcomes and guide decision-making for emergency department patients with influenza.

Millions of Americans are infected by influenza annually. A minority seek care in the emergency department (ED) and, of those, only a limited number experience severe disease or death. ED clinicians must distinguish those at risk for deterioration from those who can be safely discharged. We developed random forest machine learning (ML) models to estimate needs for critical care within 24h and inpatient care within 72h in ED patients with influenza. Predictor data were limited to those recorded prior to ED disposition decision: demographics, ED complaint, medical problems, vital signs, supplemental oxygen use, and laboratory results. Our study population was comprised of adults diagnosed with influenza at one of five EDs in our university health system between January 1, 2017 and May 18, 2022; visits were divided into two cohorts to facilitate model development and validation. Prediction performance was assessed by the area under the receiver operating characteristic curve (AUC) and the Brier score. Among8032 patientswith laboratory-confirmed influenza,incidence of critical care needs was 6.3% and incidence of inpatient care needswas 19.6%.Themost common reasons for ED visitwere symptoms of respiratory tract infection,fever,and shortness of breath. Model AUCswere0.89 (95% CI 0.86-0.93)for prediction of criticalcare and0.90(95% CI0.88-0.93)for inpatient care needs;Brier scores were0.026 and0.042, respectively.Importantpredictorsincludedshortness of breath,increasingrespiratory rate,and ahigh number of comorbid diseases. ML methods can be used to accurately predict clinical deterioration in ED patients with influenza and have potential to support ED disposition decision-making.