International Journal of Academic and Applied Research (IJAAR)   Year: 2020 | Volume: 4 | Issue: 4 | Page No.: 1-7 Determination of the Properties That are Effective in Determining the Edible State of Mushroom (Agaricus and Lepiota) by Using Datamining Methods Oznur Isci Guneri, Burcu Durmus Abstract: The aim of this study is to determine the mushroom characteristics that are effective in classification with the help of decision trees by identifying data mining algorithms that can classify mushrooms according to their edibility status. In the scope of classification analysis, 22 different characteristics of mushrooms were examined. Classification analysis was performed with different decision tree algorithms and decision tree structures were examined. All of the algorithms discussed were very successful according to correct classification rate, Kappa statistics and root mean square error statistics. On the other hand, when the decision tree structures of the algorithms were examined, it was seen that the classification structures highlight different fungal properties and various comparisons were made in this direction. As a result of the study; the decision tree algorithms were found to be successful in detecting edibility in two types of mushroom families and the 'odor' attribute was found to be more distinctive than the other characteristics in the determination of edibility. In addition, it was found that 7 out of 22 attributes did not play any role in distinguishing edible mushroom.