Comparison of Multiple Linear Regression, Cubist Regression, and Random Forest Algorithms to Estimate Daily Air Surface Temperature from Dynamic Combinations of MODIS LST Data

Phan, Thanh Noi; Degener, Jan; Kappas, Martin

doi:10.3390/rs9050398

Cited by 145 publications

(87 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This randomness and requiring no assumption of the probability distribution in predictive variables, increase model prediction accuracy and robustness against over-fitting [80,81]. Finally, an optimal prediction model is generated by aggregating all the "trees" that form the "forest" [82,83].…”

Section: Comparison To Other Modeling Methodsmentioning

confidence: 99%

Dual Activation Function-Based Extreme Learning Machine (ELM) for Estimating Grapevine Berry Yield and Quality

Maimaitiyiming

Sagan

Kwasniewski³

2019

Remote Sensing

View full text Add to dashboard Cite

Reliable assessment of grapevine productivity is a destructive and time-consuming process. In addition, the mixed effects of grapevine water status and scion-rootstock interactions on grapevine productivity are not always linear. Despite the potential opportunity of applying remote sensing and machine learning techniques to predict plant traits, there are still limitations to previously studied techniques for vine productivity due to the complexity of the system not being adequately modeled. During the 2014 and 2015 growing seasons, hyperspectral reflectance spectra were collected using a handheld spectroradiometer in a vineyard designed to investigate the effects of irrigation level (0%, 50%, and 100%) and rootstocks (1103 Paulsen, 3309 Couderc, SO4 and Chambourcin) on vine productivity. To assess vine productivity, it is necessary to measure factors related to fruit ripeness and not just yield, as an over cropped vine may produce high-yield but poor-quality fruit. Therefore, yield, Total Soluble Solids (TSS), Titratable Acidity (TA) and the ratio TSS/TA (maturation index, IMAD) were measured. A total of 20 vegetation indices were calculated from hyperspectral data and used as input for predictive model calibration. Prediction performance of linear/nonlinear multiple regression methods and Weighted Regularized Extreme Learning Machine (WRELM) were compared with our newly developed WRELM-TanhRe. The developed method is based on two activation functions: hyperbolic tangent (Tanh) and rectified linear unit (ReLU). The results revealed that WRELM and WRELM-TanhRe outperformed the widely used multiple regression methods when model performance was tested with an independent validation dataset. WRELM-TanhRe produced the highest prediction accuracy for all the berry yield and quality parameters (R2 of 0.522–0.682 and RMSE of 2–15%), except for TA, which was predicted best with WRELM (R2 of 0.545 and RMSE of 6%). The results demonstrate the value of combining hyperspectral remote sensing and machine learning methods for improving of berry yield and quality prediction.

show abstract

Section: Comparison To Other Modeling Methodsmentioning

confidence: 99%

Dual Activation Function-Based Extreme Learning Machine (ELM) for Estimating Grapevine Berry Yield and Quality

Maimaitiyiming

Sagan

Kwasniewski³

2019

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Since available, MODIS LST data has been used for various studies, such as evaluating and monitoring urban heat islands [11][12][13][14][15], estimating air surface temperatures (Ta) [16][17][18][19], retrieving soil moisture [20,21], drought assessment [22], and hydrology applications [23]. Most of these studies have shown that the changes of land surface properties (e.g., normalized difference vegetation index (NDVI), elevation, and land use/cover types) will result in the variations of LST [3].…”

Section: Introductionmentioning

confidence: 99%

“…The majority of these studies used point data from weather stations located 1.5 m to 2.0 m above the land surface [33][34][35]. LST and Ta are not equivalent, and their relationship is multifaceted due to the complex surface energy budget and the multiple related variables between them [17][18][19]36]. Therefore, the relationship between Ta and elevation is different from those of LST and elevation.…”

Section: Introductionmentioning

confidence: 99%

Land Surface Temperature Variation Due to Changes in Elevation in Northwest Vietnam

Phan

Kappas

Tran

2018

Climate

Self Cite

View full text Add to dashboard Cite

Land surface temperature (LST) is one of the most important variables for applications relating to the physics of land surface processes. LST rapidly changes in both space and time, and knowledge of LST and its spatiotemporal variation is essential to understand the interactions between human activity and the environment. This study investigates the spatiotemporal variation of LST according to changes in elevation. The newest version (version 6) of MODIS LST data for 2015 was used. An area of 40,000 km 2 (200 × 200 km 2 ) in northwest Vietnam with elevations ranging from 8 m to 3165 m was chosen as a case study. Our results showed that the drop in LST with increased elevation varied throughout the year during both the daytime and nighttime. The monthly averages in 2015 and an altitude increase of 1000 m resulted in a decrease in LST ranging from 3.8 • C to 6.1 • C and 1.5 • C to 5.8 • C for the daytime and nighttime, respectively. This suggests that in any study relating to the spatial distribution of LST, the effect of elevation on LST should be considered. In addition, the effects of land use/cover and elevation distribution on the relationship between LST and elevation are discussed.

show abstract

“…They concluded that CB was the best choice for predicting urban impervious surface cover. Noi et al (2017) compared the results of Multiple Linear Regression, Cubist Regression, and Random Forest Algorithms in estimation of daily air surface temperature. They concluded that using different combinations of data, RF or CB algorithms resulted in high accuracies.…”

Section: Introductionmentioning

confidence: 99%

Predicting tile drainage discharge using machine learning algorithms

Motarjemi

Möller

Plauborg

et al. 2020

Preprint

View full text Add to dashboard Cite

Abstract. Drainage systems can significantly improve the water management in agricultural fields. However, they may transport contaminants originating from fertilizers and pesticides and threaten ecosystems. Determining the quantity of drainage water is an important factor for constructed wetlands and other drainage mitigation techniques. This study was carried out in Denmark where tile drainage systems are implemented in more than half of the agricultural fields. The first aim of the study was to predict the annual discharge of tile drainage systems using machine-learning methods, which have been highly popular in recent years. The second objective was to assess the importance of the parameters and their impact on the predictions. Data from 53 drainage stations distributed in different regions of Denmark were collected and used for the analysis. The covariates contained 35 parameters including the calculated percolation and geographic variables such as drainage probability, clay content in different depth intervals, and elevation, all extracted from existing national maps. Random Forest and Cubist were selected as predictive models. Both models were trained on the dataset and used to predict yearly drainage discharge. Results highlighted the importance of the cross-validation methods and indicated that both Random Forest and Cubist can perform as predictive models with a low complexity and good correlation between predicted and observed discharge. Covariate importance analysis showed that among all of the used predictors, the percolation and elevation have the largest effect on the prediction of tile drainage discharge. This work opens up for a better understanding of the dynamics of tile drainage discharge and proves that machine-learning techniques can perform as predictive models in this specific concept. The developed models can be used in regard to a national mapping of expected tile drain discharge.

show abstract

Comparison of Multiple Linear Regression, Cubist Regression, and Random Forest Algorithms to Estimate Daily Air Surface Temperature from Dynamic Combinations of MODIS LST Data

Cited by 145 publications

References 44 publications

Dual Activation Function-Based Extreme Learning Machine (ELM) for Estimating Grapevine Berry Yield and Quality

Dual Activation Function-Based Extreme Learning Machine (ELM) for Estimating Grapevine Berry Yield and Quality

Land Surface Temperature Variation Due to Changes in Elevation in Northwest Vietnam

Predicting tile drainage discharge using machine learning algorithms

Contact Info

Product

Resources

About