Metrics for evaluating performance and uncertainty of Bayesian network models

Marcot, Bruce G.

doi:10.1016/j.ecolmodel.2012.01.013

Cited by 298 publications

(280 citation statements)

References 35 publications

Supporting

Mentioning

278

Contrasting

Unclassified

Order By: Relevance

“…The variance reduction estimates the impact of a change in the state of a proxy variable on the state of the target variable [51]. The variance reduction values range from 0 to 100%, where a higher value indicates a higher influence [18,49].…”

Section: Sensitivity Analysis and Model Validationmentioning

confidence: 99%

See 1 more Smart Citation

Inferring Missing Climate Data for Agricultural Planning Using Bayesian Networks

Lara-Estrada

Rasche

Sucar

et al. 2018

Land

View full text Add to dashboard Cite

Abstract:Climate data availability plays a key role in development processes of policies, services, and planning in the agricultural sector. However, data at the spatial or temporal resolution required is often lacking, or certain values are missing. In this work, we propose to use a Bayesian network approach to generate data for missing variables. As a case study, we use relative humidity, which is an important indicator of land suitability for coffee production. For the model, we first extracted climate data for the variables precipitation, maximum and minimum air temperature, wind speed, solar radiation and relative humidity from the surface reanalysis dataset Climate Forecast System Reanalysis. We then used machine learning algorithms to define the model structure and parameters from the relationships of the variables found in the dataset. Precipitation, maximum and minimum air temperature, wind speed, and solar radiation are then used as proxy variables to infer missing values for monthly relative humidity and relative humidity for the driest month. For this, we used both complete and incomplete initial data. In both scenarios of data availability, the comparison of estimated and measured values of relative humidity shows a high level of agreement. We conclude that using Bayesian Networks is a practical solution to estimate relative humidity for coffee agricultural planning.

show abstract

Section: Sensitivity Analysis and Model Validationmentioning

confidence: 99%

“…3 The Spherical Payoff is a scoring metric used to test the performance of Bayesian network models. The score goes from 0 to 1, where 1 indicates the best performance [51]. …”

Section: Model Structure and Parametersmentioning

confidence: 99%

Inferring Missing Climate Data for Agricultural Planning Using Bayesian Networks

Lara-Estrada

Rasche

Sucar

et al. 2018

Land

View full text Add to dashboard Cite

show abstract

“…We described models with Nagelkerke's R 2 (Nagelkerke 1991) and area under the receiver operating characteristic curve (AUC). AUC values 0.5 indicate models which provide more incorrect than correct predictions, whereas a value of 1.0 indicates no error in the predictions (Marcot 2012).…”

Section: Animal Use Of Compacted Snow: Field Methodsmentioning

confidence: 99%

Footload influences wildlife use of compacted trails in the snow

Whiteman

Buskirk

2013

Wildlife Biology

View full text Add to dashboard Cite

Animals moving across snow surfaces sink to varying depths, increasing the energetic cost of travel. For ease of movement, animals may follow compacted trails created by sports such as snowmobiling and snowshoeing. We tested the assumption that animals less‐adapted to snow travel (i.e. animals with a high footload (body mass/foot surface area)) are more likely to use compacted trails and follow them for greater distances than animals well‐adapted to moving on snow. We sampled animal movements on compacted and non‐compacted transects in northwestern Wyoming and southeastern Idaho, USA, during two winters. Consistent with our prediction, footload positively influenced the probability of animals following compacted transects, and positively influenced their following distance, although the latter relationship was weaker. Additionally, we found that different snow sports cause similar increases in snow density. We suggest that future researchers experimentally test whether the use of compacted trails by animals with high footloads alters their habitat selection and dispersal, and whether it affects their competitive and predatory success.

show abstract

“…It shows how the changes and uncertainties in the network structure function when data about the nodes in network structure is entered. Sensitivity analysis reveals which variable in the model has the biggest effect on the target node, and the variables are ranked in terms of their effects [30]. In other words, Here, the positive and negative effects of parent nodes on the child node (%100) were initially tested separately.…”

Section: Sensitivity Analysismentioning

confidence: 99%