In this article a feedforward error backpropagation artificial neural network is investigated and the analysis of its illogical behaviour is presented. The problem of illogical behavior arises in various models of artificial neural networks. In the presented work a classifying artificial neural network (CANN) is considered and several learning algorithms were implemented and compared. CANN was designed for automatic differentiaition of cyanobacterial strains during environmental monitoring and some of trained networks demonstrated illogical behavior in further testing. Several original techniques were elaborated for estimation of the quality and accuracy of classification in addition to the traditional ones. Novel visualization methods were suggested for classification and generalization results representation.

Abundance and biological diversity of phytoplankton communities, investigated in this work, are often used as a marker for the determination of environmental health and fresh water quality. Presently their routine analysis is very time consuming and expensive. A lot of articles are devoted to the development of a system for an in situ automated analysis of phytoplankton properties. However, the applied problems of biology, ecology and, in particular, algology usually are associated with some difficulties due to shortage and/or fuzziness of the experimental data. Hence, while using neural network modelling a set of specific problems can occur. In this article on the base of experimental data several such problems are presented and possible solutions are suggested. In particular, the illogical behavior of classifying neural network is revealed, while studying the biological diversity of cyanobacteria, and the original technique for results validation is presented. This problem is investigated on a set of spectroscopic data, recorded by means of confocal laser scanning microscopy. The generalization quality of the trained model is studied as the main learning parameter. Another problem of shortage dataset is examined in the frames of regression model for bioplankton abundance. This problem is solved by means of feed-forward back-propagation neural networks with two hidden layers. The modelling was carried out on a small experimental selection (only 39 observations were available), despite this, the relatively high determination coefficient was obtained for the training and test samples, while using dropout layout.

No abstract

In this article a neural-network regression model for prediction of the bacterioplankton abundance according to physicochemical parameters of the environmental conditions is considered and some of the peculiarities of its development are described. A particular case of small and very heterogeneous data sample, typical for biological applications, is analysed. To solve this problem, a number of multi-layer feed-forward neural networks with different architectures are studied. The regression results are estimated on the base of determination coefficient and standard deviation of predicted values in the test sample. The effect of the dropout, applied to one of the hidden layers, on learning process and obtained results is analysed.

No abstract