Modeling a System for Decision Support in Snow Avalanche Warning Using Balanced Random Forest and Weighted Random Forest

Möhle, Sibylle; Bründl, Michael; Beierle, Christoph

doi:10.1007/978-3-319-10554-3_8

Cited by 7 publications

(8 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As explained before, all these methods were combined with the same external balancing scheme (Appendix A.3). However, for comprehensiveness and consistency of the demonstration, an internal balancing approach, namely the weighted random forests (wRF, Chen et al, 2004) already used for avalanche forecasting (Möhle et al, 2014), was also implemented. The different classifiers were implemented on transformed avalanche drivers which are the first principal components responsible, in each massif, for at least 95% of the total variability in the meteorological and snow variables (Appendix A.2).…”

Section: Competing Methodsmentioning

confidence: 99%

“…Internal approaches are algorithmically explicit, meaning that one creates innovative classification algorithms (or at least adapts existing ones) in order to take the class unbalanced problem into consideration. For example, an internal approach is used in Möhle et al (2014) where avalanche forecasting is attempted using a modified class-balancing random forest algorithm. On the contrary, in external approaches, one pre-processes the data in order to equilibrate the size of each class without modifying the classification algorithm.…”

Section: Class Balancingmentioning

confidence: 99%

“…Further refinements such as inclusion of expert knowledge (Schweizer and Föhn, 1996) or explicit consideration of the topography into the analysis (McCollister et al, 2003) have been proposed, without modifying the spirit of the methodology. More recently, non-linear classification methods were introduced, notably classification trees (Davis et al, 1999;Hendrikx et al, 2005;Baggi and Schweizer, 2009;Hendrikx et al, 2014), random forestsalso called forest decision trees - (Mitterer and Schweizer, 2013;Möhle et al, 2014;Marienthal et al, 2015;Dreier et al, 2016), and support vector machines (Pozdnoukhov et al, 2008(Pozdnoukhov et al, , 2011.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Combining random forests and class-balancing to discriminate between three classes of avalanche activity in the French Alps

Sielenou

Viallon-Galinier

Hagenmuller

et al. 2021

Cold Regions Science and Technology

View full text Add to dashboard Cite

Section: Competing Methodsmentioning

confidence: 99%

Section: Class Balancingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Combining random forests and class-balancing to discriminate between three classes of avalanche activity in the French Alps

Sielenou

Viallon-Galinier

Hagenmuller

et al. 2021

Cold Regions Science and Technology

View full text Add to dashboard Cite

“…The random forest approach is robust to both outliers and noise multiple collinearities problems [ 51 ]. Therefore, random forest has a good performance in multivariate prediction and its interpretation, so that it has been widely used in many fields such as medicine, biology, and so on [ 52 ]. Actually, RF is almost similar to a black box.…”

Section: Methodsmentioning

confidence: 99%

Assessing and Predicting the Water Resources Vulnerability under Various Climate-Change Scenarios: A Case Study of Huang-Huai-Hai River Basin, China

Chen

Feng

Zhang

et al. 2020

Entropy

View full text Add to dashboard Cite

The Huang-Huai-Hai River Basin plays an important strategic role in China’s economic development, but severe water resources problems restrict the development of the three basins. Most of the existing research is focused on the trends of single hydrological and meteorological indicators. However, there is a lack of research on the cause analysis and scenario prediction of water resources vulnerability (WRV) in the three basins, which is the very important foundation for the management of water resources. First of all, based on the analysis of the causes of water resources vulnerability, this article set up the evaluation index system of water resource vulnerability from three aspects: water quantity, water quality and disaster. Then, we use the Improved Blind Deletion Rough Set (IBDRS) method to reduce the dimension of the index system, and we reduce the original 24 indexes to 12 evaluation indexes. Third, by comparing the accuracy of random forest (RF) and artificial neural network (ANN) models, we use the RF model with high fitting accuracy as the evaluation and prediction model. Finally, we use 12 evaluation indexes and an RF model to analyze the trend and causes of water resources vulnerability in three basins during 2000–2015, and further predict the scenarios in 2020 and 2030. The results show that the vulnerability level of water resources in the three basins has been improved during 2000–2015, and the three river basins should follow the development of scenario 1 to ensure the safety of water resources. The research proved that the combination of IBDRS and an RF model is a very effective method to evaluate and forecast the vulnerability of water resources in the Huang-Huai-Hai River Basin.

show abstract

“…A major challenge when developing as well as verifying statistical models, and avalanche forecasts in general, is the lack of a measurable target variable. Since avalanche occurrence seems a logical target variable, most of the previous approaches focused on the estimation of avalanche activity using typical machine learning methods such as classification trees (Davis et al, 1999;Hendrikx et al, 2014;Baggi and Schweizer, 2009), nearest neighbors (Purves et al, 2003), support vector machines (Pozdnoukhov et al, 2008(Pozdnoukhov et al, , 2011 and random forests (Mitterer and Schweizer, 2013;Möhle et al, 2014;Dreier et al, 2016;Dkengne Sielenou et al, 2021). To build and validate these models, a substantial amount of avalanche data is required.…”

Section: Introductionmentioning

confidence: 99%

Data-driven automated predictions of the avalanche danger level for dry-snow conditions in Switzerland

Pérez-Guillén¹,

Techel²,

Hendrick³

et al. 2021

Preprint

View full text Add to dashboard Cite

Abstract. Even today, the assessment of avalanche danger is by large a subjective, yet data-based decision-making process. Human experts analyze heterogeneous data volumes, diverse in scale, and conclude on the avalanche scenario based on their experience. Nowadays, modern machine learning methods and the rise in computing power in combination with physical snow cover modelling open up new possibilities for developing decision support tools for operational avalanche forecasting. Therefore, we developed a fully data-driven approach to predict the regional avalanche danger level, the key component in public avalanche forecasts, for dry-snow conditions in the Swiss Alps. Using a large data set of more than 20 years of meteorological data measured by a network of automated weather stations, which are located at the elevation of potential avalanche starting zones, and snow cover simulations driven with these input weather data, we trained two random forest (RF) classifiers. The first classifier (RF #1) was trained relying on the forecast danger levels published in the avalanche bulletin. Given the uncertainty related to a forecast danger level as a target variable, we trained a second classifier (RF #2), relying on a quality-controlled subset of danger level labels. We optimized the RF classifiers by selecting the best set of input features combining meteorological variables and features extracted from the simulated profiles. The accuracy of the danger level predictions ranged between 74 % and 76 % for RF #1, and between 72 % and 78 % for RF #2, with both models achieving better performance than previously developed methods. To assess the accuracy of the forecast, and thus the quality of our labels, we relied on nowcast assessments of avalanche danger by well-trained observers. The performance of both models was similar to the accuracy of the current experience-based Swiss avalanche forecasts (which is estimated to 76 %). The models performed consistently well throughout the Swiss Alps, thus in different climatic regions, albeit with some regional differences. A prototype model with the RF classifiers was already tested in a semi-operational setting by the Swiss avalanche warning service during the winter 2020-2021. The promising results suggest that the model may well have potential to become a valuable, supplementary decision support tool for avalanche forecasters when assessing avalanche hazard.

show abstract

Modeling a System for Decision Support in Snow Avalanche Warning Using Balanced Random Forest and Weighted Random Forest

Cited by 7 publications

References 10 publications

Combining random forests and class-balancing to discriminate between three classes of avalanche activity in the French Alps

Combining random forests and class-balancing to discriminate between three classes of avalanche activity in the French Alps

Assessing and Predicting the Water Resources Vulnerability under Various Climate-Change Scenarios: A Case Study of Huang-Huai-Hai River Basin, China

Data-driven automated predictions of the avalanche danger level for dry-snow conditions in Switzerland

Contact Info

Product

Resources

About