Search NASA⌕ Search

SEARCH · Search NASA

Results for “random forest (RF)”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Applied Machine-Learning Models to Identify Spectral Sub-Types of M Dwarfs from Photometric Surveys

M dwarfs are the most abundant stars in the Solar Neighborhood and they are prime targets for searching for rocky planets in habitable zones. Consequently, a detailed characterization of these stars is in demand. The spectral sub-type is one of the parameters that is used for the characterization and it is traditionally derived from the observed spectra. However, obtaining the spectra of M dwarfs is expensive in terms of observation time and resources due to their intrinsic faintness. We study the performance of four machine-learning (ML) models—K-Nearest Neighbor (KNN), Random Forest (RF), Probabilistic Random Forest (PRF), and Multilayer Perceptron (MLP)—in identifying the spectral sub-types of M dwarfs at a grand scale by deploying broadband photometry in the optical and near-infrared. We trained the ML models by using the spectroscopically identified M dwarfs from the Sloan Digital Sky Survey (SDSS) Data Release (DR) 7, together with their photometric colors that were derived from the SDSS, Two-Micron All-Sky Survey, and Wide-field Infrared Survey Explorer. We found that the RF, PRF, and MLP give a comparable prediction accuracy, 74%, while the KNN provides slightly lower accuracy, 71%. We also found that these models can predict the spectral sub-type of M dwarfs with ~99% accuracy within ±1 sub-type. The five most useful features for the prediction are r - z, r - i, r - J, r - H , and g - z, and hence lacking data in all SDSS bands substantially reduces the prediction accuracy. However, we can achieve an accuracy of over 70% when the r and i magnitudes are available. Since the stars in this study are nearby (d ≲ 1300 pc for 95% of the stars), the dust extinction can reduce the prediction accuracy by only 3%. Finally, we used our optimized RF models to predict the spectral sub-types of M dwarfs from the Catalog of Cool Dwarf Targets for the Transiting Exoplanet Survey Satellite, and we provide the optimized RF models for public use.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRS2023

Bhutan Agriculture III: Monitoring Cropland Changes in Bhutan using Remote Sensing to Bolster Food Security and Support Crop Monitoring

The Bhutan Agriculture III team aimed to improve agricultural efficiency in Bhutan. Bhutan is a nation heavily reliant on agriculture, but it faces challenges such as geophysical limitations and lack of scientific agricultural practice. The team partnered with a primary end user, Bhutan’s Department of Agriculture (DoA), and with collaborators; the Bhutan Foundation, National Plant Protection Centre (NPPC), Agricultural Research Department Centre (ARDC), National Statistics Bureau (NSB), and the Ugyen Wangchuck Institute for Conservation and Environment Research (UWICER). Advised by NASA SERVIR, the team developed crop masks and monitored rice distribution from 2015 to 2022 utilizing Earth observations such as Landsat 8 Operational Land Imager (OLI), Landsat 9 OLI-2, Sentinel-1 C-Band Synthetic Aperture Radar (C-SAR), Sentinel-2 MultiSpectral Instrument (MSI) and Shuttle Radar Topography Mission (SRTM). The team gathered 5,000 points from the five dzongkhags that yield the most rice in Bhutan (Paro, Punakha, Samtse, Sarpang and Wangue Phodrang) using Collect Earth Online (CEO). With the data collected, the team split the data into training and validation data on Google Earth Engine (GEE) for a random forest (RF) classifier for rice and non-rice classification. After running the data on the Random Forest (RF) model, the team got an accuracy score of 81.48%, a kappa score of 55.75% and an F1 score of 86.11%. This data supports better agricultural decision-making for the governing body of Bhutan, helps enhance farming efficiency and foster sustainable practices, assists in overcoming data inaccuracy and bolsters food security in the country.

Sonam Seldon Tshering↗
DOE OSTI2023

A chemistry-informed hybrid machine learning approach to predict metal adsorption onto mineral surfaces

Historically, surface complexation model (SCM) constants and distribution coefficients (K d ) have been employed to quantify mineral-based retardation effects controlling the fate of metals in subsurface geologic systems. Our recent SCM development workflow, based on the Lawrence Livermore National Laboratory Surface Complexation/Ion Exchange (L-SCIE) database, illustrated a community FAIR data approach to SCM development by predicting uranium(VI)-quartz adsorption for a large number of literature-mined data. Here, we present an alternative hybrid machine learning (ML) approach that shows promise in achieving equivalent high-quality predictions compared to traditional surface complexation models. At its core, the hybrid random forest (RF) ML approach is motivated by the proliferation of incongruent SCMs in the literature that limit their applicability in reactive transport models. Our hybrid ML approach implements PHREEQC-based aqueous speciation calculations; values from these simulations are automatically used as input features for a random forest (RF) algorithm to quantify adsorption and avoid SCM modeling constraints entirely. Named the LLNL Speciation Updated Random Forest (L-SURF) model, this hybrid approach is shown to have applicability to U(VI) sorption cases driven by both ion-exchange and surface complexation, as is shown for quartz and montmorillonite cases. The approach can be applied to reactive transport modeling and may provide an alternative to the costly development of self-consistent SCM reaction databases.

38 RADIATION CHEMISTRY, RADIOCHEMISTRY, AND NUCLEA↗
DOE OSTI2022

Ch3MS-RF: a random forest model for chemical characterization and improved quantification of unidentified atmospheric organics detected by chromatography–mass spectrometry techniques

Abstract. The chemical composition of ambient organic aerosols plays a critical role in driving their climate and health-relevant properties and holds important clues to the sources and formation mechanisms of secondary aerosol material. In most ambient atmospheric environments, this composition remains incompletely characterized, with the number of identifiable species consistently outnumbered by those that have no mass spectral matches in the literature or the National Institute of Standards and Technology/National Institutes of Health/Environmental Protection Agency (NIST/NIH/EPA) mass spectral databases, making them nearly impossible to definitively identify. This creates significant challenges in utilizing the full analytical capabilities of techniques which separate and generate spectra for complex environmental samples. In this work, we develop the use of machine learning techniques to quantify and characterize novel, or unidentifiable, organic material. This work introduces Ch3MS-RF (Chemical Characterization by Chromatography–Mass Spectrometry Random Forest Modeling), an open-source, R-based software tool, for efficient machine-learning-enabled characterization of compounds separated in chromatography–mass spectrometry applications but not identifiable by comparison to mass spectral databases. A random forest model is trained and tested on a known 130 component representative external standard to predict the response factors of novel environmental organics based on position in volatility–polarity space and mass spectrum, enabling the reproducible, efficient, and optimized quantification of novel environmental species. Quantification accuracy on a reserved 20 % test set randomly split from the external standard compound list indicates that random forest modeling significantly outperforms the commonly used methods in both precision and accuracy, with a median response factor percent error of −2 %, for modeled response factors, compared to > 15 %, for typically used proxy assignment-based methods. Chemical properties modeling, evaluated on the same reserved 20 % test set and an extrapolation set of species identified in ambient organic aerosol samples collected in the Amazon rainforest, also demonstrate robust performance. Extrapolation set property prediction mean absolute errors for carbon number, oxygen to carbon ratio (O : C), average carbon oxidation state (OSc‾), and vapor pressure are 1.8, 0.15, 0.25, and 1.0 (log(atm)), respectively. Extrapolation set out-of-sample R2 for all properties modeled are above 0.75, with the exception of vapor pressure. While predictive performance for vapor pressure is less robust compared to the other chemical properties modeled, random-forest-based modeling was significantly more accurate than other commonly used methods of vapor pressure prediction, decreasing the mean vapor pressure prediction error to 0.24 (log(atm)) from 0.55 (log(atm)) (chromatography-based vapor pressure prediction) and 1.2 (log(atm)) (chemical formula-based vapor pressure prediction). The random forest model significantly advances an untargeted analysis of the full scope of chemical speciation yielded by two-dimensional gas chromatography (GCxGC-MS) techniques and can be applied to gas chromatography coupled with electron ionization mass spectrometry (GC-MS) as well. It enables the accurate estimation of key chemical properties commonly utilized in the atmospheric chemistry community, which may be used to more efficiently identify important tracers for further individual analysis and to characterize compound populations uniquely formed under specific ambient conditions.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Line Faults Classification Using Machine Learning on Three Phase Voltages Extracted from Large Dataset of PMU Measurements

An end-to-end supervised learning method is developed to classify transmission line faults in a twoyear field-recorded dataset that includes synchronized measurements of three-phase voltages recorded by 38 Phasor Measurement Units (PMU) sparsely located in in the US Western Grid interconnection. Statistical analysis is performed to extract features from this large dataset to train Support Vector Machine (SVM), Random Forest (RF), and eXtreme Gradient Boosting (XGBoost) classifiers initially. The training further leverages a simulated dataset from a synthetic grid with 12 PMUs to increase the number of faults of types infrequently seen in the field-recorded dataset. Training the classification models with the combined dataset resulted in a classification accuracy of 97.7%. This is a significant improvement over 89.7% to 92.5% accuracy obtained by relying on the field-recorded dataset alone.

47 OTHER INSTRUMENTATION↗
DOE OSTI2022

Machine Learning for Improving Surface-Layer-Flux Estimates

Abstract Flows in the atmospheric boundary layer are turbulent, characterized by a large Reynolds number, the existence of a roughness sublayer and the absence of a well-defined viscous layer. Exchanges with the surface are therefore dominated by turbulent fluxes. In numerical models for atmospheric flows, turbulent fluxes must be specified at the surface; however, surface fluxes are not known a priori and therefore must be parametrized. Atmospheric flow models, including global circulation, limited area models, and large-eddy simulation, employ Monin–Obukhov similarity theory (MOST) to parametrize surface fluxes. The MOST approach is a semi-empirical formulation that accounts for atmospheric stability effects through universal stability functions. The stability functions are determined based on limited observations using simple regression as a function of the non-dimensional stability parameter representing a ratio of distance from the surface and the Obukhov length scale (Obukhov in Trudy Inst Theor Geofiz AN SSSR 1:95–115, 1946), $$z/L$$ z / L . However, simple regression cannot capture the relationship between governing parameters and surface-layer structure under the wide range of conditions to which MOST is commonly applied. We therefore develop, train, and test two machine-learning models, an artificial neural network (ANN) and random forest (RF), to estimate surface fluxes of momentum, sensible heat, and moisture based on surface and near-surface observations. To train and test these machine-learning algorithms, we use several years of observations from the Cabauw mast in the Netherlands and from the National Oceanic and Atmospheric Administration’s Field Research Division tower in Idaho. The RF and ANN models outperform MOST. Even when we train the RF and ANN on one set of data and apply them to the second set, they provide more accurate estimates of all of the fluxes compared to MOST. Estimates of sensible heat and moisture fluxes are significantly improved, and model interpretability techniques highlight the logical physical relationships we expect in surface-layer processes.

Meteorology & Atmospheric Sciences↗
DOE OSTI2020

A2Cloud‐RF : A random forest based statistical framework to guide resource selection for high‐performance scientific computing on the cloud

Summary This article proposes a random‐forest based A2Cloud framework to match scientific applications with Cloud providers and their instances for high performance. The framework leverages four engines for this task: PERF engine, Cloud trace engine, A2Cloud‐ext engine, and the random forest classifier (RFC) engine. The PERF engine profiles the application to obtain performance characteristics, including the number of single‐precision (SP) floating‐point operations (FLOPs), double‐precision (DP) FLOPs, x87 operations, memory accesses, and disk accesses. The Cloud trace engine obtains the corresponding performance characteristics of the selected Cloud instances including: SP floating point operations per second (FLOPS), DP FLOPS, x87 operations per second, memory bandwidth, and disk bandwidth. The A2Cloud‐ext engine uses the application and Cloud instance characteristics to generate objective scores that represent the application‐to‐Cloud match. The RFC engine uses these objective scores to generate two types of random forests to assist users with rapid analysis: application‐specific random forests (ARF) and application‐class based random forests. The ARF consider only the input application's characteristics to generate a random forest and provide numerical ratings to the selected Cloud instances. To generate the application‐class based random forests, the RFC engine downloads the application profiles and scores of previously tested applications that perform similar to the input application. Using these data, the RFC engine creates a random forest for instance recommendation. We exhaustively test this framework using eight real‐world applications across 12 instances from different Cloud providers. Our tests show significant statistical agreement between the instance ratings given by the framework and the ratings obtained via actual Cloud executions.

Samuel, David↗
NASA NTRS2021

Bhutan Agriculture: Developing a Crop Mask for Rice and Creating a Data Collection Protocol Utilizing Remotely Sensed Data in Bhutan

Rice cultivation in Bhutan has been increasingly threatened by deteriorating soil health and outbreaks of diseases and pests associated with the global change in climate patterns. Field surveys, which the national government of Bhutan has relied on to monitor remote agricultural lands, are becoming increasingly overwhelmed by growing threats to agricultural health. To address these concerns, NASA DEVELOP partnered with the Department of Agriculture of Bhutan, the Bhutan Foundation, and the Ugyen Wangchuck Institute of Conservation and Environmental Research (UWICER) and worked to increase the government of Bhutan’s agricultural monitoring capacity. Utilizing Earth observations including Landsat 8 Operational Land Imager (OLI), Sentinel-1 C-band Synthetic Aperture Radar (C-SAR), Shuttle Radar Topography Mission (SRTM), and Planet imagery, the DEVELOP team worked with NASA SERVIR and created a sampling protocol to identify rice plantations and supplement field surveys for more efficient agriculture monitoring. The analysis focused on districts Paro, Punakha, Samtse, Sarpang, Trongsa, Zhemgang, Wangdue Phodrang, and Samdrup Jongkhar in the year 2020 during the period of transplantation (June) to harvesting of rice (November). The team provided the partners with a sampling protocol for integrating NASA Earth observations into their crop monitoring methods, as well as a crop mask for rice identification and to aid crop management. The crop mask for rice was developed using the Random Forest (RF) classifier for the eight districts of Bhutan. Visually, the random forest model has proved to be more accurate and precise than the classification and Regression Tree model. Statistically, the Random Forest model was 91.8% accurate in identifying rice in Bhutan.

Yeshey Seldon↗
DOE OSTI2025

Transcripts and genomic intervals associated with variation in metabolite abundance in maize leaves under field conditions

Abstract Plants exhibit extensive environment-dependent intraspecific metabolic variation, which likely plays a role in determining variation in whole plant phenotypes. However, much of the work seeking to use natural variation to link genes and transcript’s impacts on plant metabolism has employed data from controlled environments. Here, we generated and analyzed data on the variation in the abundance of 26 metabolites across 660 maize inbred lines under field conditions. We employ these data and previously published transcript and whole plant phenotype data reported for the same field experiment to identify both genomic intervals (through genome-wide association studies (GWAS)) and transcripts (using both transcriptome-wide association studies (TWAS) and an explainable artificial intelligence (AI) approach based on random forest (RF)) associated with variation in metabolite abundance. Both genome-wide association and random forest-based methods identified substantial numbers of significant associations including genes with plausible links to the metabolites they are associated with. In contrast, the transcriptome-wide association identified only six significant associations. In three cases, genetic markers associated with metabolic variation in our study colocalized with markers linked to variation in non-metabolic traits scored in the same experiment. We speculate that the poor performance of transcriptome-wide association studies in identifying transcript-metabolite associations may reflect a high prevalence of non-linear interactions between transcripts and metabolites and/or a bias towards rare transcripts playing a large role in determining intraspecific metabolic variation.

Mathivanan, Ramesh Kanna↗
DOE OSTI2018

Network Scale Ubiquitous Volume Estimation Using Tree-Based Ensemble Learning Methods

Currently real-time, ubiquitous volume data for roadway networks remains the key missing dimension in traffic operations. Most volume data are average annual daily traffic (AADT) measures derived from the Highway Performance Monitoring System (HPMS). Although methods to factor the AADT to hourly averages for typical day of week exist, actual volume data is limited to a sparse collection of locations in which volumes are continuously recorded. This paper explores the use of state-of-art machine learning techniques to estimate accurate real-time volume measures that span the highway network providing ubiquitous coverage in space, and point-in-time measures for a specific date and time. Three tree-based ensemble learning models, random forest (RF), gradient boost machine (GBM), and extreme gradient boost (XGBoost), were tested for volume estimation by learning from combined dataset of commercial probe data provided by TomTom, the FHWA's Travel Monitoring Analysis System (TMAS) data, and other infrastructure attributes such as number of lanes, speed limit, and weather. The methods were tested on major corridors and freeways in the metropolitan area of Denver. All three machine learning methods were able to provide hourly volume estimates 24 hours a day, 7 days a week, and 365 days a year with around 18% mean absolute error to true volume and about 5% of error with respect to roadway capacity. The low error measures allow the potential application by transportation agencies.

33 ADVANCED PROPULSION SYSTEMS↗
DOE OSTI2020

Mapping Vegetation at Species Level with High-Resolution Multispectral and Lidar Data Over a Large Spatial Area: A Case Study with Kudzu

Mapping vegetation species is critical to facilitate related quantitative assessment, and mapping invasive plants is important to enhance monitoring and management activities. Integrating high-resolution multispectral remote-sensing (RS) images and lidar (light detection and ranging) point clouds can provide robust features for vegetation mapping. However, using multiple sources of high-resolution RS data for vegetation mapping on a large spatial scale can be both computationally and sampling intensive. Here, we designed a two-step classification workflow to potentially decrease computational cost and sampling effort and to increase classification accuracy by integrating multispectral and lidar data in order to derive spectral, textural, and structural features for mapping target vegetation species. We used this workflow to classify kudzu, an aggressive invasive vine, in the entire Knox County (1362 km2) of Tennessee (U.S.). Object-based image analysis was conducted in the workflow. The first-step classification used 320 kudzu samples and extensive, coarsely labeled samples (based on national land cover) to generate an overprediction map of kudzu using random forest (RF). For the second step, 350 samples were randomly extracted from the overpredicted kudzu and labeled manually for the final prediction using RF and support vector machine (SVM). Computationally intensive features were only used for the second-step classification. SVM had constantly better accuracy than RF, and the producer’s accuracy, user’s accuracy, and Kappa for the SVM model on kudzu were 0.94, 0.96, and 0.90, respectively. SVM predicted 1010 kudzu patches covering 1.29 km2 in Knox County. We found the sample size of kudzu used for algorithm training impacted the accuracy and number of kudzu predicted. The proposed workflow could also improve sampling efficiency and specificity. Our workflow had much higher accuracy than the traditional method conducted in this research, and could be easily implemented to map kudzu in other regions as well as map other vegetation species.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

A bi-level data-driven framework for fault-detection and diagnosis of HVAC systems

Long-term operation of heating, ventilation, and air conditioning (HVAC) systems will eventually lead to a range of HVAC system failures, resulting in excessive energy consumption and maintenance costs. Here, to avoid HVAC malfunctioning, fault detection diagnostic (FDD) is utilized as a common practice. Machine learning methods have lately received considerable interest for FDD analysis of HVAC systems due to their high detection accuracy. Meanwhile, HVAC malfunctions are regarded as rare occurrences, hence normal operating data samples are much more accessible than data samples in faulty and malfunctioning conditions. The dominating frequency of normal operation in HVAC datasets has also led to heavily biased classification algorithms within the literature. Moreover, the focus of previous literature has been on increasing the accuracy of the models which leads to a high number of false positives (misleading alarms) in the system. In order to enhance the performance of diagnostic procedures and fill the mentioned gaps, this study proposes a novel data-driven framework. A bi-level machine learning framework is developed for diagnosing faults in air handling units (AHUs) and rooftop units (RTUs) based on principal component analysis (PCA), time series anomaly detection, and random forest (RF). It is shown that PCA can reduce the dataset dimension with one principal component accounting for 95% of data variance. Also, the random forest could classify the faults with 89% precision for single-zone AHU, 85% precision for RTU, and 79% for multi-zone AHU. By proposing this framework, three persistent challenges are addressed: (I) minimizing false positives; (II) accounting for data imbalance; and (III) normal condition monitoring of equipment.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2021

A Machine Learning-Based Vulnerability Analysis for Cascading Failures of Integrated Power-Gas Systems

This article proposes a cascading failure simulation (CFS) method and a hybrid machine learning method for vulnerability analysis of integrated power-gas systems (IPGSs). The CFS method is designed to study the propagating process of cascading failures between the two systems, generating data for machine learning with initial states randomly sampled. The proposed method considers generator and gas well ramping, transmission line and gas pipeline tripping, island issue handling and load shedding strategies. Then, a hybrid machine learning model with a combined random forest (RF) classification and regression algorithms is proposed to investigate the impact of random initial states on the vulnerability metrics of IPGSs. Extensive case studies are carried out on three test IPGSs to verify the proposed models and algorithms. Simulation results show that the proposed models and algorithms can achieve high accuracy for the vulnerability analysis of IPGSs.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2024

Fair Bagging Boosting Models [SWR-24-38]

Fair Bagging Boosting Models is a software implementation of a framework for building, measuring bias and correcting bias in 3 popular forest machine learning models: gradient boosted trees (GBT), random forest (RF), and XGBoost models, using the XGBoost library. The framework takes advantage of the flexibility in XGBoost library to represent gradient boosted tree and random forest models, as well as the ability to use custom loss function.

Ugirumurera, Juliette↗
DOE OSTI2021

Weather and Random Forest-based Load Profiling Approximation Models and its Transferability across Climate Zones

This study is to provide predictive understanding of the associations of various weather attributes with residential and commercial load profiles, for a variety of climate zones and seasons. In this work, machine learning (ML) approaches were used to identify and quantify the impacts of various weather attributes on residential and commercial electricity demand and its components across the western United States. Performance and transferability of the developed ML models were then evaluated across different temperate zones (e.g., southern, middle, and northern US) and across coastal, mid-continent, and wet zones, with inputs of weather condition data from the National Oceanic and Atmospheric Administration (NOAA) at representative weather stations. The predictive models were developed based on the ranked/screened factors using the regression tree (RT) and random forest (RF) approaches, for five different scenarios (seasons).

load composite, Random Forest, regression tree, lo↗
DOE OSTI2026

The Use of Machine Learning Models for Predicting the Dielectric Strength of Gases

Technological advancements in high voltage systems have pushed sulfur hexafluoride (SF6) to its operational limits. Furthermore, this gas has other drawbacks including a high liquefaction temperature and a high global warming potential. Therefore, there has been an urgent need to find alternative gases with high dielectric strength (DS). In this work, density functional theory (DFT) is used to calculate molecular descriptors that are fed into an artificial neural network (ANN) and a random forest (RF). These machine learning (ML) models are then used to predict the DS for hundreds of molecules. A finite element model (FEM) is also used to calculate the electric field profile of multiple simple electrode geometries as the applied voltage to the system is increased. Results indicate that the random forest model has better generalization to unseen data than the neural network. The highest DS value predicted by the RF was 2.16 relative to the experimental DS of SF6. The results also demonstrate how choosing a gas with a higher DS and a geometry with minimal edges and corners can significantly increase the operating voltage of an electrical system. Due to its superior generalization, the RF represents the most promising path toward an accurate DS predictor once sufficient experimental data are available.

Mileski, Matthew [AFIT]↗
DOE OSTI2024

Geographical Insights into Suicide Mortality Through Spatial Machine Learning

Suicide mortality is a leading cause of death in the United States, with an upward trend that emphasizes its significance as a public health issue. Previous research has employed global models like ordinary least squares (OLS) regression and local models such as geographically weighted regression (GWR). While local models are useful for analyzing spatial variations in suicide mortality, they share limitations with traditional global models, particularly about their inability to handle multi-collinearity and non-linear relationships. Machine learning approaches, like random forests (RF), can address some of these limitations but often fail to account for spatial variability. This gap highlights the need for spatial ML models specifically designed to tackle suicide mortality. This research seeks to fill this void by using a geographically weighted random forest model (GWRF) to examine the associations between county-level suicide mortality in the U.S. from 2010 to 2020 and various social and environmental determinants of health. A key aspect of our methodology is disciplined feature selection, which reduces the pool of explanatory variables by about 90%. This refinement enhances the explanatory power of both global (R2 improved from 0.59 to 0.67) and local (R2 improved from 0.64 to 0.67) RF models while reducing their run times. An analysis of the importance scores for these selected features reveals that the drivers of suicide mortality vary by context. Thus, to effectively address regional disparities and inform targeted public health interventions, a holistic approach that incorporates multiple county-level characteristics is essential.

Lebakula, Viswadeep [ORNL] (ORCID:0000000152935914↗
DOE OSTI2020

Structure–activity relationship-based chemical classification of highly imbalanced Tox21 datasets

Abstract The specificity of toxicant-target biomolecule interactions lends to the very imbalanced nature of many toxicity datasets, causing poor performance in Structure–Activity Relationship (SAR)-based chemical classification. Undersampling and oversampling are representative techniques for handling such an imbalance challenge. However, removing inactive chemical compound instances from the majority class using an undersampling technique can result in information loss, whereas increasing active toxicant instances in the minority class by interpolation tends to introduce artificial minority instances that often cross into the majority class space, giving rise to class overlapping and a higher false prediction rate. In this study, in order to improve the prediction accuracy of imbalanced learning, we employed SMOTEENN, a combination of Synthetic Minority Over-sampling Technique (SMOTE) and Edited Nearest Neighbor (ENN) algorithms, to oversample the minority class by creating synthetic samples, followed by cleaning the mislabeled instances. We chose the highly imbalanced Tox21 dataset, which consisted of 12 in vitro bioassays for > 10,000 chemicals that were distributed unevenly between binary classes. With Random Forest (RF) as the base classifier and bagging as the ensemble strategy, we applied four hybrid learning methods, i.e., RF without imbalance handling (RF), RF with Random Undersampling (RUS), RF with SMOTE (SMO), and RF with SMOTEENN (SMN). The performance of the four learning methods was compared using nine evaluation metrics, among which F 1 score, Matthews correlation coefficient and Brier score provided a more consistent assessment of the overall performance across the 12 datasets. The Friedman’s aligned ranks test and the subsequent Bergmann-Hommel post hoc test showed that SMN significantly outperformed the other three methods. We also found that a strong negative correlation existed between the prediction accuracy and the imbalance ratio (IR), which is defined as the number of inactive compounds divided by the number of active compounds. SMN became less effective when IR exceeded a certain threshold (e.g., > 28). The ability to separate the few active compounds from the vast amounts of inactive ones is of great importance in computational toxicology. This work demonstrates that the performance of SAR-based, imbalanced chemical toxicity classification can be significantly improved through the use of data rebalancing.

Idakwo, Gabriel↗