Transactions on Machine Learning and Data MiningP-ISSN: 1865-6781, E-ISSN 2509-9337 ISBN 978-3-942952-72-9
Volume 12 - Number 2 - October 2019 - Pages 55-74
Evaluating Economic Performance with Soft Regression
Moti Schneider1, Arthur Yosef2, Eli Shnaider3
1Netanya Academic College, Israel
2Tel Aviv-Yaffo Academic College, Israel
3Peres Academic Center, Israel
Abstract
This study demonstrates effective data mining tool under severe limitations of data availability. We present a soft computing method for evaluating economic performance. To avoid computational explosion, we utilize intervals. This will reduce the number attributes in the dataset. Utilizing intervals allows us to overcome difficult modeling problems such as large quantity of missing data, substantial outliers, etc. Finally, case study of evaluating economic performance of the Soviet led East European bloc is presented. In spite of highly unreliable and inaccurate data provided by the officials of the bloc, the method presented here allows to reach solid and reliable conclusions.
Keywords: Data Mining, Soft Computing, Cross-national model, Soft Regression, fuzzy logic
Download Paper (379 KB)