Yapay Sinir Ağları (Ysa) Kullanılarak Taban Büyük Omurgasızlarının Modellemesi
Abstract
Models are useful to predict communities in watercourses based on the abiotic characteristics of their aquatic environment. Back-propagation Artificial Neural Networks (ANN) were tested with the aim of modelling the occurrences of benthic macroinvertebrate taxa in Yuvarlakçay and Namnam Streams which both inflow to Köyceğiz Lake in Muğla, Turkey. For that purpose ANN algorithms were used to induce predictive models on a dataset. This dataset consisted of 98 samples, collected over a year period during 1992- 1993. Fourteen environmental variables are measured at each site (total 8), as well as dominance values of the benthic macroinvertebrate taxa (Baetis, Caenis, Ephemerella, Onychogomphus, Rhithrogena).Initially different neural networks were tried to develop and optimize the best model configuration which ensure the prediction of the habitat suitability of each macroinvertebrate taxa.