Learning Random Forest Research Articles

Big data analytics on social networks for real-time depression detection

During the coronavirus pandemic, the number of depression cases has dramatically increased. Several depression sufferers disclose their actual feeling via social media. Thus, big data analytics on social networks for real-time depression detection is proposed. This research work detected the depression by analyzing both demographic characteristics and opinions of Twitter users during a two-month period after having answered the Patient Health Questionnaire-9 used as an outcome measure. Machine learning techniques were applied as the detection model construction. There are five machine learning techniques explored in this research which are Support Vector Machine, Decision Tree, Naïve Bayes, Random Forest, and Deep Learning. The experimental results revealed that the Random Forest technique achieved higher accuracy than other techniques to detect the depression. This research contributes to the literature by introducing a novel model based on analyzing demographic characteristics and text sentiment of Twitter users. The model can capture depressive moods of depression sufferers. Thus, this work is a step towards reducing depression-induced suicide rates.

A Random Forest approach to predict geology from geophysics in the Pontiac subprovince, Canada

Current “visual” integration approaches to map geology from geophysics are challenging, biased toward user assumptions, and time consuming. Applying a supervised machine learning Random Forest method to airborne geophysical data sets is an alternative solution to quickly generate interpretative maps, which can later be validated through targeted and therefore more efficient field campaigns. The approach is tested in the Malartic–Cadillac area of the Pontiac subprovince, Canada. Available airborne geophysical data sets include magnetic, frequency-domain electromagnetic, and radiometric data. Unlike the western part of the area, many studies have been done in the eastern part where the world-class Canadian Malartic gold deposit is located. The eastern part has abundant field observations and well-documented geology that are used as training data for Random Forest learning. A predicted map is built after preprocessing the data, gridding the area into a mesh of nodes and training using known geological knowledge. The results provide new information about the spatial distribution of lithological units, including diabase dikes, felsic-intermediate igneous rocks, mafic–ultramafic igneous rocks, and sedimentary rocks. Comparison between the predicted map and previous geological maps could suggest locations where future field mapping efforts should focus in the Malartic–Cadillac area.

Vulnerability Assessment of Asphalt Plant through Machine Learning Techniques

Many businesses throughout the globe have recently realized the value of Supervisory Control and Data Acquisition (SCADA) systems. Many critical infrastructures, such as electricity grids, asphalt plants, and wastewater disposals, are controlled by these systems. With the introduction of Fourth Industrial Revolution, 4IR or Industry 4.0, today’s SCADA systems cannot be separated from the outside world, making them more susceptible to hostile assaults. Conventional security systems including different antivirus software and firewalls are unable to safeguard SCADA systems as they are of distinct requirements. For this, different machine learning algorithms, i.e., SVM, KNN, and random forest, are tested to cover the anomaly detection along with security protection for SCADA systems. The dataset used in this research study was made locally in an asphalt plant by using different sensor data grouped in two classes: one is natural signal values, and the other one is attack class in which different sensor values are found out of range while in operation. Amongst the above-mentioned algorithms, KNN outperformed with an accuracy rate of 89% for anomaly detection and any kind of external attack can be detected and notified to the control room for on-time actions.

A temporal dependency feature in lower dimension for lung sound signal classification

Respiratory sounds are expressed as nonlinear and nonstationary signals, whose unpredictability makes it difficult to extract significant features for classification. Static cepstral coefficients such as Mel-frequency cepstral coefficients (MFCCs), have been used for classification of lung sound signals. However, they are modeled in high-dimensional hyperspectral space, and also lose temporal dependency information. Therefore, we propose shifted delta -cepstral coefficients in lower-subspace (SDC-L) as a novel feature for lung sound classification. It preserves temporal dependency information of multiple frames nearby same to original SDC, and improves feature extraction by reducing the hyperspectral dimension. We modified EMD algorithm by adding a stopping rule to objectively select a finite number of intrinsic mode functions (IMFs). The performances of SDC-L were evaluated with three machine learning techniques (support vector machine (SVM), k-nearest neighbor (k-NN) and random forest (RF)) and two deep learning algorithms (multilayer perceptron (MLP) and convolutional neural network (cNN)) and one hybrid deep learning algorithm combining cNN with long short term memory (LSTM) in terms of accuracy, precision, recall and F1-score. We found that the first 2 IMFs were enough to construct our feature. SVM, MLP and a hybrid deep learning algorithm (cNN plus LSTM) outperformed with SDC-L, and the other classifiers achieved equivalent results with all features. Our findings show that SDC-L is a promising feature for the classification of lung sound signals.

Impact assessment of humanitarian demining using object-based peri-urban land cover classification and morphological building detection from VHR Worldview imagery

This study utilised very high-resolution (VHR) imagery to monitor and evaluate the impact of humanitarian demining activities in a peri-urban environment in Afghanistan. Identifying buildings and mapping the spatial distribution of different types of land cover is of great practical significance for demining organisations, such as the HALO Trust, to assess the impact of their mine-clearance activities, by quantifying change and the growth of a population in a specific area over time. This study had two main objectives: (i) to map the post-clearance land cover, and (ii) to detect and quantify the change in the number and area of buildings. Two independent workflows were implemented and evaluated. To map land cover, this study investigated the implementation of various machine learning algorithms in object-based image classification (OBIA) of VHR satellite imagery (Worldview-1,2,3). Image segmentation was carried out using the Large-Scale Mean-Shift (LSMS) algorithm, before classification was performed based on a machine learning Random Forests (RF) approach. Different parameters and spatial distribution of training samples were tested to analyse the model's performance. Further analysis determined that by using only the segments' mean value per spectral band (Red, Green, Blue), data redundancy in the training stage was eliminated. The final classified map had an overall accuracy of 90.67% and a total built-area of 643,660.28 m2 was detected in the 4.11 km2 study area. To detect and quantify buildings present in the study area, an alternative, automatic, unsupervised approach based on the morphological building index (MBI) was implemented using MATLAB. Two VHR (0.5 m) panchromatic images acquired by WorldView-1 in 2008 and 2018 were processed using a series of multi-scale and multi-directional morphological operators, before a series of post-processing thresholds were applied to refine the output. Parameters were systematically optimised for the datasets and their sensitivity analysed. By comparing the output to manually labelled reference maps, producer's and user's accuracy of 78.63/81.70% and 75.30/77.91% were attained for 2008 and 2018 imagery respectively. The built-area was found to increase from 58,705 to 611,920 m2 over the ten-year period, indicating a significant increase in the number of people resident in the area. Whilst OBIA was found to be more accurate than MBI for the 2018 pansharpened imagery, with producer's and user's accuracy of 91.00/90.60% versus 84.65/79.44% respectively, it must be considered that the MBI, an unsupervised method requiring no training, offers a fast solution if the sole objective is building detection.

Development of a dynamic machine learning algorithm to predict clinical pregnancy and live birth rate with embryo morphokinetics

ObjectiveTo evaluate the feasibility of generating a center-specific embryo morphokinetic algorithm by time-lapse microscopy to predict clinical pregnancy rates.DesignA retrospective cohort analysis.SettingAcademic fertility clinic in a tertiary hospital setting.Patient(s)Patients who underwent in vitro fertilization with embryos that underwent EmbryoScope time-lapse microscopy and subsequent transfer between 2014 and 2018.Intervention(s)None.Main Outcome Measure(s)Clinical pregnancy.Result(s)A supervised, random forest learning algorithm from 367 embryos successfully predicted clinical pregnancy from a training set with overall 65% sensitivity and 74% positive predictive value, with an area under the curve of 0.7 for the test set. Similar results were achieved for live birth outcomes. For the secondary analysis, embryo growth morphokinetics were grouped into five clusters using unsupervised clustering. The clusters that had the fastest morphokinetics (time to blastocyst = 97 hours) had pregnancy rates of 54%, whereas a cluster that had the slowest morphokinetics (time to blastocyst = 122 hours) had a pregnancy rate of 71%, although the differences were not statistically significant (P=.356). Other clusters had pregnancy rates of 51%–60%.Conclusion(s)This study shows the feasibility of a clinic-specific, noninvasive embryo morphokinetic simple machine learning model to predict clinical pregnancy rates.

Research on optimization of an enterprise financial risk early warning method based on the DS-RF model

The financial risk early warning process of enterprises faces problems such as uncertainty and complexity. In the big data environment, scholars and enterprises that continue to use traditional evaluation methods will face large challenges. It is essential for an enterprise's sustainable operation to combine artificial intelligence algorithms, dynamically monitor its financial risks, and carry out financial risk early warning processes accurately and effectively. This study proposes an early warning method for corporate financial risks based on the evidence theory-random forest (DS-RF) model. The classic algorithm of machine learning—random forest was introduced into the framework of evidence theory to construct a random forest model with four dimensions: profitability, asset quality, debt risk, and operating growth. While predicting the risk, the credibility of the evidence was determined, and then the D-S synthesis rule was used for information fusion. An example was analyzed, taking JS Reclamation Group as the study subject. The comparison with the early warning results of the random forest algorithm and the traditional model shows that the DS-RF model proposed in this paper has a higher early warning accuracy and the results are presented more comprehensively and systematically, which effectively improves the efficiency of enterprise financial risk early warning and helps managers to make relevant decisions efficiently and scientifically.

Place recognition and navigation of outdoor mobile robots based on random Forest learning with a 3D LiDAR

Place recognition and navigation of outdoor mobile robots based on random Forest learning with a 3D LiDAR

Coupled online sequential extreme learning machine model with ant colony optimization algorithm for wheat yield prediction

Inadequate agricultural planning compounded by inaccurate predictions results in an inflated local market rate and prompts higher importation of wheat. To tackle this problem, this research has designed two-phase universal machine learning (ML) model to predict wheat yield (Wpred), utilizing 27 agricultural counties’ data within the Agro-ecological zone. The universal model, online sequential extreme learning machines coupled with ant colony optimization (ACO-OSELM) is developed, by incorporating the significant annual yield data lagged at (t − 1) as the model’s predictor to generate future yield at 6 test stations. In the first phase, ACO is adopted to search for suitable, statistically relevant data stations for model training, and the corresponding test station by virtue of a feature selection strategy. An annual wheat yield time-series input dataset is constructed utilizing data from each selected training station (1981–2013) and applied against 6 test stations (with each case modelled with 26 station data as the input) to evaluate the hybrid ACO-OSELM model. The partial autocorrelation function is implemented to deduce statistically significant lagged data, and OSELM is applied to generate Wpred. The two-phase hybrid ACO-OSELM model is tested within the 6 agricultural districts (represented as stations) of Punjab province, Pakistan and the results are benchmarked with extreme learning machine (ELM) and random forest (RF) integrated with ACO (i.e., hybrid ACO-ELM and hybrid ACO-RF models, respectively). The performance of the ACO-OSELM model was proven to be good in comparison to ACO-ELM and ACO-RF models. The hybrid ACO-OSELM model revealed its potential to be implemented as a decision-making system for crop yield prediction in areas where a significant association with the historical agricultural crop is well-established.

A machine learning approach for identification of gastrointestinal predictors for the risk of COVID-19 related hospitalization.

Background and aimCOVID-19 can be presented with various gastrointestinal symptoms. Shortly after the pandemic outbreak, several machine learning algorithms were implemented to assess new diagnostic and therapeutic methods for this disease. The aim of this study is to assess gastrointestinal and liver-related predictive factors for SARS-CoV-2 associated risk of hospitalization.MethodsData collection was based on a questionnaire from the COVID-19 outpatient test center and from the emergency department at the University Hospital in combination with the data from internal hospital information system and from a mobile application used for telemedicine follow-up of patients. For statistical analysis SARS-CoV-2 negative patients were considered as controls in three different SARS-CoV-2 positive patient groups (divided based on severity of the disease). The data were visualized and analyzed in R version 4.0.5. The Chi-squared or Fisher test was applied to test the null hypothesis of independence between the factors followed, where appropriate, by the multiple comparisons with the Benjamini Hochberg adjustment. The null hypothesis of the equality of the population medians of a continuous variable was tested by the Kruskal Wallis test, followed by the Dunn multiple comparisons test. In order to assess predictive power of the gastrointestinal parameters and other measured variables for predicting an outcome of the patient group the Random Forest machine learning algorithm was trained on the data. The predictive ability was quantified by the ROC curve, constructed from the Out-of-Bag data. Matthews correlation coefficient was used as a one-number summary of the quality of binary classification. The importance of the predictors was measured using the Variable Importance. A 2D representation of the data was obtained by means of Principal Component Analysis for mixed type of data. Findings with the p-value below 0.05 were considered statistically significant.ResultsA total of 710 patients were enrolled in the study. The presence of diarrhea and nausea was significantly higher in the emergency department group than in the COVID-19 outpatient test center. Among liver enzymes only aspartate transaminase (AST) has been significantly elevated in the hospitalized group compared to patients discharged home. Based on the Random Forest algorithm, AST has been identified as the most important predictor followed by age or diabetes mellitus. Diarrhea and bloating have also predictive importance, although much lower than AST.ConclusionSARS-CoV-2 positivity is connected with isolated AST elevation and the level is linked with the severity of the disease. Furthermore, using the machine learning Random Forest algorithm, we have identified the elevated AST as the most important predictor for COVID-19 related hospitalizations.

Effects of UAV-LiDAR and Photogrammetric Point Density on Tea Plucking Area Identification

High-cost data collection and processing are challenges for UAV LiDAR (light detection and ranging) mounted on unmanned aerial vehicles in crop monitoring. Reducing the point density can lower data collection costs and increase efficiency but may lead to a loss in mapping accuracy. It is necessary to determine the appropriate point cloud density for tea plucking area identification to maximize the cost–benefits. This study evaluated the performance of different LiDAR and photogrammetric point density data when mapping the tea plucking area in the Huashan Tea Garden, Wuhan City, China. The object-based metrics derived from UAV point clouds were used to classify tea plantations with the extreme learning machine (ELM) and random forest (RF) algorithms. The results indicated that the performance of different LiDAR point density data, from 0.25 (1%) to 25.44 pts/m2 (100%), changed obviously (overall classification accuracies: 90.65–94.39% for RF and 89.78–93.44% for ELM). For photogrammetric data, the point density was found to have little effect on the classification accuracy, with 10% of the initial point density (2.46 pts/m2), a similar accuracy level was obtained (difference of approximately 1%). LiDAR point cloud density had a significant influence on the DTM accuracy, with the RMSE for DTMs ranging from 0.060 to 2.253 m, while the photogrammetric point cloud density had a limited effect on the DTM accuracy, with the RMSE ranging from 0.256 to 0.477 m due to the high proportion of ground points in the photogrammetric point clouds. Moreover, important features for identifying the tea plucking area were summarized for the first time using a recursive feature elimination method and a novel hierarchical clustering-correlation method. The resultant architecture diagram can indicate the specific role of each feature/group in identifying the tea plucking area and could be used in other studies to prepare candidate features. This study demonstrates that low UAV point density data, such as 2.55 pts/m2 (10%), as used in this study, might be suitable for conducting finer-scale tea plucking area mapping without compromising the accuracy.

Can a Hierarchical Classification of Sentinel-2 Data Improve Land Cover Mapping?

Monitoring of land cover plays an important role in effective environmental management, assessment of natural resources, environmental protection, urban planning and sustainable development. Increasing demand for accurate and repeatable information on land cover and land cover changes causes rapid development of the advanced, machine learning algorithms dedicated to land cover mapping using satellite images. Free and open access to Sentinel-2 data, characterized with high spatial and temporal resolution, increased the potential to map and to monitor land surface with high accuracy and frequency. Despite a considerable number of approaches towards land cover classification based on satellite data, there is still a challenge to clearly separate complex land cover classes, for example grasslands, arable land and wetlands. The aim of this study is to examine, whether a hierarchal classification of Sentinel-2 data can improve the accuracy of land cover mapping and delineation of complex land cover classes. The study is conducted in the Lodz Province, in central Poland. The pixel-based land cover classification is carried out using the machine learning Random Forest (RF) algorithm, based on a time series of Sentinel-2 imagery acquired in 2020. The following nine land cover classes are mapped: sealed surfaces, woodland broadleaved, woodland coniferous, shrubs, permanent herbaceous (grassy cover), periodically herbaceous (i.e., arable land), mosses, non-vegetated (bare soil) and water bodies. The land cover classification is conducted following two approaches: (1) flat, where all land cover classes are classified together, and (2) hierarchical, where the stratification is applied to first separate the most stable land cover classes and then classifying the most problematic once. The national databases served as the source of the reference sampling plots for the classification process. The process of selection and verification of the reference sampling plots is performed automatically. To assess the stability of the classification models the classification processes are performed iteratively. The results of this study confirmed that the hierarchical approach gave more accurate results compared to the commonly used flat approach. The median of the overall accuracy (OA) of the hierarchical classification was higher by 3–9 percentage points compared to the flat one. Of interest, the OA of the hierarchical classification reached 0.93–0.99, whereas the flat approach reached 0.90. Individual classes are also better classified in the hierarchical approach.

Prognostic value of machine learning for acute heart failure

Abstract Funding Acknowledgements Type of funding sources: None. Background At onset of acute heart failure (AHF), various clinical fundamental parameters including vital sign, laboratory data, or initial treatment were investigated, and we can roughly estimate the prognosis. However, machine learning method for prediction of the prognosis was not studied. Purpose To elucidate prognostic value of machine learning for AHF comparing conventional statistical model. Methods We enrolled consecutive 300 patients with AHF (79.5 ± 12.1 years, 158 Males). Patients with acute coronary syndrome, mechanical circulatory support cases, and cardio-pulmonary arrest cases were excluded. The patients were randomly divided into 80% (240 cases) and 20% (60 cases), and the former was used as train data, and the latter as validation data. Objective variable was set as cardiac death in one year. First, logistic regression analysis with Akaike’s information criterion (AIC) was performed, and extracted predictive parameters. The predictive model for the cardiac prognosis was constructed by cut-off value of ROC curve analysis of propensity score was calculated. Next, machine learning (random forest method and deep learning) to build predictive model was performed with the predictors. Finally, accuracy of each predictive model was compared. Results Thirty cases showed cardiac death in one year. Logistic regression with AIC extracted 8 predictors, and the cut off-value of propensity score with the 6 parameters was 0.110. The accuracy was 0.714 and area under ROC (AUROC) was 0.836. Conversely, random forest method demonstrated the accuracy as 0.927, AUROC 0.860. On deep learning, the accuracy was 0.937 and AUROC 0.901. The top 4 high feature importance of random forest were Cl/red blood cell count/pH/Anion Gap. However, accuracy of those predictors was lower than that of machine learning. Conclusion Machine learning was a powerful tool to predict cardiac prognosis of AHF, comparing with conventional statistical model. Abstract Figure. Statistical model

Influence of Self-Efficacy on Cancer-Related Fatigue and Health-Related Quality of Life in Young Survivors of Childhood Cancer.

This study aims to elucidate how self-efficacy influences cancer-related fatigue and health-related quality of life (HRQoL) in young survivors of childhood cancer. Forty-six young survivors (age range, 8–18 years) of childhood cancer who were currently in complete remission completed measures for self-efficacy (Pediatric General Self-Efficacy Scale (PedsSE)), cancer-related fatigue (Cancer-related Fatigue Score (CRFS)), and HRQoL (Pediatric Quality of Life Inventory 4.0 Generic Core Scale, Pediatric Quality of Life Inventory (PedsQL)). Structural relationships between the PedsSE and CRFS or PedsQL, including the effects of potential demographic or clinical confounders, were examined by machine learning random forest algorithms and structural equation modeling. According to the distribution of the PedsQL, six survivors with PedsQL < 70 were determined to have compromised HRQoL (referred to as “low-PedsQL survivors”). The random forest model identified six variables for the prediction of the CRFS, with the PedsSE being the most important, and eight variables for the distinction of low-PedsQL survivors, with the CRFS being the most and the PedsSE the third most important variable. The structural equation model indicated that a direct influence of the PedsSE on the PedsQL was less detectable (β = −0.049), whereas an indirect influence of the PedsSE on the PedsQL via the CRFS was evident (β = 0.333). The model explained 51% of the variation of the CRFS and 28% of the variation of the PedsQL. The PedsSE was strongly correlated with “altered mood” in the subclass of the CRFS (r = −0.470), and “altered mood” was strongly correlated with the PedsQL (r = 0.737). In conclusion, self-efficacy is a major determinant of cancer-related fatigue and influences HRQoL via cancer-related fatigue in survivors of childhood cancer. The main pathway from self-efficacy to HRQoL is thought to be via the emotional aspect of cancer-related fatigue. However, unlike adult survivors of cancer, self-efficacy for young survivors may not contribute much to self-management behaviors that maintain HRQoL.

Providing Convenient Indoor Thermal Comfort in Real-Time Based on Energy-Efficiency IoT Network

Monitoring the thermal comfort of building occupants is crucial for ensuring sustainable and efficient energy consumption in residential buildings. It enables not only remote real-time detection of situations, but also a timely reaction to reduce the damage made by harmful situations in targeted buildings. In this paper, we first design a new Internet of Things (IoT) architecture in order to provide remote availability of both indoor and outdoor conditions, with respect to the limited energy of IoT devices. We then build a multi-output prediction model of indoor parameters using a random forest learning algorithm, and based on a longitudinal real dataset of one year. Our prediction model considers outdoor conditions to predict the indoor ones. Hence, it helps to detect discomfort situations in real-time when comparing predicted variables to real ones. Furthermore, when detecting an indoor thermal discomfort, we provide a new genetic-based algorithm to find the most suitable values of indoor parameters, enabling the improvement of the indoor occupants’ thermal comfort. Numerical results show the efficiency of our prediction scheme, reaching an accuracy of 96%, as well as our genetic-based scheme in optimizing the indoor thermal parameters by 85%.

Rapid Detection of Pretilachlor in Fish and Water Samples Using a Fluorescence Sensor Based on Dual-Emission Molecularly Imprinted Polymers and Machine Learning Random Forest Algorithm

Rapid Detection of Pretilachlor in Fish and Water Samples Using a Fluorescence Sensor Based on Dual-Emission Molecularly Imprinted Polymers and Machine Learning Random Forest Algorithm

A machine learning approach for accurate crop type mapping using combined SAR and optical time series data

A country’s food needs mainly depend upon its agriculture resources and require reliable information related to crop health, distribution, and acreage estimation to manage and monitor resources to implement a sustainable agricultural system. Different methodologies have been used to collect this information. However, the availability of earth resource satellite data with improved spatial, spectral, and temporal resolutions, such as the European Space Agency's (ESA) Copernicus program satellites Sentinel-1 (S1) and Sentinel-2 (S2), are creating more practicability to generate crop type maps. S1 and S2, both operating in a constellation of twin satellites, carry a C-band Synthetic Aperture Radar (SAR) and Multispectral Instrument (MSI). Vertical transmit and horizontal receive (VH), and vertical transmit and vertical receive (VV) channels of S1 were used to exploit the temporal backscatter of crops present in the study area. In this research, a machine learning random forest classification algorithm is used for accurate crop type mapping through the combination of SAR (S1) and optical (S2) time-series data. Random forest classifier has produced considerable improved accuracies of crop type mapping in previous studies as it uses ensemble decision trees trained on sample data which permit the vote in favor of the most popular land use/ land cover class. The key objectives of this study are to investigate the classification accuracies for different data combinations. Three plots of data are tested (i) S1 (ii) S2 (iii) S1 & S2. A combination of SAR and optical data turn out with the best overall accuracy of 97% and a kappa coefficient of 0.97. Space-borne SAR and optical data add a new aspect of crop type mapping, which increases the classification accuracy by including valuable parameters and beating the drawbacks of each other. By comparing the results, it can be concluded that combining all-weather accessible SAR and spectrally rich optical data accomplished more accurate outcomes. It would be an imperative advance for the future endeavor to estimate crop biomass and biophysical parameters.

Detecting Real-Time Fall of Elderly People Using Machine Learning

Abstract: Fall is a significant national health issue for the elderly people, generally resulting in severe injuries when the person lies down on the floor over an extended period without any aid after experiencing a great fall. Thus, elders need to be cared very attentively. A supervised-machine learning based fall detection approach with accelerometer, gyroscope is devised. The system can detect falls by grouping different actions as fall or non-fall events and the care taker is alerted immediately as soon as the person falls. The public dataset SisFall with efficient class of features is used to identify fall. The Random Forest (RF) and Support Vector Machine (SVM) machine learning algorithms are employed to detect falls with lesser false alarms. The SVM algorithm obtain a highest accuracy of 99.23% than RF algorithm. Keywords: Fall detection, Machine learning, Supervised classification, Sisfall, Activities of daily living, Wearable sensors, Random Forest, Support Vector Machine

Spatio-temporal distribution of sea-ice thickness using a machine learning approach with Google Earth Engine and Sentinel-1 GRD data

Knowledge of ice thickness and its distribution is of great interest to plan ship and offshore operations in ice areas. It is also a major direct factor in climate change. However, it is currently the most inaccurate sea-ice parameter according to a large majority of the scientific community. Satellite-borne synthetic aperture radar (SAR) data are highly valuable for monitoring daily ice-covered oceans. In this study, we develop a new technique based on a machine learning Random Forest (RF) regression approach using the combination of the in-situ thickness measurements and the backscattering information from Sentinel-1 to retrieve the level first-year ice (FYI) thickness. By applying the technique over the Sentinel-1 ground range detected (GRD) data set available in Google Earth Engine (GEE) over the Beaufort Sea spanning a time period from Apr 2015 to Sep 2018, a thickness of up to 1.5 m with a root mean square error (RMSE) of 22 cm is retrieved.

Mixture-based weight learning improves the random forest method for hyperspectral estimation of soil total nitrogen

Visible and near-infrared reflectance spectroscopy is a promising technique to estimate soil total nitrogen, but because of many uncontrolled variations like soil water, developing a high-accuracy soil total nitrogen model is still challenging. This study proposed a new methodology called mixture-based weight learning to reduce the problems stated above. This method improves the estimation accuracy of soil total nitrogen by mixing soil total nitrogen with soil water, then using the random forest method to model soil total nitrogen. A series of different mixtures of soil water and soil total nitrogen were made, and based on the visible and near-infrared spectra measured under laboratory conditions, the optimal model (j = 0.03) was constructed and used as the final mixture-based weight learning–random forest model, which produced better results (R2 of validation = 0.757; root mean square error of validation = 0.235 g/kg; mean relative error of validation = 10.0%; ratio of performance to interquartile range of validation = 2.419) than a random forest model. Our mixture-based weight learning method combined with the random forest method has great potential for the accurate remote retrieval of soil total nitrogen and enhances the available ways to estimate soil properties.