Downloads: 3
Research Paper | Computer Science | Volume 15 Issue 5, May 2026 | Pages: 241 - 245 | India
Exemplification of Local Regression at Leaf Nodes of Decision Tree
Abstract: CART algorithm is used to generate decision trees. CART algorithm is implemented in RStudio by means of RPART package. In a regression tree created using RPART, the average value of the dependent variable column for the observations located in each terminal node is shown. There may exist different patterns at different leaf nodes. This paper proposes to apply linear regression to the leaf nodes of the regression tree developed by RPART. The MSE (mean squared error) of all leaf nodes in regression tree obtained by RPART is higher as compared to the MSE (mean squared error) of applying local regression to RPART leaf nodes.
Keywords: Decision trees, Regression, Leaf nodes, Mean squared error, CART, rpart
How to Cite?: A. D. Mankar, S. D. Bhoite, K. G. Kharade, K. A. Raskar, "Exemplification of Local Regression at Leaf Nodes of Decision Tree", Volume 15 Issue 5, May 2026, International Journal of Science and Research (IJSR), Pages: 241-245, https://www.ijsr.net/getabstract.php?paperid=SR26503235151, DOI: https://dx.dx.doi.org/10.21275/SR26503235151