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.
Nowadays cyanobacterial blooms in open reservoirs and estuaries became one of the most important ecological problem. The optimal way to solve this problem is to develop innovative methods for controlling the number of bloom-forming cyanobacteria based on weak external actions, which have no serious consequences for the entire ecological system. A novel efficient technique for in vivo estimation of cyanobacterial viability for online ecological monitoring of the results of weak external actions was elaborated by using a combination of different spectroscopic methods. It has been shown that the results obtained by means of conventional spectrophotometry and fluorimetry for cyanobacterial culture as a whole and the data obtained by fluorescent microscopic spectroscopy applied to a single cell are strictly related.
In the last few decades cyanbacteria became a valuable object of biotechnology. During their industrial cultivation growth rate, physiological state and algological purity of the culture should be controled permanently. One of the methods that can provide on-line monitoring of cyanobacterial cultures is a fluorescence spectroscopy.
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.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.