Search

Flood mitigation behavior is essential for effective flood risk management, particularly in Australia, where the federal government has increasingly emphasized individual responsibility for preventing and preparing for flood disasters. Despite this emphasis, around 60% of flood-prone residents hesitate to adopt private mitigation measures. Their reluctance highlights the complexities of decision making regarding the implementation mitigation measures, emphasizing the need for a comprehensive understanding of flood mitigation behavior among Australian residents. In order to address this knowledge gap, we conducted household surveys in flood-prone regions of Australia and used six machine learning algorithms—logistic regression, k -nearest neighbor, support vector machine, random forest, extreme gradient boosting, and artificial neural network—to analyze the proposed framework for flood mitigation behavior. The results of five-fold cross-validation demonstrated that the random forest algorithm provided superior accuracy for predicting protection motivation compared to the other five algorithms. Furthermore, seven of the 37 predictors utilized in the model had a higher impact on protection motivation than the other predictors. These influential predictors included self-efficacy, response efficacy, response costs, key variables within the coping appraisal factor, fear or worry about future flood risk, one variable associated with the threat appraisal factor, past experiences of flooding inside the house linked to the prior flood experience factor, and the implementation of flood mitigation measures by individuals connected to the householder, such as family, friends, and neighbors, which pertains to the social environment factor within the proposed framework for flood mitigation behavior. These predictors were deemed adequate to be used as input variables in the random forest model, because they provided accuracy and performance similar to when all 37 predictors were included. These insights into influential predictors can be used to develop an optimal prediction model for protection motivation in flood-prone areas of Australia and reduce the effort needed to collect the required data during postdisaster surveys. This research is of value to policymakers and floodplain managers in designing effective flood mitigation strategies and promoting the adoption of protective measures in flood-prone communities to reduce flood risk. [ABSTRACT FROM AUTHOR]

Assessment of the spatial distribution of potential pathways of sediment transport and the degree of linkage between sediment sources and the channel network within a watershed represents a valuable analysis for informing management decisions on sediment yield and transfer. Given the limitations of conventional methods for determining index of sediment connectivity (IC), there is a need to provide a flexible and efficient approach with the ability to apply different factors. In this regard, five decision tree-based machine learning models: M5 prime (M5P), random tree (RT), random forest (RF), alternating model tree (AMT), and reduced error pruning tree (REPT) were tested using geomorphic and climatic factors. Two databases were constructed with 200 and 1600 classes at 50 watersheds in Queensland, Australia. In these models, IC was assessed as an output parameter and six attributes that affect IC were assigned as input parameters (i.e., elevation, slope, area, length of stream channel, normalized difference vegetation index, and rainfall). Statistical validation and comparison of model predictions with calculated IC values based on the approach of Borselli et al. (Catena 75:268–277, 2008) were performed. Based on the statistical criteria, the RF model produced the most robust estimations of IC compared to other models and performed very well for IC modelling, especially in smaller subsections of watersheds. Accordingly, these findings can play an effective role for implementing watershed management and soil and water resources management measures. [ABSTRACT FROM AUTHOR]

Spatially explicit prediction of soil organic carbon (SOC) serves as a crucial foundation for effective land management strategies aimed at mitigating soil degradation and assessing carbon sequestration potential. Here, using more than 1000 in-situ observations, we trained two machine learning models (random forest, and K-means coupled with multiple linear regression), and one process-based model (the vertically resolved MIcrobial-MIneral Carbon Stabilization (MIMICS)) to predict SOC content of the top 30 cm of soil in Australia. Parameters of MIMICS were optimized for different site groupings, using two distinct approaches, plant functional types (MIMICS-PFT), and the most influential environmental factors (MIMICS-ENV). We found that at the continental scale, soil bulk density and mean annual temperature are the dominant controls of SOC variation, and that dominant controls vary for different vegetation types. All models showed good performance in SOC predictions with R2 greater than 0.8 during out-of-sample validation with random forest being the most accurate, and SOC in forests is more predictable than that in non-forest soils. Parameter optimization approaches made a notable difference in the performance of MIMICS SOC prediction with MIMICS-ENV performing better than MIMICS-PFT especially in non-forest soils. Digital maps of terrestrial SOC stocks generated using all the models showed similar spatial distribution with higher values in southeast and southwest Australia, but the magnitude of estimated SOC stocks varied. The mean ensemble estimate of SOC stocks was 30.08 t/ha with K-means coupled with multiple linear regression generating the highest estimate (mean SOC stocks at 38.15 t/ha) and MIMICS-PFT generating the lowest estimate (mean SOC stocks at 24.29 t/ha). We suggest that enhancing process-based models to incorporate newly identified drivers that significantly influence SOC variations in different environments could be key to reducing the discrepancies in these estimates. Our findings underscore the considerable uncertainty in SOC estimates derived from different modelling approaches and emphasize the importance of rigorous out-of-sample validation before applying any one approach in Australia. [ABSTRACT FROM AUTHOR]

Background: Operating rooms (ORs) are one of the costliest units in a hospital, therefore the cumulative consequences of any kind of inefficiency in OR management lead to a significant loss of revenue for the hospital, staff dissatisfaction, and patient care disruption. One of the possible solutions to improving OR efficiency is knowing a reliable estimate of the duration of operations. The literature suggests that the current methods used in hospitals, e.g., a surgeon's estimate for the given surgery or taking the average of only five previous records of the same procedure, have room for improvement. Methods: We used over 4 years of elective surgery records (n = 52,171) from one of the major metropolitan hospitals in Australia. We developed robust Machine Learning (ML) approaches to provide a more accurate prediction of operation duration, especially in the absence of surgeon's estimation. Individual patient characteristics and historic surgery information attributed to medical records were used to train predictive models. A wide range of algorithms such as Extreme Gradient Boosting (XGBoost) and Random Forest (RF) were tested for predicting operation duration. Results: The results show that the XGBoost model provided statistically significantly less error than other compared ML models. The XGBoost model also reduced the total absolute error by 6854 min (i.e., about 114 h) compared to the current hospital methods. Conclusion: The results indicate the potential of using ML methods for reaching a more accurate estimation of operation duration compared to current methods used in the hospital. In addition, using a set of realistic features in the ML models that are available at the point of OR scheduling enabled the potential deployment of the proposed approach. [ABSTRACT FROM AUTHOR]

Mineral prospectivity mapping is a crucial technique for discovering new economic mineral deposits. However, detailed knowledge-based geological exploration and interpretations generally involve significant costs, time, and human resources. In this study, an ensemble machine learning approach was tested using geoscience datasets to map Cu-Au and Pb-Zn mineral prospectivity in the Cobar Basin, NSW, Australia. The input datasets (magnetic, gravity, faults, electromagnetic, and magnetotelluric data layers) were chosen by considering their association with Cu-Au and Pb-Zn mineralization patterns. Three machine learning algorithms, namely random forest (RF), support vector machine (SVM), and maximum-likelihood (MaxL) classification, were applied to the input data. The results of the three algorithms were ensembled to produce Cu-Au and Pb-Zn prospectivity maps over the Cobar Basin with improved classification accuracy. The findings demonstrate good agreement with known mineral occurrence points and existing mineral prospectivity maps developed using the weights-of-evidence (WofE) method. The ability to capture training points accurately and the simplicity of the proposed approach make it advantageous over complex mineral prospectivity mapping methods, to serve as a preliminary evaluation technique. The methodology can be modified with different datasets and algorithms, facilitating the investigations of mineral prospectivity in other regions and providing guidance for more detailed, high-resolution geological investigations. [ABSTRACT FROM AUTHOR]

Wildlife strikes in aviation represent a serious economic concern; however, in some jurisdictions, the costs associated with this phenomenon are not collected or shared. This hampers the industry's ability to quantify the risk and assess the potential benefit from investment in effective wildlife hazard management activities. This research project has applied machine learning to the problem by training a random forest algorithm on wildlife strike cost data collected in the United States and predicting the costs associated with wildlife strikes in Australia. This method estimated a mean annual figure of AUD 7.9 million in repair costs and AUD 4.8 million in other costs from 2008 to 2017. It also provided year-on-year estimates showing variability through the reporting period that was not correlated with strike report numbers. This research provides a baseline figure for the Australian aviation industry to assess and review current and future wildlife hazard management practices. It also provides a technique for other countries, airlines, or airports to estimate the cost of wildlife strikes within their jurisdictions or operational environments. [ABSTRACT FROM AUTHOR]

Accurate pre-harvest yield forecasting of mango is essential to the industry as it supports better decision making around harvesting logistics and forward selling, thus optimizing productivity and reducing food waste. Current methods for yield forecasting such as manually counting 2–3% of the orchard can be accurate but are very time inefficient and labour intensive. More recent evaluations of technological solutions such as remote (satellite) and proximal (on ground) sensing have provided very encouraging results, but they still require infield in-season sampling for calibration, the technology comes at a significant cost, and commercial availability is limited, especially for vehicle-mounted sensors. This study presents the first evaluation of a "time series"—based remote sensing method for yield forecasting of mango, a method that does not require infield fruit counts and utilizes freely available satellite imagery. Historic yield data from 2015 to 2022 were sourced from 51 individual orchard blocks from two farms (AH and MK) in the Northern Territory of Australia. Time series measures of the canopy reflectance properties of the blocks were obtained from Landsat 7 and 8 satellite data for the 2015–2022 growing seasons. From the imagery, the following vegetation indices (VIs) were derived: EVI, GNDVI, NDVI, and LSWI, whilst corresponding weather variables (rainfall (Prec), temperature (Tmin/Tmax), evapotranspiration (ETo), solar radiation (Rad), and vapor pressure deficit (vpd)) were also sourced from SILO data. To determine the relationships among weather and remotely sensed measures of canopy throughout the growing season and the yield achieved (at the block level and the farm level), six machine learning (ML) algorithms, namely random forest (RF), support vector regression (SVR), eXtreme gradient boosting (XGBOOST), RIDGE, LASSO and partial least square regression (PLSR), were trialed. The EVI/GNDVI and Prec/Tmin were found to be the best RS and weather predictors, respectively. The block-level combined RS/weather-based RF model for 2021 produced the best result (MAE = 2.9 t/ha), marginally better than the RS only RF model (MAE = 3.4 t/ha). The farm-level model error (FLEM) was generally lower than the block-level model error, for both the combined RS/weather-based RF model (farm = 3.7%, block (NMAE) = 33.6% for 2021) and the RS-based model (farm = 4.3%, block = 38.4% for 2021). Further testing of the RS/weather-based RF models over six additional orchards (other than AH and MK) produced errors ranging between 24% and 39% from 2016 to 2020. Although accuracies of prediction did vary at both the block level and the farm level, this preliminary study demonstrates the potential of a "time series" RS method for predicting mango yields. The benefits to the mango industry are that it utilizes freely available imagery, requires no infield calibration, and provides predictions several months before the commercial harvest. Therefore, this outcome not only presents a more adoptable option for the industry, but also better supports automation and scalability in terms of block-, farm-, regional, and national level forecasting. [ABSTRACT FROM AUTHOR]

One of the most valuable and nutritionally essential agricultural commodities worldwide is milk. The European Union and New Zealand are the second- and third-largest exporting regions of milk products and rely heavily on pasture-based production systems. They are comparable to the Australian systems investigated in this study. With projections of herd decline, increased milk yield must be obtained from a combination of animal genetics and feed efficiencies. Accurate pasture biomass estimation across all seasons will improve feed efficiency and increase the productivity of dairy farms; however, the existing time-consuming and manual methods of pasture measurement limit improvements to utilisation. In this study, Sentinel-2 (S2) band and spectral index (SI) information were coupled with the broad season and management-derived datasets using a Random Forest (RF) machine learning (ML) framework to develop a perennial ryegrass (PRG) biomass prediction model accurate to +/−500 kg DM/ha, and that could predict pasture yield above 3000 kg DM/ha. Measurements of PRG biomass were taken from 11 working dairy farms across southeastern Australia over 2019–2021. Of the 68 possible variables investigated, multiple simulations identified 12 S2 bands and 9 SI, management and season as the most important variables, where Short-Wave Infrared (SWIR) bands were the most influential in predicting pasture biomass above 4000 kg DM/ha. Conditional Latin Hypercube Sampling (cLHS) was used to split the dataset into 80% and 20% for model calibration and internal validation in addition to an entirely independent validation dataset. The combined internal model validation showed R2 = 0.90, LCCC = 0.72, RMSE = 439.49 kg DM/ha, NRMSE = 15.08, and the combined independent validation had R2 = 0.88, LCCC = 0.68, RMSE = 457.05 kg DM/ha, NRMSE = 19.83. The key findings of this study indicated that the data obtained from the S2 bands and SI were appropriate for making accurate estimations of PRG biomass. Furthermore, including SWIR bands significantly improved the model. Finally, by utilising an RF ML model, a single 'global' model can automate PRG biomass prediction with high accuracy across extensive regions of all seasons and types of farm management. [ABSTRACT FROM AUTHOR]

Background: Thirty-day readmission is an increasingly important problem for total knee arthroplasty (TKA) patients. The aim of this study was to develop a risk prediction model using machine learning and clinical insight for 30-day readmission in primary TKA patients. Method: Data used to train and internally validate a multivariable predictive model were obtained from a single tertiary referral centre for TKA located in Victoria, Australia. Hospital administrative data and clinical registry data were utilised, and predictors were selected through systematic review and subsequent consultation with clinicians caring for TKA patients. Logistic regression and random forest models were compared to one another. Calibration was evaluated by visual inspection of calibration curves and calculation of the integrated calibration index (ICI). Discriminative performance was evaluated using the area under the receiver operating characteristic curve (AUC-ROC). Results: The models developed in this study demonstrated adequate calibration for use in the clinical setting, despite having poor discriminative performance. The best-calibrated readmission prediction model was a logistic regression model trained on administrative data using risk factors identified from systematic review and meta-analysis, which are available at the initial consultation (ICI = 0.012, AUC-ROC = 0.589). Models developed to predict complications associated with readmission also had reasonable calibration (ICI = 0.012, AUC-ROC = 0.658). Conclusion: Discriminative performance of the prediction models was poor, although machine learning provided a slight improvement. The models were reasonably well calibrated, meaning they provide accurate patient-specific probabilities of these outcomes. This information can be used in shared clinical decision-making for discharge planning and post-discharge follow up. [ABSTRACT FROM AUTHOR]

Annual turnover of home care workers represents a huge loss of revenue and is a key source of inefficiency in the home health care industry. In this article, we propose a data-driven approach to monitor employee churn and to capture the evolution of employee intent to leave. Unlike most papers in the literature, we use machine learning techniques to analyze over 2 million visits in the US, Canada, and Australia between 2016 and 2019. Results show that the gap between the number of hours worked and in the contract is the most important factor to predict employee intent to leave, which means an employee should be given as many hours as requested in the contract to improve retention. Secondary results show that having diverse shift lengths and continuity in services and patients seem to be associated with less turnover. [ABSTRACT FROM AUTHOR]

The characterisation of fuel distribution across heterogeneous landscapes is important for wildfire mitigation, validating fuel models, and evaluating fuel treatment outcomes. However, efficient fuel mapping at a landscape scale is challenging. Fuel hazard metrics were obtained using Terrestrial Laser Scanning (TLS) and the current operational approach (visual fuel assessment) for seven sites across south-eastern Australia. These point-based metrics were then up-scaled to a continuous fuel map, an area relevant to fire management using random forest modelling, with predictor variables derived from Airborne Laser Scanning (ALS), Sentinel 2A images, and climate and soil data. The model trained and validated with TLS observations (R 2 = 0.51 for near-surface fuel cover and 0.31 for elevated fuel cover) was found to have higher predictive power than the model trained with visual fuel assessments (R 2 = −0.1 for the cover of both fuel layers). Models for height derived from TLS observations exhibited low-to-moderate performance for the near-surface (R 2 = 0.23) and canopy layers (R 2 = 0.25). The results from this study provide practical guidance for the selection of training data sources and can be utilised by fire managers to accurately generate fuel maps across an area relevant to operational fire management decisions. [ABSTRACT FROM AUTHOR]

Population-level uptake of direct-acting antiviral (DAA) treatment for hepatitis C virus (HCV) infection, including retreatment, can be estimated through administrative pharmaceutical dispensation data. However, the reasons for retreatment are not captured in these data. We developed a machine learning model to classify retreatments as reinfection or treatment failure at a national level. Retreatment data from the REACH-C cohort (n = 10,843 treated with DAAs; n = 320 retreatments with known reason), were used to train a random forest model. Nested cross validation was undertaken to assess model performance and to optimise hyperparameters. The model was applied to data on DAA retreatment dispensed during 2016-2021 in Australia, to identify the reason for retreatment (treatment failure or reinfection). Average predictive accuracy, precision, sensitivity, specificity and F 1 -score for the model were 96.3%, 96.5%, 96.3%, 96.3% and 96.3%, respectively. Nationally, 95,272 individuals initiated DAAs, with treatment uptake declining from 32,454 in 2016 to 6,566 in 2021. Of those treated, 6,980 (7%) were retreated. Our model classified 51.8% (95% CI 46.7–53.6%; n = 3,614) of cases as reinfection and 48.2% (95% CI 46.4–53.3%; n = 3,366) as treatment failure. Retreatment for reinfection increased steadily over the study period from 14 in 2016 to 1,092 in 2020, stabilising in 2021. Retreatment for treatment failure increased from 73 in 2016 to 1,077 in 2019, then declined to 515 in 2021. Among individuals retreated for treatment failure, 50% had discontinued initial treatment. We used a novel methodology with high classification accuracy to evaluate DAA retreatment patterns at a national level. Increases in retreatment uptake for treatment failure corresponded to the availability of pangenotypic and salvage regimens. Increasing retreatment uptake for reinfection likely reflects increasing reinfection incidence. This study used machine learning methodologies to analyse national administrative data and characterise trends in HCV retreatment due to reinfection and treatment failure. Retreatment for reinfection increased over time, reflecting increasing numbers of people at risk for reinfection following HCV cure. Increased retreatment for treatment failure corresponded to the availability of pangenotypic and salvage DAA regimens. The findings of this study can be used by public health agencies and policy makers to guide and assess HCV elimination strategies, while the novel methodology for monitoring trends in HCV retreatment has the potential to be used in other settings, and health conditions. [Display omitted] • Seven percent of people treated for HCV in Australia were retreated. • Machine learning was used to classify retreatments based on national administrative data. • Overall, 52% of retreatment was for reinfection and 48% was for treatment failure. • Retreatment for reinfection increased over time, reflecting increasing numbers of people at-risk for reinfection following cure. • Retreatment for treatment failure increased when salvage DAAs became available. [ABSTRACT FROM AUTHOR]

The early detection and rapid extinguishing of forest fires are effective in reducing their spread. Based on the MODIS Thermal Anomaly (MOD14) algorithm, we propose an early stage fire detection method from low-spatial-resolution but high-temporal-resolution images, observed by the Advanced Himawari Imager (AHI) onboard the geostationary meteorological satellite Himawari-8. In order to not miss early stage forest fire pixels with low temperature, we omit the potential fire pixel detection from the MOD14 algorithm and parameterize four contextual conditions included in the MOD14 algorithm as features. The proposed method detects fire pixels from forest areas using a random forest classifier taking these contextual parameters, nine AHI band values, solar zenith angle, and five meteorological values as inputs. To evaluate the proposed method, we trained the random forest classifier using an early stage forest fire data set generated by a time-reversal approach with MOD14 products and time-series AHI images in Australia. The results demonstrate that the proposed method with all parameters can detect fire pixels with about 90% precision and recall, and that the contribution of contextual parameters is particularly significant in the random forest classifier. The proposed method is applicable to other geostationary and polar-orbiting satellite sensors, and it is expected to be used as an effective method for forest fire detection. [ABSTRACT FROM AUTHOR]

Summary: Landscape rehabilitation following mining is required to be resilient to disturbance impacts such as fire, drought and disease. As mining companies undergo the process of rehabilitation certification and mine closure, there are notable knowledge gaps on the ecological risks associated with mature rehabilitated landscapes, based largely on the assumption that rehabilitation is analogous to reference communities. However, the response to fire disturbance across a range of landscapes remains largely untested and in particular there is limited understanding of recovery traits of plant species that occur naturally or are commonly seeded into rehabilitation. In August 2018, a controlled fire was applied to 37 hectares of 12‐year‐old coal‐mine rehabilitation in central Queensland, Australia. We used a combination of (i) ground plot surveys and (ii) drone imagery to compare the vegetation response of burnt woody species to unburnt controls prior to, and for, two years following the fire. The survival of the most dominant shrub species found on the rehabilitation site was significantly impacted by the fire. Old Man Saltbush (Atriplex nummularia Lindl. subsp. nummularia) recorded significant post‐fire mortality, with ground surveys recording an average reduction of 89% of stems per hectare across the burnt site, while unburnt controls remained unchanged. The plot data analysis was supported with high spatial and temporal resolution drone imagery, classified using a Random Forest machine‐learning approach. Change analysis of these maps showed a significant decline of 82% in Old Man Saltbush plant density and 92% reduction in foliage cover following the fire. In addition, the mean canopy area of individual Old Man Saltbush shrubs reduced significantly from a pre‐fire mean of 11.3 to 4.8 m2 two years following the fire. A spatial proximity analysis showed that those individuals that survived the fire were located significantly closer to unburnt areas and bare spoil, indicating that discontinuous ground fuel loads can greatly improve the survivability of individuals. This study provides new evidence on the contested fire sensitivity of Old Man Salt bush and demonstrates the risk that future climate‐driven extreme events may have on the resilience of novel ecosystems. [ABSTRACT FROM AUTHOR]

The power of Machine Learning is demonstrated for automatic interpretation of well logs and determining reservoir properties for volume of shale, porosity, and water saturation respectively for tight clastic sequences. Random Forest algorithms are reputed for their efficiency as they belong to a class of algorithms called ensemble methods, which are traditionally seen as weak learners, but can be transformed into strong performers and they promise to deliver highly accurate results. The study area is located offshore Australia in the Poseidon and Crown fields situated in the Browse Basin, which are gas fields in tight complex clastic reservoirs. There are 5 wells used in this study with one well manually interpreted which is subsequently used in developing a machine learning model which predicts the output for the other 4 wells. The basic open hole logs namely Natural gamma ray, Resistivity, Neutron Porosity, Bulk Density, P-wave and S-wave sonic travel-time, are used in interpretation. One of the wells has a missing S-wave travel-time log which was also predicted by developing a Random Forest Machine Learning model. The results indicate a very robust improvement in performance when Random Forest algorithm was combined with Adaptive Boosting when interpreting the well logs. The training accuracy using Random Forest alone was 98.21%, but testing was 77.62% which suggested over-fitting by the Random Forest model. The Adaptive Boosting of the Random Forest algorithm resulted in the overall training accuracy of 99.40% and an overall testing accuracy of 97.03%, indicating a drastic improvement in performance. S-wave travel-time log was predicted by preparing a training set consisting of Natural gamma ray, Resistivity, Neutron Porosity, Bulk Density, and P-wave travel-time logs for the 4 wells using Random Forest which gave a training accuracy of 99.79% and a testing accuracy of 98.54%. Machine learning algorithms can be successfully applied for interpreting well log data in complex sedimentary environment and their performance can be drastically improved using Adaptive Boosting. [ABSTRACT FROM AUTHOR]

This study evaluates the ability of machine learning algorithms to classify tropical depressions (TDs) and tropical storms (TSs) in the western region of the southwest Pacific Ocean (SWPO). Decision rules are generated to predict the environment required for a depression to fully develop into a mature storm, and the most influential predictors in the classification decision are ranked. TD and TS are discriminated based on a maximum sustained wind speed threshold (≥17 ms−1). Various aerosol, thermodynamic, and dynamic parameters are extracted closest to the initiation point of each non‐developing and developing sample. The covariates associated with each labelled sample are used to train a decision tree and random forest model. Results using a testing dataset suggest the random forest approach more accurately distinguishes between non‐developing and developing samples. The classification accuracy of the decision tree and random forest are 72% and 91%, respectively. Random forest outperformed the decision tree by providing higher accuracy in test data. The most important variables for binary classification are sea salt aerosol optical depth (AOD), 1,000 mb relative humidity, and sea surface temperature. AOD is a quantitative estimate of the aerosols presents in the air through the extinction of a ray of light as it passes through the atmosphere. Mean composite maps constructed in an unsupervised manner have been created for the most important variables identified by the random forest classifier during TD and TS events to highlight the difference in geophysical and aerosol variables' climatology during the two different classifications. This work will advance the risk management strategies for northeastern Australia and other SWPO basin islands to control their tropical cyclone related losses through prioritizing forecasting variables that are the strongest predictors of the strengthening of tropical depressions into tropical cyclones. [ABSTRACT FROM AUTHOR]

There is considerable interest and value in identifying the gap between crop yields that have actually been achieved, and yields that could have potentially been achieved. A suite of methods currently exist to estimate the yield potential of a crop, but there are no approaches that predict the site- and season-specific yield potential using datasets that are readily available and easily accessible for farmers. The aim of this study was to fill this need and develop a novel approach to identify crop yield gaps through site- and season-specific models of crop yield potential. The study focused on cotton lint yield, with data from 14 different seasons and 68 different fields from a collection of large, irrigated cotton farms in eastern Australia. This abundance of yield data was then joined with other spatial and temporal datasets that describe yield, such as rainfall, temperature, soil, and management. A quantile random forest machine learning model was then used to model yield at 30 m resolution, where the 95th percentile predictions were treated as the yield potential. The yield gaps at a 30 m resolution were then estimated for all seasons and sites. The results were compared to a more traditional 'historical maximum yield' approach, where no data modelling and only empirical yield data was used to estimate the yield potential. This revealed that there was a general agreement between the two approaches, although the quantile machine learning approach is both site- and season-specific, not just site-specific. Overall, there is a great need for alternative approaches to estimate yield potential and yield gaps, as the approaches currently available possess many limitations. The approach developed in this study has the potential for wide-spread adoption in broadacre cropping systems, and if the causes of yield gaps are identified, could lead to the implementation of management strategies to close them. [ABSTRACT FROM AUTHOR]

Meteorological phenomena is an area in which a large amount of data is generated and where it is more difficult to make predictions about events that will occur due to the high number of variables on which they depend. In general, for this, probabilistic models are used that offer predictions with a margin of error, so that in many cases they are not very good. Due to the aforementioned conditions, the use of machine learning algorithms can serve to improve predictions. This article describes an exploratory study of the use of machine learning to make predictions about the phenomenon of rain. To do this, a set of data was taken as an example that describes the measurements gathered on rainfall in the main cities of Australia in the last 10 years, and some of the main machine learning algorithms were applied (knn, decision tree, random forest, and neural networks). The results show that the best model is based on neural networks. [ABSTRACT FROM AUTHOR]

Almonds (Prunus dulcis Mill.) are consumed worldwide and their geographical origin plays a crucial role in determining their market value. In the present study, a total of 250 almond reference samples from six countries (Australia, Spain, Iran, Italy, Morocco, and the USA) were non-polar extracted and analyzed by UPLC-ESI-IM-qToF-MS. Four harvest periods, more than 30 different varieties, including both sweet and bitter almonds, were considered in the method development. Principal component analysis showed that there are three groups of samples with similarities: Australia/USA, Spain/Italy and Iran/Morocco. For origin determination, a random forest achieved an accuracy of 88.8 %. Misclassifications occurred mainly between almonds from the USA and Australia, due to similar varieties and similar external influences such as climate conditions. Metabolites relevant for classification were selected using Surrogate Minimal Depth, with triacylglycerides containing oxidized, odd chained or short chained fatty acids and some phospholipids proven to be the most suitable marker substances. Our results show that focusing on the identified lipids (e. g., using a QqQ-MS instrument) is a promising approach to transfer the origin determination of almonds to routine analysis. [Display omitted] • Origin determination of almonds by lipidomics analysis via UPLC-ESI-IM-qToF-MS. • 250 almond samples from six different countries were used for analysis. • Random forest analysis achieved an accuracy of 88.8 %. • Metabolites were selected by Surrogate Minimal Depth and identified as triacylglycerols. [ABSTRACT FROM AUTHOR]

A basalt lava plain covers much of central and western Victoria in Australia. Many of the younger flows (<1 Ma) have hummocky surfaces with abundant tumuli and lava rises, known locally as stony rises due to the protruding basalt boulders. Stony rises are noteworthy geologically but have ecological importance and are culturally significant to Aboriginal Australians. Ecologically, stony rises host many native plant species. Culturally, stony rises provide evidence of Aboriginal land use and occupation through the presence of artefact deposits. Currently, mapping of stony rises is carried out by conducting field surveys on a study-by-study basis. This could be performed more rapidly using remote sensing datasets such as LiDAR, satellite and aerial imagery, and geophysical data. This research mapped stony rises in an area that is experiencing rapid urban development using a combination of remotely sensed datasets that characterise the morphology of stony rises through a Machine Learning approach that utilises Object Based Image Analysis (OBIA) segmentation and Random Forests (RF) classification techniques. The research successfully identified stony rises in the landscape to an accuracy of 89.97 %, with 314 potential new stony rise sites identified. The research identified the most important predictor variables were slope, local elevation, the concentrations of thorium and uranium (obtained from airborne radiometric data), aeromagnetic data (total magnetic intensity), the Normalised Difference Water Index (NDWI), and the Clay Mineral Ratio. The results from this study lay a foundation for standardising the definition of a stony rise within Australia. Furthermore, by utilising various predictor variables, this study highlights the success of a multi-faceted approach into detecting a specific landform at a landscape level. • Stony rises detected to an accuracy of 89.97 % using our novel method. • 314 potential new stony rises were detected using our novel method. • Using multifaceted predictor variables for landform classification and detection [ABSTRACT FROM AUTHOR]

Robust quantification of vegetative biomass using satellite imagery using one or more forms of machine learning (ML) has hitherto been hindered by the extent and quality of training data. Here, we showcase how ML predictive demonstrably improves when additional training data is used. We collated field datasets of pasture biomass obtained via destructive sampling, 'C-Dax' reflective measurements and rising plate meters (RPM) from ten livestock farms across four States in Australia. Remotely sensed data from the Sentinel-2 constellation was used to retrieve aboveground biomass using a novel machine learning paradigm hereafter termed "SPECTRA-FOR" (Spectral Pasture Estimation using Combined Techniques of Random-forest Algorithm for Features Optimisation and Retrieval). Using this framework, we show that the low temporal resolution of Sentinel-2 in high latitude regions with persistent cloud cover leads to extensive gaps between cloud-free images, hindering model performance and, thus, contemporaneous ability to forecast real-time pasture biomass. By leveraging the spectral consistency between Sentinel-2 and Planet Lab SuperDove to overcome this limitation, we used ten spectral bands of Sentinel-2, four bands of Sentinel-2 as a proxy for pre-2022 SuperDove (referred to as synthetic SuperDove or SSD), and the actual SuperDove (ASD), given that SuperDove imagery has a higher resolution and more frequent passage compared with Sentinel-2. Using their respective bands as input features to SPECRA-FOR, model performance for the ten bands of Sentinel-2 were R2 = 0.87, root mean squared error (RMSE) of 439 kg DM/ha and mean absolute error (MAE) of 255 kg DM/ha, while that for SSD increased to an R2 of 0.92, RMSE of 346 kg DM/ha and MAE = 208 kg DM/ha. The study revealed the importance of robust data mining, imagery harmonisation and model validation for accurate real-time modelling of pasture biomass with ML. [Display omitted] • New method to optimise Sentinel-2 imagery using Planet Labs SuperDove data. • The near-infrared band is more sensitive for retrieving aboveground biomass. • The random forest algorithm can capture the grazing regimes of individual farms. • Model performance and evaluation rely on quality data and ample field observations. • The evaluation of the model with monthly data outperformed the varied temporal scales. [ABSTRACT FROM AUTHOR]

This paper presents the results of a research project which investigated permeability prediction for the Precipice Sandstone of the Surat Basin. Machine learning techniques were used for permeability estimation based on multiple wireline logs. This information improves the prediction of CO2 injectivity in this formation. Well logs and core data were collected from 5 boreholes in the Surat Basin, where extensive core data and complete sets of conventional well logs exist for the Precipice Sandstone. Four different machine learning (ML) techniques, including Random Forest (RF), Artificial neural network (ANN), Gradient Boosting Regressor (GBR), and Support Vector Regressor (SVR), were independently trained with a wide range of hyper-parameters to ensure that not only is the best model selected, but also the right combination of model parameters is selected. Cross-validation for 20 different combinations of the seven available input logs was used for this study. Based on the performances in the validation and blind testing phases, the ANN with all seven logs used as input was found to give the best performance in predicting permeability for the Precipice Sandstone with the coefficient of determination (R2) of about 0.93 and 0.87 for the training and blind data sets respectively. Multi-regression analysis also appears to be a successful approach to calculate reservoir permeability for the Precipice Sandstone. Models with a complete set of well logs can generate reservoir permeability with R2 of more than 90%. [ABSTRACT FROM AUTHOR]

Road crash fatality is a universal problem of the transportation system. A massive death toll caused annually due to road crash incidents, and among them, vulnerable road users (VRU) are endangered with high crash severity. This paper focuses on employing machine learning-based classification approaches for modelling injury severity of vulnerable road users—pedestrian, bicyclist, and motorcyclist. Specifically, this study aims to analyse critical features associated with different VRU groups—for pedestrian, bicyclist, motorcyclist and all VRU groups together. The critical factor of crash severity outcomes for these VRU groups is estimated in identifying the similarities and differences across different important features associated with different VRU groups. The crash data for the study is sourced from the state of Queensland in Australia for the years 2013 through 2019. The supervised machine learning algorithms considered for the empirical analysis includes the K-Nearest Neighbour (KNN), Support Vector Machine (SVM) and Random Forest (RF). In these models, 17 distinct road crash parameters are considered as input features to train models, which originate from road user characteristics, weather and environment, vehicle and driver condition, period, road characteristics and regions, traffic, and speed jurisdiction. These classification models are separately trained and tested for individual and unified VRU to assess crash severity levels. Afterwards, model performances are compared with each other to justify the best classifier where Random Forest classification models for all VRU modes are found to be comparatively robust in test accuracy: (motorcyclist: 72.30%, bicyclist: 64.45%, pedestrian: 67.23%, unified VRU: 68.57%). Based on the Random Forest model, the road crash features are ranked and compared according to their impact on crash severity classification. Furthermore, a model-based partial dependency of each road crash parameters on the severity levels is plotted and compared for each individual and unified VRU. This clarifies the tendency of road crash parameters to vary with different VRU crash severity. Based on the outcome of the comparative analysis, motorcyclists are found to be more likely exposed to higher crash severity, followed by pedestrians and bicyclists. [ABSTRACT FROM AUTHOR]

Background: Implementation research has delved into barriers to implementing change and interventions for the implementation of innovation in practice. There remains a gap, however, that fails to connect implementation barriers to the most effective implementation strategies and provide a more tailored approach during implementation. This study aimed to explore barriers for the implementation of professional services in community pharmacies and to predict the effectiveness of facilitation strategies to overcome implementation barriers using machine learning techniques.Methods: Six change facilitators facilitated a 2-year change programme aimed at implementing professional services across community pharmacies in Australia. A mixed methods approach was used where barriers were identified by change facilitators during the implementation study. Change facilitators trialled and recorded tailored facilitation strategies delivered to overcome identified barriers. Barriers were coded according to implementation factors derived from the Consolidated Framework for Implementation Research and the Theoretical Domains Framework. Tailored facilitation strategies were coded into 16 facilitation categories. To predict the effectiveness of these strategies, data mining with random forest was used to provide the highest level of accuracy. A predictive resolution percentage was established for each implementation strategy in relation to the barriers that were resolved by that particular strategy.Results: During the 2-year programme, 1131 barriers and facilitation strategies were recorded by change facilitators. The most frequently identified barriers were a 'lack of ability to plan for change', 'lack of internal supporters for the change', 'lack of knowledge and experience', 'lack of monitoring and feedback', 'lack of individual alignment with the change', 'undefined change objectives', 'lack of objective feedback' and 'lack of time'. The random forest algorithm used was able to provide 96.9% prediction accuracy. The strategy category with the highest predicted resolution rate across the most number of implementation barriers was 'to empower stakeholders to develop objectives and solve problems'.Conclusions: Results from this study have provided a better understanding of implementation barriers in community pharmacy and how data-driven approaches can be used to predict the effectiveness of facilitation strategies to overcome implementation barriers. Tailored facilitation strategies such as these can increase the rate of real-time implementation of innovations in healthcare, leading to an industry that can confidently and efficiently adapt to continuous change. [ABSTRACT FROM AUTHOR]

Lightning-caused wildfires are a significant contributor to burned areas, with lightning ignitions remaining one of the most unpredictable aspects of the fire environment. There is a clear connection between fuel moisture and the probability of ignition; however, the mechanisms are poorly understood and predictive methods are underdeveloped. Establishing a lightning-ignition relationship would be useful in developing a model that would complement early warning systems designed for fire control and prevention. A machine learning (ML) approach was used to define a predictive model for wildfire ignition based on lightning forecasts and environmental conditions. Three different binary classifiers were adopted: a decision tree, an AdaBoost and a Random Forest, showing promising results, with both ensemble methods (Random Forest and AdaBoost) exhibiting an out-ofsample accuracy of 78%. Data provided by a Western Australia wildfire database allowed a comprehensive verification on over 145 lightning-ignited wildfires in regions of Australia during 2016. This highlighted that in a minimum of 71% of the cases the ML models correctly predicted the occurrence of an ignition when a fire was actually initiated. The super-learner developed is planned to be used in an operational context to the enhance information connected to fire management. [ABSTRACT FROM AUTHOR]

Objective: This research aimed to (i) assess the effects of time‐varying predictors (day of the week, month, year, holiday, temperature) on daily ED presentations and (ii) compare the accuracy of five methods for forecasting ED presentations, including four statistical methods and a machine learning approach. Methods: Predictors of ED presentations were assessed using generalised additive models (GAMs), generalised linear models, multiple linear regression models, seasonal autoregressive integrated moving average models and random forest. The accuracy of short‐term (14 days), mid‐term (30 days) and long‐term (365 days) forecasts were compared using two measures of forecasting error. Results: The data are the numbers of presentations to public hospital EDs in South‐East Queensland, Australia, from 2009 to 2015. ED presentations are largely affected by year of presentation, and to a lesser extent by month, day of the week and holidays. Maximum daily temperature is also a significant predictor of ED presentations. Of the four statistical models considered, the GAM had the greatest forecasting accuracy, and produced consistent and coherent forecasts, likely due to its flexibility in modelling complex time‐varying effects. The random forest machine learning approach had the lowest forecasting accuracy, likely due to overfitting the data. Conclusions: Calendar and temperature variables, not previously considered in the Australian literature, were found to significantly impact ED presentations. This study also demonstrates the potential of GAMs as a dual explanatory and forecasting method for the modelling, and more accurate prediction, of ED presentations. [ABSTRACT FROM AUTHOR]

• The accuracy of machine learning algorithms to predict injury locations were evaluated. • 16,878 construction accident records were analyzed. • Significant contributing factors were identified using permutation importance method. The construction industry is a hazardous industry with significant injuries and fatalities. Few studies have used data-driven analysis to investigate injuries due to construction accidents. This study aims to deploy machine learning (ML) models to predict four injury types (ITs): Upper limbs, lower limbs, head/neck, and back/trunk. A total of 16,878 construction accident records in Australia were collected and fed into several ML algorithms, including fine trees, ensemble of boosted trees, xgboost, random forest, two types of support vector machines, and logistic regression. Six performance metrics of precision, recall, accuracy, F1 score, the area under the receiver operating curve (AUROC), and the area under precision recall curve (AUPRC) were used to evaluate modeling outputs. Random forest showed superior performance in predicting injury types (accuracy 79.3%; recall 78.0%; F1 score 78.5%; precision 77.1%; AUROC 0.98; and AUPRC 0.78). The critical features of injury types were analyzed using the feature importance method and accident nature and mechanism had significant impacts. The study's findings contribute to safety enhancement by providing quantitative prediction models of injury types and subsequent development of safety controls in construction. [ABSTRACT FROM AUTHOR]

Ecological condition continues to decline in arid and semi-arid river basins globally due to hydrological over-abstraction combined with changing climatic conditions. Whilst provision of water for the environment has been a primary approach to alleviate ecological decline, how to accurately monitor changes in riverine trees at fine spatial and temporal scales, remains a substantial challenge. This is further complicated by constantly changing water availability across expansive river basins with varying climatic zones. Within, we combine rare, fine-scale, high frequency temporal in-situ field collected data with machine learning and remote sensing, to provide a robust model that enables broadscale monitoring of physiological tree water stress response to environmental changes via actual evapotranspiration (ET). Physiological variation of Eucalyptus camaldulensis (River Red Gum) and E. largiflorens (Black Box) trees across 10 study locations in the southern Murray-Darling Basin, Australia, was captured instantaneously using sap flow sensors, substantially reducing tree response lags encountered by monitoring visual canopy changes. Actual ET measurement of both species was used to bias correct a national spatial ET product where a Random Forest model was trained using continuous timeseries of in-situ data of up to four years. Precise monthly AMLETT (A ustralia-wide M achine L earning ET for T rees) ET outputs in 30 m pixel resolution from 2012 to 2021, were derived by incorporating additional remote sensing layers such as soil moisture, land surface temperature, radiation and EVI and NDVI in the Random Forest model. Landsat and Sentinal-2 correlation results between in-situ ET and AMLETT ET returned R2 of 0.94 (RMSE 6.63 mm period−1) and 0.92 (RMSE 6.89 mm period−1), respectively. In comparison, correlation between in-situ ET and a national ET product returned R2 of 0.44 (RMSE 34.08 mm period−1) highlighting the need for bias correction to generate accurate absolute ET values. The AMLETT method presented here, enhances environmental management in river basins worldwide. Such robust broadscale monitoring can inform water accounting and importantly, assist decisions on where to prioritize water for the environment to restore and protect key ecological assets and preserve floodplain and riparian ecological function. [Display omitted] • Data to quantify riverine tree evapotranspiration is underpinned by high replication of in-situ sap flow measurements. • Evapotranspiration from various field environments is reproduced using in-situ data and machine learning - R2 of 0.94 • Spatiotemporal data created to report long-term woody tree evapotranspiration (from 2012) in monthly intervals. • Random Forest model includes environmental parameters to estimate evapotranspiration across a variety of environments. [ABSTRACT FROM AUTHOR]

Conventional housing price prediction methods rarely consider the spatiotemporal non-stationary problem in a large data volumes. In this study, four machine learning (ML) models are used to explore the impacts of various features – i.e., property attributes and neighborhood quality - on housing price variations at different geographical scales. Using a 32-year (1984–2016) housing price dataset of Metropolitan Adelaide, Australia, this research relies on 428,000 sale transaction records and 38 explanatory variables. It is shown that non-linear tree-based models, such as Decision Tree, have perform better than linear models. In addition, ensemble machine learning techniques, such as Gradient-Boosting and Random Forest, are better at predicting future housing prices. A spatiotemporal lag (ST-lag) variable was added to improve the prediction accuracy of the models. The study demonstrates that ST-lag (or similar spatio-temporal indicator) can be a useful moderator of spatio-temporal effects in ML applications. This paper will serve as a catalyst for future research into the dynamics of the Australian property market, utilizing the benefits of cutting-edge technologies to develop models for business and property valuation at various geographical levels. • Spatio-temporal non-stationary is significant in explanation of the variations in housing price; • A proposed spatiotemporal lag is incorporated to increase the accuracy of the models; • Non-linear tree-based models have better performance than the linear model. • Ensemble ML techniques are powerful methods to produce better predictive performance. [ABSTRACT FROM AUTHOR]

This study describes the supervised support vector regression method and cuckoo optimization algorithm (COA-SVR), a newly developed mineral potential modelling (MPM) technique, and a case study of its application to predicting gold potential in the Granites-Tanami Orogen (GTO), Australia. COA-SVR, which was borne out of a computer program coded in MATLAB®, incorporates a popular radial basis function (RBF) known as the Gaussian kernel function. The COA-SVR model was trained using a series of predictor maps previously generated and described by Roshanravan et al. (2020a), and corroborated by way of a 10-fold cross-validation. The modelling results indicate that the COA-SVR approach to MPM outperformed the previous data-driven (i.e., random forest) and continuous (i.e., fuzzy gamma, geometric average, and data-driven index overlay) models of Roshanravan et al. (2020a). As such, the newly proposed COA-SVR method presents a valid approach to MPM. To further constrain the results, the gold prospective domains identified via COA-SVR and the previous models of Roshanravan et al. (2020a) were delimited, using the student's t-value and concentration-area (C-A) fractal techniques. Following the above, indices of overall performance (O e), accuracy (A), kappa (K) and normalized density (N d) were applied to further assess the gold prospective domains as delimited by the student's t-value and C-A fractal techniques. The results are consistent with the COA-SVR and random forest models having outperformed the more traditional modelling approaches documented by Roshanravan et al. (2020a). Given the positive results obtained in this study and similarly positive results reported by others, we believe that the apparent superior nature of machine-learning algorithms strongly merits further consideration, research and more widespread application in MPM. [Display omitted] • New approach to mineral potential modelling called a COA-SVR • Approach was tested on orogenic gold systems in the GTO, Australia. • Results were compared to previously generated mineral potential models. • Results indicate that the COA-SVR method presents a valid approach to MPM. • Study highlighted there is a direct correlation between the indices of Oe , A and K. [ABSTRACT FROM AUTHOR]

Recent studies have suggested that due to climate change, the number of wildfires across the globe have been increasing and continue to grow even more. The recent massive wildfires, which hit Australia during the 2019–2020 summer season, raised questions to what extent the risk of wildfires can be linked to various climate, environmental, topographical, and social factors and how to predict fire occurrences to take preventive measures. Hence, the main objective of this study was to develop an automatized and cloud-based workflow for generating a training dataset of fire events at a continental level using freely available remote sensing data with a reasonable computational expense for injecting into machine learning models. As a result, a data-driven model was set up in Google Earth Engine platform, which is publicly accessible and open for further adjustments. The training dataset was applied to different machine learning algorithms, i.e., Random Forest, Naïve Bayes, and Classification and Regression Tree. The findings show that Random Forest outperformed other algorithms and hence it was used further to explore the driving factors using variable importance analysis. The study indicates the probability of fire occurrences across Australia as well as identifies the potential driving factors of Australian wildfires for the 2019–2020 summer season. The methodical approach and achieved results and drawn conclusions can be of great importance to policymakers, environmentalists, and climate change researchers, among others. [ABSTRACT FROM AUTHOR]

The use of spectral data is seen as a fast and non-destructive method capable of monitoring pasture biomass. Although there is great potential in this technique, both end users and sensor manufacturers are uncertain about the necessary sensor specifications and achievable accuracies in an operational scenario. This study presents a straightforward parametric method able to accurately retrieve the hyperspectral signature of perennial ryegrass (Lolium perenne) canopies from multispectral data collected within a two-year period in Australia and the Netherlands. The retrieved hyperspectral data were employed to generate optimal indices and continuum-removed spectral features available in the scientific literature. For performance comparison, both these simulated features and a set of currently employed vegetation indices, derived from the original band values, were used as inputs in a random forest algorithm and accuracies of both methods were compared. Our results have shown that both sets of features present similar accuracies (root mean square error (RMSE) ≈490 and 620 kg DM/ha) when assessed in cross-validation and spatial cross-validation, respectively. These results suggest that for pasture biomass retrieval solely from top-of-canopy reflectance (ranging from 550 to 790 nm), better performing methods do not rely on the use of hyperspectral or, yet, in a larger number of bands than those already available in current sensors. [ABSTRACT FROM AUTHOR]

Fairness in policing, driven by the effective and transparent investigation and remediation of police misconduct, is vital to maintaining the legitimacy of policing agencies, and the capacity for police to function within society. Research into police misconduct in Australia has traditionally been performed on an ad-hoc basis, with limited access to law enforcement data. This research seeks to identify the antecedents of serious police misconduct, resulting in the dismissal or criminal charge of officers, among a large police misconduct dataset. Demographic and misconduct data were sourced for a sample of 600 officers who have committed instances of serious misconduct, and a matched sample of 600 comparison officers across a 13-year period. A machine learning analysis, random forest, was utilised to produce a robust predictive model, with Partial Dependence Plots employed to demonstrate within variable interaction with serious misconduct. Prior instances of serious misconduct were particularly predictive of further serious misconduct, while misconduct was most prominent around mid-career. Secondary employment, and performance issues were important predictors, while demographic variables typically outperformed complaint variables. This research suggests that serious misconduct is similarly prevalent among experienced officers, as it is junior officers, while secondary employment is an important indicator of misconduct risk. Findings provide guidance for misconduct prevention policy among policing agencies. [ABSTRACT FROM AUTHOR]

Identification of the extent of heat altered coals is important for coal mining and resource estimation because alteration directly affects key economic properties such as ash content, volatile matter, specific energy, sulphur, free swell index and beneficiation characteristics. The most reliable method to identify altered coal is through geochemical analysis of drill core samples; however this method is time consuming and costly. Cheaper alternative methods to identify altered coal is by macroscopic observation of non-core drill cuttings, but this is less reliable than core. Geophysical wireline log data is also used to identify altered coal, but with variable success. Over the past decades, Machine Learning methods have proven popular to automatically classify a variety of lithotypes from geophysical log data. However rarely has there been focus on predicting altered coal. In this paper we apply the most recent machine learning methods which include gradient boosted machines, random forests and artificial neural networks to automatically predict altered and non-altered lithotypes using geophysical log data. We use a massive data set comprising of 1230 samples from 263 boreholes from a highly intruded deposit in the Bowen Basin, Eastern Australia. To do this, we calculate each sample's distance from intrusion and predict their relative density from geophysical log inputs including gamma ray, caliper and compensated density. We then train our machine learning methods on 80% of the data to predict alteration class using the calculated distance to intrusion, predicted relative density, and geophysical log gamma ray, as primary inputs. We evaluated each of these machine learning methods on the remaining 20% of the data to determine the best performing model. Finally, using the best performing model we further split the altered and non-altered classification into lithotypes using a decision tree based on geological knowledge of the case study area. The results indicate that of the machine learning algorithms the random forest produced the best results with only 11 misclassifications across the entire data set of 1230 samples which represents <1% error. • Prediction of heat affected coals with a massive dataset and machine learning methods in the Bowen Basin. • Prediction of relative density and classification of altered and non-altered coals. • Lithology group definition based on a decision tree with <1% error. [ABSTRACT FROM AUTHOR]