scholarly journals Improving the Performance of Vegetable Leaf Wetness Duration Models in Greenhouses Using Decision Tree Learning

Water ◽  
2019 ◽  
Vol 11 (1) ◽  
pp. 158 ◽  
Author(s):  
Hui Wang ◽  
Jorge Sanchez-Molina ◽  
Ming Li ◽  
Francisco Rodríguez Díaz
Keyword(s):  
Threshold Model ◽  
Estimation Error ◽  
Characteristic Curve ◽  
Regression Tree ◽  
Independent Set ◽  
Classification And Regression Tree ◽  
Leaf Wetness ◽  
Tree Model ◽  
Leaf Wetness Duration ◽  
Input Variables

Leaf wetness duration (LWD) is a key driving variable for peat and disease control in greenhouse management, and depends upon irrigation, rainfall, and dewfall. However, LWD measurement is often replaced by its estimation from other meteorological variables, with associated uncertainty due to the modelling approach used and its calibration. This study uses the decision learning tree method (DLT) for calibrating four LWD models—RH threshold model (RHM), the dew parameterization model (DPM), the classification and regression tree model (CART) and the neural network model (NNM)—whose performances in reproducing measured data are assessed using a large dataset. The relative importance of input variables in contributing to LWD estimation is also computed for the models tested. The LWD models were evaluated at two different greenhouse locations: in a Chinese (CN) greenhouse over three planting seasons (April 2014–October 2015) and in a Spanish (ES) greenhouse over four planting seasons (April 2016–February 2018). Based on multi-evaluation indicators, the models were given a ranking for their assessment capabilities during calibration (in the Spanish greenhouse from April 2016 to December 2016 and in the Chinese greenhouse from April 2014 to November 2014). The models were then evaluated on an independent set of data, and the obtained areas under the receiver operating characteristic curve (AUC) of the LWD models were over 0.73. Therein, the best LWD model in this case was the NNM, with positive predict values (PPVs) of 0.82 (SP) and 0.90 (CN), and mean absolute errors (MAEs) of 1.85 h (SP) and 1.30 h (CN). Consequently, the DLT can decrease LWD estimation error by calibrating the model threshold and choosing black box model input variables.

scholarly journals Model to Enhance Site-Specific Estimation of Leaf Wetness Duration

Plant Disease ◽  
2002 ◽  
Vol 86 (2) ◽  
pp. 179-185 ◽  
Author(s):  
K. S. Kim ◽  
S. E. Taylor ◽  
M. L. Gleason ◽  
K. J. Koehler
Keyword(s):  
Wind Speed ◽  
Regression Tree ◽  
Classification And Regression Tree ◽  
Estimation Accuracy ◽  
Leaf Wetness ◽  
Wind Model ◽  
Site Specific ◽  
Leaf Wetness Duration ◽  
Linear Discriminant ◽  
The Hierarchical Structure

The ability of empirical models to enhance accuracy of site-specific estimates of leaf wetness duration (LWD) was assessed for 15 sites in Iowa, Nebraska, and Illinois during May to September of 1997, 1998, and 1999. Enhanced estimation of LWD was obtained by applying a 0.3-m height correction to SkyBit wind-speed estimates for input to the classification and regression tree/stepwise linear discriminant (CART/SLD) model (CART/SLD/Wind model), compared to either a proprietary model (SkyBit wetness) or to the CART/SLD model using wind speed estimates for a 10-m height. The CART/SLD/Wind model estimated LWD more accurately than the other models during dew-eligible (20:00 to 9:00) as well as dew-ineligible (10:00 to 19:00) periods, and for the period 20:00 to 9:00 regardless of rain events. Improvement of LWD estimation accuracy was ascribed to both the hierarchical structure of decision-making in the CART procedure and wind speed correction. Accuracy of the CART/SLD/Wind model identifying hours as wet or dry varied little among the 15 sites, suggesting that this model may be desirable for estimating LWD from site-specific data throughout the midwestern United States.

scholarly journals Forecasting Site-Specific Leaf Wetness Duration for Input to Disease-Warning Systems

Plant Disease ◽  
2006 ◽  
Vol 90 (5) ◽  
pp. 650-656 ◽  
Author(s):  
K. S. Kim ◽  
M. L. Gleason ◽  
S. E. Taylor
Keyword(s):  
Regression Tree ◽  
Empirical Models ◽  
Classification And Regression Tree ◽  
Weather Data ◽  
Leaf Wetness ◽  
Warning Systems ◽  
Site Specific ◽  
Leaf Wetness Duration ◽  
Regression Tree Analysis ◽  
Cart Model

Empirical models based on classification and regression tree analysis (CART model) or fuzzy logic (FL model) were used to forecast leaf wetness duration (LWD) 24 h into the future, using site-specific weather data estimates as inputs. Forecasted LWD and air temperature then were used as inputs to simulate performance of the Melcast and TOM-CAST disease-warning systems. Overall, the CART and FL models underpredicted LWD with a mean error (ME) of 2.3 and 3.9 h day-1, respectively. The CFL model, a corrected version of the FL model using a weight value, reduced ME in LWD forecasts to -1.1 h day-1. In the Melcast and TOM-CAST simulations, the CART and CFL models predicted timing of occurrence of action thresholds similarly to thresholds derived from on-site weather data measurements. Both models forecasted the exact spray dates for approximately 45% of advisories derived from measurements. When hindcast and forecast estimates derived from site-specific estimates provided by SkyBit Inc. were used as inputs, the CART and CFL models forecasted spray advisories within 3 days for approximately 70% of simulation periods for the Melcast and TOM-CAST disease-warning systems. The results demonstrate that these models substantially enhance the accuracy of commercial site-specific LWD estimates and, therefore, can enhance performance of disease-warning systems using LWD as an input.

scholarly journals Estimating leaf wetness duration over turfgrass, and in a 'Niagara Rosada' vineyard, in a subtropical environment

2008 ◽  
Vol 65 (spe) ◽  
pp. 10-17 ◽  
Author(s):  
Jorge Lulu ◽  
Paulo Cesar Sentelhas ◽  
Mário José Pedro Júnior ◽  
José Ricardo Macedo Pezzopane ◽  
Gabriel Constantino Blain
Keyword(s):  
High Accuracy ◽  
Training System ◽  
Dew Point ◽  
Classification And Regression Tree ◽  
Leaf Wetness ◽  
Good Precision ◽  
Weather Station ◽  
Leaf Wetness Duration ◽  
Confidence Index ◽  
Niagara Rosada

Leaf wetness duration (LWD) is a key parameter in agrometeorology because it is related to plant disease occurrence. As LWD is seldomly measured in a standard weather station it must be estimated to run warning systems for schedule chemical disease control. The objective of the present study was to estimate LWD over turfgrass considering different models with data from a standard weather station, and to evaluate the correlation between estimated LWD over turfgrass and LWD measured in a 'Niagara Rosada' vineyard, cultivated in a hedgerow training system, in Jundiaí, São Paulo State, Brazil. The wetness sensors inside the vineyard were located at the top of the plants, deployed at an inclination angle of 45º and oriented southwest, with three replications. The methods used to estimate LWD were: number of hours with relative humidity above 90% (NHRH > 90%), dew point depression (DPD), classification and regression tree (CART) and Penman-Monteith (PM). The CART model had the best performance to estimate LWD over turfgrass, with a good precision (R² = 0.82) and a high accuracy (d = 0.94), resulting in a good confidence index (c = 0.85). The results from this model also presented a good correlation with measured LWD inside the vineyard, with a good precision (R² = 0.87) and a high accuracy (d = 0.96), resulting in a high confidence index (c = 0.93), showing that LWD in a 'Niagara Rosada' vineyard can be estimated with data from a standard weather station.

scholarly journals Species Diversity and Community Assembly of Cladocera in the Sand Ponds of the Ulan Buh Desert, Inner Mongolia of China

Diversity ◽  
2021 ◽  
Vol 13 (10) ◽  
pp. 502
Author(s):  
Yang-Liang Gu ◽  
Qi Huang ◽  
Lei Xu ◽  
Eric Zeus Rizo ◽  
Miguel Alonso ◽  
...  
Keyword(s):  
Species Richness ◽  
Species Diversity ◽  
Community Assembly ◽  
Inner Mongolia ◽  
Regression Tree ◽  
Environmental Variable ◽  
Classification And Regression Tree ◽  
Tree Model ◽  
Environmental Selection ◽  
Ulan Buh Desert

In deserts, pond cladocerans suffer harsh conditions like low and erratic rainfall, high evaporation, and highly variable salinity, and they have limited species richness. The limited species can take advantage of ephippia or resting eggs for being dispersed with winds in such habitats. Thus, environmental selection is assumed to play a major role in community assembly, especially at a fine spatial scale. Located in Inner Mongolia, the Ulan Buh desert has plenty of temporary water bodies and a few permanent lakes filled by groundwater. To determine species diversity and the role of environmental selection in community assembly in such a harsh environment, we sampled 37 sand ponds in June 2012. Fourteen species of Cladocera were found in total, including six pelagic species, eight littoral species, and two benthic species. These cladocerans were mainly temperate and cosmopolitan fauna. Our classification and regression tree model showed that conductivity, dissolved oxygen, and pH were the main factors correlated with species richness in the sand ponds. Spatial analysis using a PCNM model demonstrated a broad-scale spatial structure in the cladoceran communities. Conductivity was the most significant environmental variable explaining cladoceran community variation. Two species, Moina cf. brachiata and Ceriodaphnia reticulata occurred commonly, with an overlap at intermediate conductivity. Our results, therefore, support that environmental selection plays a major role in structuring cladoceran communities in deserts.

scholarly journals Case Fatality Ratio Estimates for the 2013–2016 West African Ebola Epidemic: Application of Boosted Regression Trees for Imputation

2019 ◽  
Vol 70 (12) ◽  
pp. 2476-2483 ◽  
Author(s):  
Alpha Forna ◽  
Pierre Nouvellet ◽  
Ilaria Dorigatti ◽  
Christl A Donnelly
Keyword(s):  
Characteristic Curve ◽  
Regression Tree ◽  
Case Fatality ◽  
West African ◽  
Boosted Regression Trees ◽  
Survival Outcomes ◽  
Tree Model ◽  
Boosted Regression Tree ◽  
Case Fatality Ratio ◽  
Out Of Sample

Abstract Background The 2013–2016 West African Ebola epidemic has been the largest to date with >11 000 deaths in the affected countries. The data collected have provided more insight into the case fatality ratio (CFR) and how it varies with age and other characteristics. However, the accuracy and precision of the naive CFR remain limited because 44% of survival outcomes were unreported. Methods Using a boosted regression tree model, we imputed survival outcomes (ie, survival or death) when unreported, corrected for model imperfection to estimate the CFR without imputation, with imputation, and adjusted with imputation. The method allowed us to further identify and explore relevant clinical and demographic predictors of the CFR. Results The out-of-sample performance (95% confidence interval [CI]) of our model was good: sensitivity, 69.7% (52.5–75.6%); specificity, 69.8% (54.1–75.6%); percentage correctly classified, 69.9% (53.7–75.5%); and area under the receiver operating characteristic curve, 76.0% (56.8–82.1%). The adjusted CFR estimates (95% CI) for the 2013–2016 West African epidemic were 82.8% (45.6–85.6%) overall and 89.1% (40.8–91.6%), 65.6% (61.3–69.6%), and 79.2% (45.4–84.1%) for Sierra Leone, Guinea, and Liberia, respectively. We found that district, hospitalisation status, age, case classification, and quarter (date of case reporting aggregated at three-month intervals) explained 93.6% of the variance in the naive CFR. Conclusions The adjusted CFR estimates improved the naive CFR estimates obtained without imputation and were more representative. Used in conjunction with other resources, adjusted estimates will inform public health contingency planning for future Ebola epidemics, and help better allocate resources and evaluate the effectiveness of future inventions.

scholarly journals Decision Tree Analysis: A Retrospective Analysis of Postoperative Recurrence of Adhesions in Patients with Moderate-to-Severe Intrauterine

2019 ◽  
Vol 2019 ◽  
pp. 1-8 ◽  
Author(s):  
Ru Zhu ◽  
Hua Duan ◽  
Sha Wang ◽  
Lu Gan ◽  
Qian Xu ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Decision Tree ◽  
Endometrial Thickness ◽  
Characteristic Curve ◽  
Multivariate Logistic Regression Analysis ◽  
Decision Tree Model ◽  
Classification And Regression Tree ◽  
Multivariate Logistic Regression Model ◽  
Multivariate Logistic Regression ◽  
Tree Model

Objective. To establish and validate a decision tree model to predict the recurrence of intrauterine adhesions (IUAs) in patients after separation of moderate-to-severe IUAs. Design. A retrospective study. Setting. A tertiary hysteroscopic center at a teaching hospital. Population. Patients were retrospectively selected who had undergone hysteroscopic adhesion separation surgery for treatment of moderate-to-severe IUAs. Interventions. Hysteroscopic adhesion separation surgery and second-look hysteroscopy 3 months later. Measurements and Main Results. Patients’ demographics, clinical indicators, and hysteroscopy data were collected from the electronic database of the hospital. The patients were randomly apportioned to either a training or testing set (332 and 142 patients, respectively). A decision tree model of adhesion recurrence was established with a classification and regression tree algorithm and validated with reference to a multivariate logistic regression model. The decision tree model was constructed based on the training set. The classification node variables were the risk factors for recurrence of IUAs: American Fertility Society score (root node variable), isolation barrier, endometrial thickness, tubal opening, uterine volume, and menstrual volume. The accuracies of the decision tree model and multivariate logistic regression analysis model were 75.35% and 76.06%, respectively, and areas under the receiver operating characteristic curve were 0.763 (95% CI 0.681–0.846) and 0.785 (95% CI 0.702–0.868). Conclusions. The decision tree model can readily predict the recurrence of IUAs and provides a new theoretical basis upon which clinicians can make appropriate clinical decisions.

Sign in / Sign up

Export Citation Format

Share Document