Choosing Text Description Language Features for Training and Testing Adaptive Resonance Theory: A Case Study for Botanical Classification

The practical methodologies of stable pattern classification using artificial intelligence as advisory tools are researched here according to studies in the flowering plant genera Lithops N. E. Br. (Aizoaceae). In this paper, the use of a neural network model with the adaptive resonance theory is a practical generation of groups as a classifier for botanical taxa. In order to provide comparisons for this study of effective classification performance, the study here in the succulent plant genus Lithops involved the classification of 87 records that comprise about 35 species. It is demonstrated that the proposed system using artificial neural networks technique with statistical property of grouping method can achieve a classification rate of 85.71% separated records into 35 groups referred to the traditional plant taxonomic groups.