In the feature subset selection problem, a learning algorithm is faced with the problem of selecting a relevant subset of features upon which to focus its attention, while ignoring the rest. To achieve the best possible performance with a particular learning algorithm on a particular training set, a feature subset selection method should consider how the algorithm and the training set interact. We explore the relation between optimal feature subset selection and relevance. Our wrapper method searches for an optimal feature subset tailored to a particular algorithm and a domain. We study the strengths and weaknesses of the wrapper approach and show a series of improved designs. We compare the wrapper approach to induction without feature subset selection and to Relief, a filter approach to feature subset selection. Significant improvement in accuracy is achieved for some datasets for the two families of induction algorithms used: decision trees and

This paper introduces ICET, a new algorithm for cost-sensitive classification. ICET uses a genetic algorithm to evolve a population of biases for a decision tree induction algorithm. The fitness

In this doctoral dissertation, we study three basic problems in machine learning and two new hypothesis spaces with corresponding learning algorithms. The problems we investigate are: accuracy estimation, feature subset selection, and parameter tuning. The latter two problems are related and are studied under the wrapper approach. The hypothesis spaces we investigate are: decision tables with a default majority rule (DTMs) and oblivious read-once decision graphs (OODGs).

Several recent machine learning publications demonstrate the utility of using feature selection algorithms in supervised learning tasks. Among these, scqucnlial feature s1ion algorithms are receiving attention. The most frequently studied variants of these algorithms are forward and backward sequential selection. Many studies on supervised learning with sequential feature selection report applications of these algorithms, but do not consider variants of them that might be more appropriate for some performance tasks. This paper reports positive empirical results on such variants, and argues for their serious consideration in similar learning tasks.

A central problem in machine learning is identifying a representative set of features from which to construct a classification model for a particular task. This thesis addresses the problem of feature selection for machine learning through a correlation based approach. The central hypothesis is that good feature sets contain features that are highly correlated with the class, yet uncorrelated with each other. A feature evaluation formula, based on ideas from test theory, provides an operational definition of this hypothesis. CFS (Correlation based Feature Selection) is an algorithm that couples this evaluation formula with an appropriate correlation measure and a heuristic search strategy. CFS was evaluated by experiments on artificial and natural datasets. Three machine learning algorithms were used: C4.5 (a decision tree learner), IB1 (an instance based learner), and naive Bayes. Experiments on artificial datasets showed that CFS quickly identifies and screens irrelevant, redundant, and noisy features, and identifies relevant features as long as their relevance does not strongly depend on other features. On natural domains, CFS typically eliminated well over half the features. In most cases, classification accuracy using the reduced feature set equaled or bettered accuracy using the complete feature set.

Abstract. Outlier detection has been used for centuries to detect and, where appropriate, remove anomalous observations from data. Outliers arise due to mechanical faults, changes in system behaviour, fraudulent behaviour, human error, instrument error or simply through natural deviations in populations. Their detection can identify system faults and fraud before they escalate with potentially catastrophic consequences. It can identify errors and remove their contaminating effect on the data set and as such to purify the data for processing. The original outlier detection methods were arbitrary but now, principled and systematic techniques are used, drawn from the full gamut of Computer Science and Statistics. In this paper, we introduce a survey of contemporary techniques for outlier detection. We identify their respective motivations and distinguish their advantages and disadvantages in a comparative review.

Several well-developed approaches to inductive learning now exist, but each has specific limitations that are hard to overcome. Multi-strategy learning attempts to tackle this problem by combining multiple methods in one algorithm. This article describes a unification of two widely-used empirical approaches: rule induction and instance-based learning. In the new algorithm, instances are treated as maximally specific rules, and classification is performed using a best-match strategy. Rules are learned by gradually generalizing instances until no improvement in apparent accuracy is obtained. Theoretical analysis shows this approach to be efficient. It is implemented in the RISE 3.1 system. In an extensive empirical study, RISE consistently achieves higher accuracies than state-of-the-art representatives of both its parent approaches (PEBLS and CN2), as well as a decision tree learner (C4.5). Lesion studies show that each of RISE's components is essential to this performance. Most signi...

In this paper, we identify two issues involved in developing an automated feature subset selection algorithm for unlabeled data: the need for finding the number of clusters in conjunction with feature selection, and the need for normalizing the bias of feature selection criteria with respect to dimension. We explore the feature selection problem and these issues through FSSEM (Feature Subset Selection using Expectation-Maximization (EM) clustering) and through two different performance criteria for evaluating candidate feature subsets: scatter separability and maximum likelihood. We present proofs on the dimensionality biases of these feature criteria, and present a cross-projection normalization scheme that can be applied to any criterion to ameliorate these biases. Our experiments show the need for feature selection, the need for addressing these two issues, and the effectiveness of our proposed solutions.

High sensitivity to irrelevant features is arguably the main shortcoming of simple lazy learners. In response to it, many feature selection methods have been proposed, including forward sequential selection (FSS) and backward sequential selection (BSS). Although they often produce substantial improvements in accuracy, these methods select the same set of relevant features everywhere in the instance space, and thus represent only a partial solution to the problem. In general, some features will be relevant only in some parts of the space; deleting them may hurt accuracy in those parts, but selecting them will have the same effect in parts where they are irrelevant. This article introduces RC, a new feature selection algorithm that uses a clusteringlike approach to select sets of locally relevant features (i.e., the features it selects may vary from one instance to another). Experiments in a large number of domains from the UCI repository show that RC almost always improves accuracy with...

This paper presents a new approach to inductive learning that combines aspects of instancebased learning and rule induction in a single simple algorithm. The RISE system searches for rules in a specific-to-general fashion, starting with one rule per training example, and avoids some of the difficulties of separate-andconquer approaches by evaluating each proposed induction step globally, i.e., through an efficient procedure that is equivalent to checking the accuracy of the rule set as a whole on every training example. Classification is performed using a best-match strategy, and reduces to nearest-neighbor if all generalizations of instances were rejected. An extensive empirical study shows that RISE consistently achieves higher accuracies than state-of-the-art representatives of its "parent" paradigms (PEBLS and CN2), and also outperforms a decision-tree learner (C4.5) in 13 out of 15 test domains (in 10 with 95% confidence). 1 Introduction Several well-developed approaches to indu...