Employing linear regression in regression tree leaves
ECAI '92 Proceedings of the 10th European conference on Artificial intelligence
C4.5: programs for machine learning
C4.5: programs for machine learning
Machine Learning
Data mining: practical machine learning tools and techniques with Java implementations
Data mining: practical machine learning tools and techniques with Java implementations
Functional Models for Regression Tree Leaves
ICML '97 Proceedings of the Fourteenth International Conference on Machine Learning
SECRET: a scalable linear regression tree algorithm
Proceedings of the eighth ACM SIGKDD international conference on Knowledge discovery and data mining
Top-Down Induction of Model Trees with Regression and Splitting Nodes
IEEE Transactions on Pattern Analysis and Machine Intelligence
Lookahead and pathology in decision tree induction
IJCAI'95 Proceedings of the 14th international joint conference on Artificial intelligence - Volume 2
ReMauve: A Relational Model Tree Learner
Inductive Logic Programming
Stepwise Induction of Multi-target Model Trees
ECML '07 Proceedings of the 18th European conference on Machine Learning
EDLRT: Entropy-based dummy variables logistic regression tree
Intelligent Data Analysis
Multi-objective optimization with surrogate trees
Proceedings of the 15th annual conference on Genetic and evolutionary computation
Incremental linear model trees on massive datasets: keep it simple, keep it fast
Proceedings of the 28th Annual ACM Symposium on Applied Computing
| Hi-index | 0.00 |
The term "model trees" is commonly used for regression trees that contain some non-trivial model in their leaves. Popular implementations of model tree learners build trees with linear regression models in their leaves. They use reduction of variance as a heuristic for selecting tests during the tree construction process. In this article, we show that systems employing this heuristic may exhibit pathological behaviour in some quite simple cases. This is not visible in the predictive accuracy of the tree, but it reduces its explanatory power. We propose an alternative heuristic that yields equally accurate but simpler trees with better explanatory power, and this at little or no additional computational cost. The resulting model tree induction algorithm is experimentally evaluated and compared with simpler and more complex approaches on a variety of synthetic and real world data sets.