Spatio-temporal Bayesian network models with latent variables for revealing trophic dynamics and functional networks in fisheries ecology
详细信息 查看全文
- 作者：Neda Trifonovaa ; Neda.Trifonova@brunel.ac.uk" class="auth_mail" title="E-mail the corresponding author ; Andrew Kennyb ; David Maxwellb ; Daniel Dupliseac ; Jose Fernandesd ; Allan Tuckera
- 关键词：BN ; Bayesian network ; IBTS ; International Bottom Trawl Survey ; ICES ; International Council for the Exploration of the Sea ; CPUE ; catch per unit effort ; P ; pelagics ; SP ; small piscivorous ; LP ; large piscivorous and top predators ; Net PP ; net primary production ; DAG ; directed acyclic graph ; CPD ; conditional probability distribution ; CPT ; conditional probability table ; DBN ; dynamic Bayesian network ; HMM ; hidden Markov model ; ARHMM ; autoregressive hidden Markov model ; HV ; hidden variable ; EM ; Expect
- 刊名：Ecological Informatics
- 出版年：2015
- 出版时间：November 2015
- 年：2015
- 卷：30
- 期：Complete
- 页码：142-158
- 全文大小：2458 K
文摘
- •
A model that allows for two hidden variables and spatial autocorrelation is proposed.
- •
Accuracy in predicting species biomass varies within the spatially resolved areas.
- •
The specific hidden variable models spatial unmeasured effect.
- •
We found temporally and spatially differentiated functional networks.
- •
Combining structure learning from data and experts' knowledge in the model architecture is optimum.