Abstract

Abstract
This paper presents a system for induction of forest of functional trees from data streams able to detect concept drift. The Ultra Fast Forest of Trees (UFFT) is an incremental algorithm, which works online, processing each example in constant time, and performing a single scan over the training examples. It uses analytical techniques to choose the splitting criteria, and the information gain to estimate the merit of each possible splitting-test. For multi-class problems the algorithm builds a binary tree for each possible pair of classes, leading to a forest of trees. Decision nodes and leaves contain naive-Bayes classifiers playing different roles during the induction process. Naive-Bayes in leaves are used to classify test examples. Naive-Bayes in inner nodes play two different roles. They can be used as multivariate splitting-tests if chosen by the splitting criteria, and used to detect changes in the class-distribution of the examples that traverse the node. When a change in the class-distribution is detected, all the sub-tree rooted at that node will be pruned. The use of naive-Bayes classifiers at leaves to classify test examples, the use of splitting-tests based on the outcome of naive-Bayes, and the use of naive-Bayes classifiers at decision nodes to detect changes in the distribution of the examples are directly obtained from the sufficient statistics required to compute the splitting criteria, without no additional computations. This aspect is a main advantage in the context of high-speed data streams. This methodology was tested with artificial and real-world data sets. The experimental results show a very good performance in comparison to a batch decision tree learner, and high capacity to detect drift in the distribution of the examples.

Abstract

Abstract

Abstract
Discretization of continuous attributes is an important task for certain types of machine learning algorithms. Bayesian approaches, for instance, require assumptions about data distributions. Decision Trees on the other hand, require sorting operations to deal with continuous attributes, which largely increase learning times. This paper presents a new method of discretization, whose main characteristic is that it takes into account interdependencies between attributes. Detecting interdependencies can be seen as discovering redundant attributes. This means that our method performs attribute selection as a side effect of the discretization. Empirical evaluation on five benchmark datasets from UCI repository, using C4.5 and a naive Bayes, shows a consistent reduction of the features without loss of generalization accuracy.

Abstract