Biological Monitoring: a Comparison between Bayesian, Neural and Machine Learning Methods of Water Quality Classification

Walley, William J.; Džeroski, Sašo

doi:10.1007/978-0-387-34951-0_20

Cited by 21 publications

(16 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When compared to other modelling methodologies for running water bodies, such as Artificial Intelligence (Walley and Džeroski, 1995;Džeroski et al, 1997Džeroski et al, , 2000Walley et al, 1998;Walley and Fontama, 2000;Broekhoven et al, 2006), the StDM is more intuitive, namely in mathematical terms, providing easy explanations for the underlying relations between independent and dependent variables and because is based on conventional linear methods that allowed a more direct development of testable hypotheses (Manel et al, 1999).…”

Section: Discussionmentioning

confidence: 99%

Applying a stochastic-dynamic methodology (StDM) to facilitate ecological monitoring of running waters, using selected trophic and taxonomic metrics as state variables

Cabecinha

Silva-Santos

Cortes

et al. 2007

Ecological Modelling

View full text Add to dashboard Cite

Ecological indicatorsBenthic macroinvertebrates Biological metricsStochastic-dynamic methodology a b s t r a c tAs an improvement of a previous work [Cabecinha, E., Cortes, R., Cabral, J.A., 2004. Performance of a stochastic-dynamic modelling methodology for running waters ecological assessment. Ecol. Modell. 175,[303][304][305][306][307][308][309][310][311][312][313][314][315][316][317], the present paper examined the applicability of a holistic stochastic-dynamic methodology (StDM) in predicting the tendencies of benthic macroinvertebrate metrics from mountain streams facing expected scenarios either: (1) of pollution increase due to the agricultural intensification; or (2) of farming activity abandonment becoming less pollutant as a non-point source. The StDM is a sequential modelling process developed in order to predict the ecological status of changed ecosystems, from which management strategies can be designed. These procedures focus on the interactions between conceptually isolated key-components, such as some relevant trophic and taxonomic metrics and changes in local environmental conditions. The dataset recorded from the field included true gradients of environmental changes. The samples of aquatic macroinvertebrate, environmental and physical-chemical data were collected from four watersheds of mountain rivers in Northeast Portugal, between 1983 and 1985. The dynamic model developed was preceded by a conventional multivariate statistical procedure performed to discriminate the significant relationships between the selected components of the studied watersheds. The model validation was based on independent data from a watershed not included in the model construction. Overall, the simulation results are encouraging since they seem to demonstrate the StDM reliability in capturing the stochastic environmental dynamics of the studied aquatic ecosystems facing agricultural scenarios that will characterize the region, namely by predicting credible behavioural patterns for the selected metrics.

show abstract

Section: Discussionmentioning

confidence: 99%

Applying a stochastic-dynamic methodology (StDM) to facilitate ecological monitoring of running waters, using selected trophic and taxonomic metrics as state variables

Cabecinha

Silva-Santos

Cortes

et al. 2007

Ecological Modelling

View full text Add to dashboard Cite

show abstract

“…It has to be mentioned however that a large inherent uncertainty is present in these predictions. Other studies on the contrary have found BBNs to perform well as predictive models (Walley & Dzeroski 1995, Trigg et al 2000, Fleishman et al 2001). However, in many of these studies, the general evaluation was based on comparing the class with the highest predicted probability value to the measured class, ignoring the information inherent to the prediction outcome as expressed by probabilities or no rigorous validation was done at all (Fleishman et 2002).…”

Section: Discussionmentioning

confidence: 99%

Application of Bayesian Belief Networks for the prediction of macroinvertebrate taxa in rivers

Adriaenssens

Goethals

Charles

et al. 2004

Ann. Limnol. - Int. J. Lim.

View full text Add to dashboard Cite

Integrated ecological models are of great potential as predictive tools in decision support of river management. Such models need to be transparent and consistent with the existing expert knowledge, and give the river manager adequate information regarding their inherent uncertainty. One way to fulfil these needs is through the use of Bayesian Belief Networks (BBNs). Such networks represent cause-and-effect assumptions between system variables in a graphical structure. To establish the potential of Bayesian Belief Networks in river management, a small-scale study was performed with the aim of assessing the success of prediction of macroinvertebrate taxa in rivers by means of a selected number of environmental variables. Gammaridae and Asellidae were chosen because of their high relative abundances in small and large brooks in contrast to other macroinvertebrate taxa. Based on one-layered BBN networks, the predictive capacity of the models was assessed by means of the number of Correctly Classified Instances (CCI) and Cohen's Kappa (K). The performance of these models was moderate to good for presence/absence classifications but showed a low to moderate performance when predicting abundance classes. When extending the former BBN network to a two-layered one, enhancing the number of links and variables, no obvious improvement in model performance was detected. The results indicate that thoughtful input variable selection as well as sensitivity analysis will improve the models for practical use in river restoration management.

show abstract

“…In comparison with other predictive modelling techniques such as ANN and fuzzy logic, BBN networks showed a relative good predictive success based on only three input variables (Goethals, 2005). Other studies have found BBNs to perform well as predictive models (Fleishman et al, 2001;Landuyt et al, 2015;Walley and Dzeroski, 1995). However, in many of these studies, rigorous validation was not done or did not receive enough attention (Fleishman et al, 2002).…”

Section: Applicability Of Bbn Modelling For Decision Support In Rivermentioning

confidence: 93%

Bayesian belief network models to analyse and predict ecological water quality in rivers

Forio

Landuyt

Bennetsen

et al. 2015

Ecological Modelling

View full text Add to dashboard Cite

Biological Monitoring: a Comparison between Bayesian, Neural and Machine Learning Methods of Water Quality Classification

Cited by 21 publications

References 1 publication

Applying a stochastic-dynamic methodology (StDM) to facilitate ecological monitoring of running waters, using selected trophic and taxonomic metrics as state variables

Applying a stochastic-dynamic methodology (StDM) to facilitate ecological monitoring of running waters, using selected trophic and taxonomic metrics as state variables

Application of Bayesian Belief Networks for the prediction of macroinvertebrate taxa in rivers

Bayesian belief network models to analyse and predict ecological water quality in rivers

Contact Info

Product

Resources

About