Lowest Root Mean Square Error Research Articles

Ensemble Voting Regression Based on Machine Learning for Predicting Medical Waste: A Case from Turkey

Predicting medical waste (MW) properly is vital for an effective waste management system (WMS), but it is difficult because of inadequate data and various factors that impact MW. This study’s primary objective was to develop an ensemble voting regression algorithm based on machine learning (ML) algorithms such as random forests (RFs), gradient boosting machines (GBMs), and adaptive boosting (AdaBoost) to predict the MW for Istanbul, the largest city in Turkey. This was the first study to use ML algorithms to predict MW, to our knowledge. First, three ML algorithms were developed based on official data. To compare their performances, performance measures such as mean absolute deviation (MAE), root mean squared error (RMSE), mean absolute percentage error (MAPE), and coefficient of determination (R-squared) were calculated. Among the standalone ML models, RF achieved the best performance. Then, these base models were used to construct the proposed ensemble voting regression (VR) model utilizing weighted averages according to the base models’ performances. The proposed model outperformed three baseline models, with the lowest RMSE (843.70). This study gives an effective tool to practitioners and decision-makers for planning and constructing medical waste management systems by predicting the MW quantity.

Nondestructive detection of kiwifruit textural characteristic based on near infrared hyperspectral imaging technology

ABSTRACT The nondestructive detection of kiwifruit texture has become an important necessity that influences kiwifruit economic efficiency and consumer recognition of the kiwifruit. This study investigated the feasibility of using hyperspectral imaging technology to identify textural characteristics of kiwifruit, and to establish the model with best performance. Firstly, a near-infrared hyperspectral online grading system (1000–2500 nm) was constructed to acquire spectral images. Textural characteristics were measured via three different textural analysis tests: textural profile analysis, puncture test, and shear test. Then, samples were divided into a calibration set and a prediction set based on rank sampling at a ratio of 3:1. Mean centralization, standard normal variate, and multiplicative scatter correction methods were applied to pre-process obtained spectra. Finally, the partial least squares method was used to establish prediction models of kiwifruit textural characteristics. The models of hardness1, chewiness, resilience, peel hardness, average hardness, and corrected hardness achieved good performance. Both the correlation coefficients of calibration (rc) and prediction (rp) values exceeded 0.9. The difference between the root mean square error (RMSE) of calibration and prediction was small, and the ratio of prediction to deviation (RPD) value exceeded 2, which was used to predict kiwifruit quality. However, the model of average shear force had a low accuracy with an RPD of 1.680 and the model of shear force also had an RPD below 1.5 and a low RMSE. The results showed that near-infrared hyperspectral imaging technique can be used as a nondestructive method for the textural characteristics kiwifruit.

Bayesian data analysis for sky-averaged 21-cm experiments in the presence of ionospheric effects

ABSTRACT The ionosphere introduces chromatic distortions on low frequency radio waves, and thus poses a hurdle for 21-cm cosmology. In this paper, we introduce time-varying chromatic ionospheric effects on simulated antenna temperature data of a global 21-cm data analysis pipeline, and try to detect the injected global signal. We demonstrate that given turbulent ionospheric conditions, more than 5 per cent error in our knowledge of the ionospheric parameters could lead to comparatively low evidence and high root-mean-square error (RMSE), suggesting a false or null detection. When using a constant antenna beam for cases that include data at different times, the significance of the detection lowers as the number of time samples increases. It is also shown that for observations that include data at different times, readjusting beam configurations according to the time-varying ionospheric conditions should greatly improve the significance of a detection, yielding higher evidences and lower RMSE, and that it is a necessary procedure for a successful detection when the ionospheric conditions are not ideal.

Efficacy of linear multiple regression and artificial neural network for long-term rainfall forecasting in Western Australia

Precipitation is one of the most intrinsic resources for manifold industrial activities all over Western Australia; consequently, immaculate rainfall prediction is indispensable for flood mitigation as well as water resources management. This study investigated the performance of artificial neural networks (ANN) and Linear multiple regression (LMR) analysis to forecast long-term seasonal spring rainfall in Western Australia, using lagged El Nino Southern Oscillation (ENSO) and Indian Ocean Dipole (IOD) as potential climatic phenomena. The ANN was developed in the form of multilayer perceptron using Levenberg–Marquardt algorithm and subsequently LMR was used with statistical significance for future spring rainfall forecast. The total climatic dataset has been divided into calibration and testing phases to determine the efficacy of the developed models. Different statistical skill tests such as root mean square error (RMSE), mean absolute error (MAE), and Willmott index of agreement ‘d’ were used to assess the efficacy of LMR and ANN modelling. In general, LMR has lower MAE and RMSE values as compared to ANN for most of the stations during calibration and testing periods, whereas ANN models performed better than LMR models based on ‘d’ values. The overall statistical analysis paradigm suggests the efficacy of LMR over ANN models for rainfall forecasting using more climatic variables. As a result, the developed LMR model, incorporated with lagged global climate indices, will facilitate the adequate preparedness for the risks associated with potential droughts in the study region.

Magnetometer-free Kalman filter for motor-based assessment of prodromal Parkinson's disease.

Observing the kinematics of specific motor tasks, such as finger tapping (FT), provides an objective and consistent quantification of the severity of neurodegenerative diseases. However, the current clinical practice mostly relies on visual observations performed by the clinician. Thus, the assessment is subjective. In this paper, we propose a magnetometer-free Kalman filter (KF) to assess FT features using wearable, inertial sensors. The KF was used to assess features during two different FT tasks, namely forefinger tapping (FTAP) and thumb-forefinger tapping (THFF). The proposed KF was validated against a camera-based reference and compared with a strap-down integration-based method. Comparison between KF method and camera reference showed low discrepancies in terms of root mean square error (RMSE) for considered features: namely number of repetitions (RMSE < 0.7), tapping frequency (RMSE < 0.1 Hz), and amplitude (RMSE < 2.6 deg). An high correlation coefficient between amplitudes was also obtained. The proposed KF performed better than the strap-down integration method on both FT tasks, showing lower RMSE on every feature as well as a higher correlation coefficient. Clinical Relevance- The wearable setup, as well as the proposed magnetometer-free KF, may provide a low-cost, easyto- use, non-invasive motion tracking system for protocols aimed to assess motor performances in neurodegenerative disorders.

A comprehensive comparison among metaheuristics (MHs) for geohazard modeling using machine learning: Insights from a case study of landslide displacement prediction

Machine learning (ML) has been extensively applied to model geohazards, yielding tremendous success. However, researchers and practitioners still face challenges in enhancing the reliability of ML models. In the present study, a systematic framework combining k-fold cross-validation (CV), metaheuristics (MHs), support vector regression (SVR), and Friedman and Nemenyi tests was proposed to improve the reliability and performance of geohazard modeling. The average normalized mean square error (NMSE) from k-fold CV sets was adopted as the fitness metric. Twenty of the most well-established MHs and the most recent MHs were adopted to tune the hyperparameters of SVR and were evaluated through nonparametric Friedman and post hoc Nemenyi tests to identify significant differences. Observations from a typical reservoir landslide were selected as a benchmark dataset, and the accuracy, robustness, computational time, and convergence speed of the MHs were compared. Significant performance differences among the twenty MHs were identified by Friedman and post hoc Nemenyi tests of the mean absolute error (MAE), root mean squared error (RMSE), Kling–Gupta efficiency (KGE), and computational time, with p values lower than 0.05. The comparison of results demonstrated that the multiverse optimizer (MVO) is among the highest-performing, most stable, and computationally efficient algorithms, providing superior performance to other methods, with nearly optimum values of the correlation coefficient (R), a low MAE (23.5086 versus 23.9360), a low mean RMSE (48.6946 versus 50.1882), and a high mean KGE (0.9803 versus 0.9893) in predicting the displacement of the Shuping landslide. This paper considerably enriches the literature regarding hyperparameter optimization algorithms and the enhancement of their reliability. In addition, Friedman and post hoc Nemenyi tests have the potential for evaluating and comparing various ML-based geohazard models.

Mapping PROMIS physical function and pain interference to the modified low back pain disability questionnaire.

The Modified Low Back Pain Disability Questionnaire (MDQ) is a commonly used tool to assess functioning of patients with low back pain (LBP). Recently, the Patient-Reported Outcomes Measurement Information System (PROMIS) was suggested as an alternative platform to assess LBP patient-reported health. We sought to map between the MDQ and PROMIS Physical Function (PROMIS-PF) and Pain Interference (PROMIS-PI) scales using multiple methods. In a retrospective analysis of LBP patients seen at Cleveland Clinic 11/14/18-12/11/19, T-scores from each PROMIS scale were mapped to MDQ total score individually and together. MDQ item and total scores were mapped to each PROMIS scale. Linear regression as well as linear and equipercentile equating were used. Split sample internal validation using root mean squared error (RMSE), mean absolute error (MAE), and correlations were used to assess accuracy of mapping equations. 13585 patients completed the three scales. In the derivation cohort, average age was 59.0 (SD = 15.8); 53.3% female and 82.9% white. Average MDQ total, PROMIS-PF, and PROMIS-PI T-scores were 40.3 (SD = 19.0), 37.2 (SD = 7.6), and 62.9 (SD = 7.2), respectively. For estimating MDQ total scores, methods that used both PROMIS-PF and PROMIS-PI had closest estimated means, lowest RMSE and MAE, and highest correlations. For estimating each of PROMIS-PF and PROMIS-PI T-scores, the best performing method was equipercentile equating using the MDQ items. We created and internally validated maps between MDQ and PROMIS-PF and PROMIS-PI using linear regression, linear and equipercentile equating. Our equations can be used by researchers wishing to translate scores between these scales.

Optimal Strategy on Radiation Estimation for Calculating Universal Thermal Climate Index in Tourism Cities of China.

The Universal Thermal Climate Index (UTCI) is believed to be a very powerful tool for providing information on human thermal perception in the domain of public health, but the solar radiation as an input variable is difficult to access. Thus, this study aimed to explore the optimal strategy on estimation of solar radiation to increase the accuracy in UTCI calculation, and to identify the spatial and temporal variation in UTCI over China. With daily meteorological data collected in 35 tourism cities in China from 1961 to 2020, two sunshine-based Angstrom and Ogelman models, and two temperature-based Bristow and Hargreaves models, together with neural network and support vector machine-learning methods, were tested against radiation measurements. The results indicated that temperature-based models performed the worst with the lowest NSE and highest RMSE. The machine-learning methods performed better in calibration, but the predictive ability decreased significantly in validation due to big data requirements. In contrast, the sunshine-based Angstrom model performed best with high NSE (Nash–Sutcliffe Efficiency) of 0.84 and low RMSE (Root Mean Square Error) of 35.4 J/m2 s in validation, which resulted in a small RMSE of about 1.2 °C in UTCI calculation. Thus, Angstrom model was selected as the optimal strategy on radiation estimation for UTCI calculation over China. The spatial distribution of UTCI showed that days under no thermal stress were high in tourism cities in central China within a range from 135 to 225 days, while the largest values occurred in Kunming and Lijiang in southwest China. In addition, days under no thermal stress during a year have decreased in most tourism cities of China, which could be attributed to the asymmetric changes in significant decrease in frost days and slightly increase in hot days. However, days under no thermal stress in summer time have indeed decreased, accompanying with increasing days under strong stress, especially in the developed regions such as Yangze River Delta and Zhujiang River Delta. Based on the study, we conclude that UTCI can successfully depict the overall spatial distribution and temporal change of the thermal environments in the tourism cities over China, and can be recommend as an efficient index in the operational services for assessing and predicting thermal perception for public health. However, extreme cold and heat stress in the tourism cities of China were not revealed by UTCI due to mismatch of the daily UTCI with category at hourly scale, which makes it an urgent task to redefine category at daily scale in the next research work.

Comparison of Existing Methods for Characterizing Bi-Linear Natural Ankle Quasi-Stiffness.

Natural ankle quasi-stiffness (NAS) is a mechanical property of the ankle joint during dynamic motion. NAS has been historically calculated as the average slope (linear regression) of the net ankle moment versus ankle angle during discrete phases of stance. However, recent work has shown that NAS is nonlinear during the stance phase. Specifically, during the loading phase of stance (∼10 to 60% of total stance), plantarflexion moment increases at an accelerating rate compared to dorsiflexion angle. Updated models have been developed to better capture this inherent nonlinearity. One type of model called bi-linear NAS (BL-NAS) divides the loading phase of stance into two subphases, called early loading (EL) and late loading (LL) NAS. Two papers, written by Crenna and Frigo (2011, "Dynamics of the Ankle Joint Analyzed Through Moment-Angle Loops During Human Walking: Gender and Age Effects," Hum. Mov. Sci., 30(6), pp. 1185-1198) and Shamaei et al. (2013, "Estimation of Quasi-Stiffness and Propulsive Work of the Human Ankle in the Stance Phase of Walking," PLoS One, 8(3), p. e59935), outline different BL-NAS models. Both models fit measured data better (lower root-mean-squared error (RMSE)) than standard single linear NAS (SL-NAS) models but have not been widely adopted, possibly because of methodological discrepancies and lack of applicability to physical devices at the time. This paper compares and contrasts these existing BL-NAS models and translates those findings to possible orthotic device designs. Results showed that both BL-NAS models had lower RMSE than SL-NAS, EL-NAS was not significantly different across walking speeds, and LL-NAS increased significantly at faster walking speeds. These improved models of NAS much better approximate natural human movement than commonly used SL-NAS models, and thus provide a basis to design ankle-foot devices with multiple stiffness properties to emulate and facilitate natural human motion.

Comparison of Two Corrector Surface Models of Orthometric Heights from GPS/Levelling Observations and Global Gravity Model

The advent of space-based measurement systems such as the Global Positioning System (GPS) offers a new alternative in orthometric height determination over conventional spirit levelling. The ellipsoidal height (h) obtained from GPS observations can be transformed into orthometric height if the geoid undulation (N) is known from a national gravimetric geoid model. However, the lack of a national geoid model in Nigeria hinders the use of the method. This study compares two corrector surface models of orthometric heights from GPS/levelling observations and the Global Gravity Model. Model A (7-parameter) and Model B (8-parameter) are based on the general 7-parameter similarity datum shift transformation. A network of twenty-one (21) GPS/levelling benchmarks within the study area were used and their geoidal heights were computed using GeoidEval utility software with reference to Global Gravitational Model (EGM08). Least squares adjustment was used to compute the coefficients of the models. Root mean square error (RMSE) was used to assess the accuracy of the models with model A having RMSE=0.171m and model B having RMSE=0.169m. Model B with the lowest RMSE is hence the better of the two models. The t-test and hypothesis test conducted at a 95% confidence level, however, revealed that the two models did not differ significantly. The study shows that the use of corrective surface to combine the gravity field model EGM08 and GPS/levelling significantly improves the determination of heights as observed from GPS in the study area.

Investigation of the Missing Data Imputation Methods on Characteristic Curve Transformation Methods Used in Test Equating

In this research, the aim was to evaluate the effect of zero imputation and multiple imputation missing data handling methods on item response theory (IRT) based test equating methods under different conditions. Data in this study was obtained from the administration of the TIMSS 2019 eighth-grade science test. Data sets were formed by randomly selecting a sample of 1000 students with full data from booklets 7 and 8. By deleting data under a completely random missing data mechanism within the scope of common-item nonequivalent groups (CINEG) design, four different data sets were obtained with the missing data rates of 10% or 20% in the new test or in both tests. The missing da ta problem was solved by using zero imputation and multiple imputation methods from these data sets. In this way, 8 different data sets were formed. Then, scaling transformation was performed by using characteristic curve transformation methods (Haebara, Stocking-Lord). Test equating results were reported in terms of observed scores. The root mean square error (RMSE) was used as the evaluation criterion to determine the error involved in test equating. As a result, it was determined that in the case of 10% missing data in both tests, generally lower RMSE values were obtained. It was observed that the multiple imputation method, one of the methods for handling missing data, was the method that produced RMSE values that were both the lowest and closer to the full data set as a reference value compared to the zero-imputation method. In addition, it was determined that, when compared to the Haebara method, Stocking-Lord method, one of the characteristic curve transformation methods, produced lower RMSE values and these values were closer to the full data set, which was taken as a reference value.

Evaluation of Selected Empirical Schemes of Calculating Sensible Heat Flux from Routinely-Measured Meteorological Parameters in a Tropical Location

The accurate estimation of sensible heat flux (Hs) is imperative in the exchange of energy and mass between the earth’s systems. However, the process of acquiring the sensible heat flux data is challenging due to the complexity in its direct measurement hence, not readily available. The alternative is to use estimations which are derived from specific models. This study evaluated the performances of five selected schemes of estimating Hs from routinely-measured meteorological parameters using statistical methods such as Mean Bias Error (MBE), Mean Percentage Difference (MPD), Root Mean Square Error (RMSE), Index of Agreement (IA), Correlation Coefficient (R) and Coefficient of Determination (R2). The empirical schemes selected are: Berkowicz and Prahm (BP), Bowen Ratio Energy Balance (BREB), Holstlag and Van Ulden (HU), Smith (ST) and Surface Temperature (ST) schemes. The results obtained revealed that in the dry season, MBE values of 6.9 Wm-2, 7.2 Wm-2, 7.5 Wm-2, 6.3 Wm-2 and -3.8 Wm-2 were obtained for BP, BREB, HU, SMT and ST respectively. In the wet season, BP, BREB, HU, SMT and ST had MBE values of 37.4 Wm-2, 14.2 Wm-2, 4.2 Wm-2, 12.0 Wm-2 and -1.1 Wm-2 respectively. The MBEs, RMSEs and MPDs obtained for the schemes were higher in the wet season (about 70 %) than in the dry season; implying that the schemes performed better in the dry season than in the wet season. The study rated BREB as the best method of estimating the sensible heat flux at the study location having the lowest MBE, lowest MPD, lowest RMSE, high R, high R2 and high IA in both seasons.

Cross-Comparison of channel parameters between FY-3E/MERSI-LL and Himawari-8/AHI in China

ABSTRACT Fengyun-3E (FY-3E) is the world’s first operational meteorological satellite on dawn-dusk orbit. In order to realize the generality of algorithms for different satellites and provide a basis for subsequent inversion studies on remote sensing parameters, such as land surface temperature, fire, and atmospheric water vapour, we perform a cross-comparison of the brightness temperature (TB) of the six infrared (IR) channels between the Medium Resolution Spectral Imager Level L onboard the FY-3E satellite (FY-3E/MERSI-LL) and the Advanced Himawari Imager onboard the Himawari-8 satellite (Himawari-8/AHI). The statistical indicators, correlation coefficient, root mean square error (RMSE) and mean bias, are used to analyse the consistency and differences of the channel parameters between the MERSI-LL and AHI in terms of overall data, different regions and different TB. The results show that the TB data observed by the FY-3E/MERSI-LL and Himawari-8/AHI are generally consistent, with correlation coefficients of more than 0.99, mean biases of less than 1.60 K and RMSEs of less than 1.90 K. In terms of different channels, the TB in channel IR 038 has the best consistency between the two sensors, while the results from the MERSI-LL are higher than those of the AHI for channels IR038, IR086, IR108 and IR120, and lower than those of the AHI for channel IR041. Analysing the situation in different regions, we find that the MERSI-LL and AHI data are highly consistent and relatively stable in the Huang-Huai Region, with low RMSEs, low mean biases and high correlation coefficients (more than 0.95) at all channels. For the different TB ranges, the RMSEs and mean biases for the TB ranges of 270–280 K and 280–290 K between the MERSI-LL and AHI are small, and the data in the 290–300 K TB range has great differences and fluctuations between the two sensors.

Generating CT images in delayed PET scans using a multi-resolution registration convolutional neural network

A Delayed scan refers to the reacquisition of CT and PET after a regular scan to improve the sensitivity and specificity of PET/CT examination. However, an additional CT scan will lead much more X-ray radiation to the patient. Therefore, developing a method to generate delayed CT (T2CT) images to avoid additional CT scans is particularly important in clinic. This paper aims to generate T2CT images from delayed PET (T2PET), regular PET (T1PET) and regular CT (T1CT) images using deep learning methods. However, it may encounter difficulties such as intrinsic differences between multi-modal images and large deformations caused by two scans. To address these issues, a multi-resolution registration convolutional neural network (MRR-CNN) is introduced to improve the accuracy of generating CT images. MRR-CNN employs three models to separately predict deformation vector field (DVF) in different resolution levels. In this method, the global deformation is evaluated firstly, and then local deformations are gradually fused to generate accurate T2CT images. We selected a recently published deep learning-based method (VoxelMorph) to compare the effectiveness of our method on 10 clinical patient data, using mean absolute error (MAE) and root mean square error (RMSE) as evaluation metrics. Compared with VoxelMorph, the proposed MRR-CNN achieves lower MAE (61.26 vs. 67.24) and lower RMSE (118.74 vs. 126.13). The experimental results indicate that our proposed method outperforms VoxelMorph in generating T2CT images.

Examination of rehydration ratio, color properties and drying kinetics of microwave dried garlic (Allium sativum L.)

In this study, color values and rehydration ratios of dried garlic slices at four different microwave power (300W, 450W, 600W and 700W) were determined. Thin layer drying models were used for drying kinetics. It was determined that L* and b* values were lower and a* values were higher in microwave-dried garlic compared to fresh garlic slices. It was observed that the C* (Chroma) and h (hue angle) values of garlic decreased as the microwave power increased compared to fresh garlic. It was determined that the rehydration ratio decreased significantly with increasing microwave power. Therefore, the highest rehydration ratio was detected in the garlic samples applied at 300W. Moisture ratio (MR) versus time calculated as a result of microwave drying was compared with 5 (Lewis, Page, Henderson, and Pabis, Midilli et al. and Logarithmic) drying models in the literature. R2 (coefficient of expression), χ2 (chi-square) and RMSE (root mean square error) were calculated by SPSS and nonlinear regression analysis of dried garlic samples. It was determined that the highest R2 (0.994-0.998, the lowest RMSE (0.016-0.022) and χ2 (0.00033-0.00064) values belonged to the Lewis model.

Assessment and Prediction of Free Fatty Acids Changes in Maize Based on Effective Accumulated Temperature in Large Granaries

ABSTRACT In large granaries, maize storage period could last 3–5 years in China. Free fatty acids (FFA) content is commonly used as a sensitivity indicator of grain quality changes during storage. Samples of the stored grain are taken manually and tested in the laboratory monthly for regular quality monitoring. Although substantial labor and time are consumed, the testing way lacks of real-time. Temperature is a main factor influencing the quality of stored grain. In this study, it was analyzed that the effective accumulated temperature (EAT) had a strong correlation coefficient with FFA. Classification and regression trees (CART) and model trees (ML) in machine learning (ML) methods were used to estimate FFA by EAT based on data collected from 11 large granaries in northeastern China. While the minimum number of samples for segmentation was set as 20, the two models both had the optimal performance. The two models were evaluated by mean absolute error (MAE), root mean square error (RMSE) and the coefficient of determination (R2). The MAE, RMSE and R2 of CART and MT are 1.296, 1.761, 0.759; 1.247, 1.821, and 0.741, respectively. As CART had the lower RMSE and the larger R2, the model performance of CART was better. The model provided a method to estimate the changes of FFA based on the exist temperature monitoring system in large granaries as a way of stored maize quality real-time monitoring.

Evaluation of air temperature estimated by ERA5-Land reanalysis using surface data in Pernambuco, Brazil.

This study aimed to analyze the average air temperature data estimated by ERA5-Land reanalysis over 10years (01/01/2011 to 12/31/2020), based on data from 12 automatic weather stations located in different cities in the state of Pernambuco, northeast of Brazil. For more careful evaluation, the average air temperature data were stratified by mesoregions of the state. ERA5-Land reanalysis data were validated using statistical indices that evaluated the accuracy. The average air temperature estimated by ERA5-Land reanalysis agree well with weather stations in almost the entire state of Pernambuco. The highest accuracy of the average air temperature estimated by ERA5-Land occurred in the city of Ibimirim (R2 = 0.98), while the lowest accuracy was measured in the city of Caruaru (R2 = 0.57). Root-mean-square error (RMSE) generated by the ERA5-Land reanalysis was lower than 0.60°C in most Pernambuco. The highest RMSE of ERA5-Land for average air temperature was calculated using data from the city of Caruaru (1.11°C), whereas the lowest RMSE was obtained with data from the city of Recife (0.41°C). According to mean bias errors, (MBE) the ERA5-Land underestimated the average air temperature data compared to automatic weather stations data, especially in Ouricuri, which had the highest underestimation (-0.80°C). On the other hand, among the municipalities where ERA5-Land overestimated the temperature values, the highest overestimation was identified in Garanhuns (0.35°C). Therefore, based on the results of this study, ERA5-Land reanalysis successfully estimated the average air temperature for the state of Pernambuco.

Prediction of geoid undulations: Random forest versus classic interpolation techniques

AbstractLocal geoid determination studies are commonly carried out today to establish the relationship between the ellipsoidal height determined by satellite geodesy methods and the orthometric height found using geoid undulation . The aim of this study was to determine a local geoid using the kriging, local polynomial (LP), and inverse distance to a power (IDP) interpolation methods along with the random forest (RF) regression method and to compare the performance. For the application, 193 Global Navigation Satellite System (GNSS)/leveling points homogeneously distributed over the study area were selected as reference data, and 70 GNSS/leveling points were selected as test data. The results were compared in terms of accuracy using well‐known performance metrics, namely, the root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R2) values. According to the results, the highest R2 and the lowest RMSE and MAE values were obtained with the RF regression method. These findings demonstrated the superiority of the RF regression over the classic interpolation methods applied in the local geoid determination for the study area.

A comparative analysis of data mining techniques for agricultural and hydrological drought prediction in the eastern Mediterranean

Drought is a natural hazard which affects ecosystems in the eastern Mediterranean. However, limited historical data for drought monitoring and forecasting are available in the eastern Mediterranean. Thus, implementing machine learning (ML) algorithms could allow for the prediction of future drought events. In this context, the main goals of this research were to capture agricultural and hydrological drought trends by using the Standardized Precipitation Index (SPI) and to assess the applicability of four ML algorithms (bagging (BG), random subspace (RSS), random tree (RT), and random forest (RF)) in predicting drought events in the eastern Mediterranean based on SPI-3 and SPI-12. The results reveal that hydrological drought (SPI-12, −24) was more severe over the study area, where most stations showed a significant (p < 0.05) negative trend. The accuracy of ML algorithms in drought prediction varied in relation to the implementation stage. In the training stage, RT outperformed the other algorithms (Root mean square error (RMSE) = 0.3, Correlation Coefficient (r) = 0.97); the performance of the algorithms can be ranked as follows: RT > RF > BG > RSS for both SPI-3 and SPI-12. In the testing stage, both the BG and RF algorithms had the highest correlation r(observed vs. predicted) (0.58–0.64) and lowest RMSE (0.68–0.88). In contrast, the lowest correlation r(observed vs. predicted) (0.3–0.41) and highest RMSE (0.94–1.10) was calculated for the RT algorithm. However, BG was more dynamic in drought capturing, with the lowest RMSE and highest correlation. In the validation stage, the BG performance was satisfactory (RMSE = 0.62–0.83, r = 0.58–0.79). The output of this research will help decision-makers with drought mitigation plans by using the new four machine learning algorithms.

An enhanced spatiotemporal fusion method – Implications for DNN based time-series LAI estimation by using Sentinel-2 and MODIS

The consequent and accurate monitoring of the seasonal dynamics of crop leaf area index (LAI) is critical to yield estimation and agriculture policy development. It is difficult for a single sensor to balance spatial and temporal resolution. Spatiotemporal fusion is an effective way to meet the need for high spatial and temporal applications. Among the fusion methods, regression model Fitting, spatial Filtering and residual Compensation (Fit-FC) may be recommended for vegetation dynamic monitoring because of its outperformance for the cases with considerable phenological changes. However, it is not good at capturing image structure and textures. To overcome the limitations, an enhanced version of Fit-FC, referred as Enhanced-Fit-FC (EFF), was developed. The EFF method can be applied with one (unidirectional prediction, Uni-EFF) or two (bidirectional prediction, Bi-EFF) coarse-fine image pairs as input for near real-time or post-growth applications. The visual and quantitative assessments indicated that EFF mitigated the blurring effect of Fit-FC and generated accurate reflectance, especially Bi-EFF contributed to capturing land cover changes. Compared with Fit-FC, the correlation coefficient (CC) and quality index (QI) of EFF increased by more than 0.12, and the root mean square error (RMSE) decreased by 0.16 at the maximum. Further, we identified the robustness and adaptability of deep neural network (DNN) model for time-series LAI estimation. The results substantiated the effectiveness of DNN in dealing with nonlinear problems and alleviating spectral saturation with higher CC and lower RMSE and relative RMSE (rRMSE) at whole growth-stages (CC= 0.91, RMSE= 0.28, and rRMSE= 7.18%) and vegetative stage (CC=0.94, RMSE=0.24, and rRMSE= 5.81%). In conclusion, the EFF method proposed in this study is competent in constructing time-series synthetic images for near real-time or post-growth applications. Moreover, the DNN model shows the potential for LAI estimation at whole growth-stages. This research not only contributes to dense time-series remote sensing-based applications (e.g., land cover change monitoring, yield forecast), but also is valuable for high-spatial-precision farmland management. • A framework to enhance spatiotemporal fusion methods was constructed for near real-time or post-growth applications. • The Enhanced-Fit-FC (EFF) method provided more accurate reflectance and spatial details compared with Fit-FC. • The bidirectional prediction of EFF (Bi-EFF) contributed to capturing land cover changes. • The DNN model was potential in alleviating spectral saturation for LAI estimation at whole growth-stages.