Web proceedings papers


Andreja Naumoski and Kosta Mitreski


The challenge to discover knowledge from environmental data that has led to usage of methods and techniques such as data mining tools, can bridge the knowledge gap between the biological experts and organisms. This research aimed to assess relationships between the diatoms and the indicators of the environment with Na?ve Bayes method. Diatoms are ideal indicators of certain physical-chemical parameters and they can be classified into one of the water quality classes (WQCs). The classification models are induced by using Na?ve Bayes technique. The input dataset that is supplied for the na?ve Bayes method is discretised. Based on the evaluation results, several models are presented and discussed. The obtain results from the models are verified with existing diatom ecological preference and for some diatoms new knowledge is added. To best of our knowledge, this is the first time the prosed method to be applied for diatom classification of any ecosystem.


Naïve Bayes, diatom classification, indicators, water ecosystem