scholarly journals Development of a Wilks feature importance method with improved variable rankings for supporting hydrological inference and modelling

2021 ◽  
Vol 25 (9) ◽  
pp. 4947-4966
Author(s):  
Kailong Li ◽  
Guohe Huang ◽  
Brian Baetz
Keyword(s):  
Predictive Accuracy ◽  
Yellow River ◽  
Yellow River Basin ◽  
The United States ◽  
Test Statistic ◽  
Relative Significance ◽  
Feature Importance ◽  
The Yellow River Basin ◽  
Abstract Feature ◽  
Stable Variable

Abstract. Feature importance has been a popular approach for machine learning models to investigate the relative significance of model predictors. In this study, we developed a Wilks feature importance (WFI) method for hydrological inference. Compared with conventional feature importance methods such as permutation feature importance (PFI) and mean decrease impurity (MDI), the proposed WFI aims to provide more reliable variable rankings for hydrological inference. To achieve this, WFI measures the importance scores based on Wilks Λ (a test statistic that can be used to distinguish the differences between two or more groups of variables) throughout an inference tree. Compared with PFI and MDI methods, WFI does not rely on any performance measures to evaluate variable rankings, which can thus result in less biased criteria selection during the tree deduction process. The proposed WFI was tested by simulating monthly streamflows for 673 basins in the United States and applied to three interconnected irrigated watersheds located in the Yellow River basin, China, through concrete simulations for their daily streamflows. Our results indicated that the WFI could generate stable variable rankings in response to the reduction of irrelevant predictors. In addition, the WFI-selected predictors helped random forest (RF) achieve its optimum predictive accuracy, which indicates that the proposed WFI could identify more informative predictors than other feature importance measures.

scholarly journals Gaining Hydrological Insights Through Wilk's Feature Importance: A Test-Statistic Interpretation method for Reliable and Robust Inference

2021 ◽  
Author(s):  
Kailong Li ◽  
Guohe Huang ◽  
Brian Baetz
Keyword(s):  
Machine Learning ◽  
Predictive Accuracy ◽  
Yellow River ◽  
Yellow River Basin ◽  
Machine Learning Algorithms ◽  
Recursive Feature Elimination ◽  
Test Statistic ◽  
Relative Significance ◽  
Testing Dataset ◽  
Feature Importance

Abstract. Feature importance has been a popular approach for machine learning models to investigate the relative significance of model predictors. In this study, we developed a Wilk's feature importance (WFI) method for hydrological inference. Compared with conventional feature importance methods such as permutation feature importance (PFI) and mean decrease in impurity (MDI), the proposed WFI aims to provide more reliable importance scores that could partially address the equifinality problem in hydrology. To achieve this, the WFI measures the importance scores based on Wilk's Ʌ (a test-statistic that can be used to distinguish the differences between two or more groups of variables) throughout a decision tree. The WFI has an advantage over PFI and MDI as it does not account for predictive accuracy so the risk of overfitting will be greatly reduced. The proposed WFI was applied to three interconnected irrigated watersheds located in the Yellow River Basin, China. By employing the recursive feature elimination approach, our results indicated that the WFI could generate more stable relative importance scores in response to the reduction of irrelevant predictors, as compared with PFI and MDI embedded in three different machine learning algorithms. In addition, the comparative study also shows that the predictors identified by WFI achieved the highest predictive accuracy on the testing dataset, which indicates the proposed WFI could identify more informative predictors among many irrelevant ones. We also extended the WFI to the local importance scores for reflecting the varying characteristics of a predictor in the hydrological processes. The related findings could help to gain insights into different hydrological behaviours.

scholarly journals Inheritors of the Yellow River: the relationship of heritage making practices to cultural self-confidence in China

2021 ◽  
Vol 5 (1) ◽  
Author(s):  
Wanlin Zhang ◽  
Stuart Walker ◽  
Martyn Evans ◽  
Julia Bennett
Keyword(s):  
Cultural Awareness ◽  
Yellow River ◽  
Yellow River Basin ◽  
The Yellow River ◽  
Chinese Government ◽  
Active Involvement ◽  
Self Confidence ◽  
Support Mechanisms ◽  
The Yellow River Basin ◽  
Relationship Of

AbstractThrough active involvement in UNESCO’s ICH (Intangible Cultural Heritage) programme, China has developed its own framework to support traditional making practices. To examine the ‘characteristics’ of heritage crafts preservation in China, we undertook empirical research in the Yellow River basin. Our research involved in-depth observations and key informant interviews with a range of highly accomplished craftspeople who have inherited their particular expertise from their family and have been officially designated ICH Inheritors. Through our qualitative research with inheritors, a businessperson and a government official, we identified various support mechanisms employed by the Chinese government that aim to protect traditional culture while also recognising outstanding individuals. We also found that ICH Inheritors play an important role in raising cultural awareness and enhancing cultural confidence through their creative activities and making practices, craft businesses and transfer of expertise. Based on the findings, potential areas where designers may collaborate with inheritors were identified.

scholarly journals Nitrogen Fertilization Effects on Physiology of the Cotton Boll–Leaf System

Agronomy ◽  
2019 ◽  
Vol 9 (6) ◽  
pp. 271
Author(s):  
Jing Chen ◽  
Liantao Liu ◽  
Zhanbiao Wang ◽  
Hongchun Sun ◽  
Yongjiang Zhang ◽  
...  
Keyword(s):  
River Basin ◽  
Nitrogen Fertilization ◽  
Yellow River ◽  
Yellow River Basin ◽  
Growth And Yield ◽  
The Yellow River ◽  
Nitrogen Rate ◽  
The Yellow River Basin ◽  
Source Sink ◽  
Subtending Leaves

The objective of this study was to assess the impacts of nitrogen on the physiological characteristics of the source–sink system of upper fruiting branches under various amounts of nitrogen fertilization. A two-year field experiment was conducted with a Bt cotton cultivar in the Yellow River Basin of China. The growth and yield of cotton of the upper fruiting branches were compared under four nitrogen levels: Control (N0, 0 kg ha−1), low nitrogen (N1, 120 kg ha−1), moderate nitrogen (N2, 240 kg ha−1), and high nitrogen (N3, 480 kg ha−1). The results indicated that in the subtending leaves in upper fruiting branches, chlorophyll content, protein content, and peroxidase (POD) activity dramatically increased with nitrogen application, reaching the highest under the moderate nitrogen treatment. The physiological characters in the seeds had the same trends as in the subtending leaves. Furthermore, the moderate nitrogen rate (240 kg ha−1) had a favorable yield and quality. Our results supported that a moderate nitrogen rate (240 kg ha−1) could coordinate the source–sink growth of cotton in the late stage, enhance the yield and fiber quality, and decrease the cost of fertilizer in the Yellow River Basin of China and other similar ecological areas.

Sign in / Sign up

Export Citation Format

Share Document