Studies on CART’s Performance in Rule Induction and Comparisons by STRIM
العنوان: | Studies on CART’s Performance in Rule Induction and Comparisons by STRIM |
---|---|
المؤلفون: | Tetsuro Saeki, Shoya Kawaguchi, Yuichi Kato |
المصدر: | Rough Sets ISBN: 9783319993676 IJCSR |
بيانات النشر: | Springer International Publishing, 2018. |
سنة النشر: | 2018 |
مصطلحات موضوعية: | Cart, Test data generation, Computer science, Rule induction, 02 engineering and technology, computer.software_genre, 01 natural sciences, Regression, 010104 statistics & probability, Variable (computer science), Tree structure, Product (mathematics), 0202 electrical engineering, electronic engineering, information engineering, 020201 artificial intelligence & image processing, Data mining, 0101 mathematics, computer, Statistical hypothesis testing |
الوصف: | The tree based method is a conventional statistical method that involves constructing a tree structure for a classification model through recursively splitting a dataset by explanatory variables to minimize some impurity criteria for the response variable. This tree structure induces many if-then rules with product forms. In this paper, we study a basic tree based approach — the classification and regression trees (CART) method — based on a simulation model for data generation and verification for induced rules. We compare CART with the statistical test rule induction method (STRIM) to clarify its performance and problems. We also apply both methods to a real-world dataset and consider their performances based on the simulation results. |
ردمك: | 978-3-319-99367-6 |
URL الوصول: | https://explore.openaire.eu/search/publication?articleId=doi_________::b04a518b48cd18f6ac958cd33571baae https://doi.org/10.1007/978-3-319-99368-3_12 |
حقوق: | CLOSED |
رقم الأكسشن: | edsair.doi...........b04a518b48cd18f6ac958cd33571baae |
قاعدة البيانات: | OpenAIRE |
ردمك: | 9783319993676 |
---|