Statistical modeling of tree failures during storms

Kabir, Elnaz; Guikema, Seth D.; Kane, Brian

doi:10.1016/j.ress.2018.04.026

Cited by 48 publications

(51 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our results did not show improved predictive performance of the map with the more flexible 579 methods GAM and BRT compared to the logistic regression model (GLM). This is somewhat 580 surprising, especially in the case of BRTs, because several recent studies have shown good 581 performance of random forest for modelling storm disturbances (Albrecht et al, 2019;Hart et 582 al., 2019;Kabir et al, 2018). Yet, in our results BRT did not lead to better predictive 583 performance in cross-validation or with test data, even though it is a tree-based ensemble 584 method very similar to random forest.…”

Section: Comparison Of Methods 572mentioning

confidence: 99%

“…In addition, different approaches allowing more flexible model behaviour 81 than fully parametric GLMs have been used, such as generalized additive models (GAM; 82 Schmidt et al, 2010) that use non-parametric smooth functions to allow more flexibility in the 83 relationship of response variable and predictors (Hastie et al, 2009). Machine learning 84 approaches have also been successfully applied to wind disturbance modeling (see 85 Hanewinkel et al 2004 for an early example) and recently especially tree-based ensemble 86 models, such as random forests, have been shown to perform well in predicting wind 87 damage (Albrecht et al, 2019;Hart et al, 2019;Kabir et al, 2018;Schindler et al, 2016). 88…”

mentioning

confidence: 99%

See 1 more Smart Citation

High-resolution mapping of forest vulnerability to wind for disturbance-aware forestry and climate change adaptation

Suvanto

Peltoniemi

Tuominen

et al. 2019

Preprint

View full text Add to dashboard Cite

10Windstorms cause major disturbances in European forests and forest management can play 11 a key role in making forests more persistent to disturbances. However, better information is 12 needed to support decision making that effectively accounts for wind disturbances. Here we 13show how empirical probability models of wind damage, combined with existing spatial 14 datasets, can be used to provide fine-scale spatial information about disturbance probability 15 over large areas. First, we created stand-level damage probability models with predictors 16 describing forest characteristics, recent forest management history and local wind, soil, site 17and climate conditions. We tested three different methods for creating the damage 18 probability models -generalized linear models (GLM), generalized additive models (GAM) 19 and boosted regression trees (BRT). Then, the damage probability maps were calculated by 20 combining the models (GLM, GAM and BRT) with GIS data sets representing the model 21predictors. Finally, we demonstrated the predictive performance of the maps with a large, 22 high spatial resolution (16 x 16 m 2 raster resolution), making it useful in assessing the 28 vulnerability of individual forest stands when making management decisions. The map is 29 also a powerful tool for communicating disturbance risks to forest owners and managers and 30 it has the potential to steer forest management practices to a more disturbance aware 31 direction. Our study showed that in spite of the inherent stochasticity of the wind and 32 damage phenomena at all spatial scales, it can be modelled with good accuracy across 33 large spatial scales when existing ground and earth observation data sources are combined 34 smartly. With improving data quality and availability, map-based risk assessments can be 35 extended to other regions and other disturbance types. 36

show abstract

Section: Comparison Of Methods 572mentioning

confidence: 99%

mentioning

confidence: 99%

High-resolution mapping of forest vulnerability to wind for disturbance-aware forestry and climate change adaptation

Suvanto

Peltoniemi

Tuominen

et al. 2019

Preprint

View full text Add to dashboard Cite

show abstract

“…Based on the comparative analysis results, the proposed square wave-activated WASDNN is more satisfactory than the WASDNN activated by other activation functions and other neural networks. In addition to these models, the proposed model is compared with other classification methods, such as logistic regression and AdaBoost [28]. However, the results are similarly not satisfying and convincing.…”

Section: Analysis On Comparisonsmentioning

confidence: 99%

“…Table 7 lists the data and the corresponding binarizations. By using the C4.5 decision tree [28], the classification results under different attributes are shown in Figure 16, and the information entropy gains of the attributes are calculated and displayed as ''entropy'' in the corresponding nodes. One can see from the figure that using these three factors to classify the presidential parties would generate errors, whereas classifying the results with the proposed WASDNN's predictions would simplify the tree complexity and generate the predictions without errors.…”

Section: Appendix B About Election Prediction Factorsmentioning

confidence: 99%

Predictions of USA Presidential Parties From 2021 to 2037 Using Historical Data Through Square Wave-Activated WASD Neural Network

Zeng

Zhang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

The United States of America (USA) has been the most powerful country in the world for decades. The country's global leadership and power are critical for international security and intercountry relationships, in which the political events yield differences. With the deepening development of globalization and the recovery of the global economy, the future policies and presidential parties of the USA should be predicted to avoid potential crises. However, making unbiased and robust predictions is challenging. This study presents a weight-and-structure-determination (WASD) algorithm-based feedforward neural network activated by a set of square wave functions to predict the presidential parties in the USA from 2021 to 2037. The historical data of presidential parties from 1853 to 2017 are used as the training set in the numerical experiments, and the results show that the proposed neural network can predict 9 terms of future presidential parties. Compared with other models, the proposed model is more efficient, robust, purely time-driven, and bias-independent. Through the square wave-activated WASD neural network, the Democratic Party is predicted to win in 2025, 2029, and 2033, whereas the Republican Party might win in 2021 and 2037. INDEX TERMS Weight-and-structure-determination algorithm, feedforward neural network, USA, presidential parties, prediction.

show abstract

“…It is mainly divided into shallow network model and deep neural network model. The shallow model mainly includes random forest [11], Bayesian model, ensemble model [12] and artificial neural network, etc, literature [13] proposed a driving risk status prediction method based on Bayesian network prediction model; in literature [14], neural network model is used to predict the nitrogen status of wheat plants. Compared with the conventional gray-world and scale-by-max approaches, the performance of neural network model is better.…”

Section: Introductionmentioning

confidence: 99%

Genetic Optimization Method of Pantograph and Catenary Comprehensive Monitor Status Prediction Model Based on Adadelta Deep Neural Network

Yuan

Chi

et al. 2019

IEEE Access

View full text Add to dashboard Cite

The status of the Pantograph and Catenary is the guarantee for the safe operation of the railway. However, the traditional Pantograph and Catenary status judgment efficiency is not satisfactory, and it is impossible to timely repair the catenary, which may lead to the greater economic loss. In this paper, a new GA-ADNN-based (genetic algorithm-Adadelta deep neural network-based) optimization method for the prediction model for catenary comprehensive pantograph and catenary monitor (CPCM) status is proposed. According to the status values of the CPCM parameters such as height, stagger, hard point, contact force, and height difference within span, the status value of the pillars in the catenary has been calculated by the analytic hierarchy process, and then the prediction model for predicting catenary CPCM status has been established and then optimized by genetic algorithm to avoid prediction model falling into local optimum. Finally, the CPCM test parameters of each pillar of the catenary in the actual example are input and the CPCM status value of the corresponding pillar is predicted. With the smallest prediction error found, the genetic algorithm is used for optimization, the optimal learning rate of the prediction model is 0.0559, and the optimal number of the hidden layer of the CPCM status prediction model is determined to be 14. The experimental results show the feasibility of GA-ADNN-based prediction model for predicting the catenary CPCM status, and that compared with the support vector machine and traditional artificial neural network prediction methods, the GA-ADNN-based prediction model has higher prediction precision and better generalization ability.

show abstract

Statistical modeling of tree failures during storms

Cited by 48 publications

References 38 publications

High-resolution mapping of forest vulnerability to wind for disturbance-aware forestry and climate change adaptation

High-resolution mapping of forest vulnerability to wind for disturbance-aware forestry and climate change adaptation

Predictions of USA Presidential Parties From 2021 to 2037 Using Historical Data Through Square Wave-Activated WASD Neural Network

Genetic Optimization Method of Pantograph and Catenary Comprehensive Monitor Status Prediction Model Based on Adadelta Deep Neural Network

Contact Info

Product

Resources

About