Application of fuzzy C-Mean clustering base tree for measuring the effectiveness of corporate

• Main Authors: Teyakome J., Eiamkanitchat N., Suriya K., Napook P.
• Format: Book Series
• Published: 2017
• Online Access: https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=84959107880&origin=inward; http://cmuir.cmu.ac.th/jspui/handle/6653943832/42197
• Institution: Chiang Mai University

© Springer Science+Business Media Singapore 2016. The operations of the companies often have many different types of indicators to measure the performance. In this work the 6 standard criteria are used, including current ratio, equity debt, return on assets (ROA), return on equity (ROE), net profit and return on investment (ROI). With the consideration of generalization, the objective function is the maximum sum of all criteria except the equity debt. This paper proposes a Fuzzy C-Mean Clustering Base Tree (FCMT) method for measuring the effectiveness of corporate in Thailand. The 6 standard criteria calculated from annual report are used for data set creation. The Fuzzy C-means algorithm is used to analyze the 982 companies and clustered into 3 clusters, including “excellent”, “good” and “fair” performance. In order to verify the correctness of clustering methodology 4 standard datasets from the UCI machine learning repository are used in the experiment. The results are trained by the decision tree algorithm to construct the classification tree. The experimental results show the 97.05 percentages of classification accuracy of the decision tree. The rules extracted from the decision tree not only can use as classification rules, the addition benefit is it can use as the guidelines to raise the effectiveness of corporations in the future.