Application of artificial neural network into the freshwater fish caught in Turkey

Abstract

Artificial neural networks (ANNs) are computational intelligence techniques, which are used in many applications, such as forecast. The aim of this study was to evaluate artificial neural networks created for freshwater fish caught in Turkey between the years of 2003 to 2012. As a decision system, ANNs are an important tool for forecast in fisheries. A feedforward neural network was selected, with two layers, sigmoid functions, and adaption learning function for the training of the ANNs. The results of the application of created neural networks for fishery products of forecast based on test cases validated by MAPE. Estimates of 2015, data was found to be 15147.6 tons. Freshwater fish caught in Turkey is forecasted in the next years. This result shows us that, in the coming year’s aquaculture and freshwater products are alarming. However, fish catch data are not regularly collected because of lack of fish catch information collected by officials.