Data Mining: Practical Machine Learning Tools and Techniques offers a thorough grounding in machine learning concepts as well as practical advice on applying machine learning tools and techniques in real-world data mining situations. This highly anticipated third edition of the most acclaimed work on data mining and machine learning will teach you everything you need to know about preparing inputs, interpreting outputs, evaluating results, and the algorithmic methods at the heart of successful data mining. Thorough updates reflect the technical changes and modernizations that have taken place in the field since the last edition, including new material on Data Transformations, Ensemble Learning, Massive Data Sets, Multi-instance Learning, plus a new version of the popular Weka machine learning software developed by the authors. Witten, Frank, and Hall include both tried-and-true techniques of today as well as methods at the leading edge of contemporary research. *Provides a thorough grounding in machine learning concepts as well as practical advice on applying the tools and techniques to your data mining projects *Offers concrete tips and techniques for performance improvement that work by transforming the input or output in machine learning methods *Includes downloadable Weka software toolkit, a collection of machine learning algorithms for data mining tasks-in an updated, interactive interface. Algorithms in toolkit cover: data pre-processing, classification, regression, clustering, association rules, visualization

Data Mining: Practical Machine Learning Tools and Techniques

https://ai.stanford.edu/~ronnyk/wrappersPrint.pdf

Wrappers for feature subset selection

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. Feature selection degraded machine learning performance in cases where some features were eliminated which were highly predictive of very small areas of the instance space. Further experiments compared CFS with a wrapper—a well known approach to feature selection that employs the target learning algorithm to evaluate feature sets. In many cases CFS gave comparable results to the wrapper, and in general, outperformed the wrapper on small datasets. CFS executes many times faster than the wrapper, which allows it to scale to larger datasets. Two methods of extending CFS to handle feature interaction are presented and experimentally evaluated. The first considers pairs of features and the second incorporates iii feature weights calculated by the RELIEF algorithm. Experiments on artificial domains showed that both methods were able to identify interacting features. On natural domains, the pairwise method gave more reliable results than using weights provided by RELIEF.

/pdf/correlation-based-feature-selection-for-machine-learning-578rirmvg5.pdf

Correlation-based Feature Selection for Machine Learning

http://library.utia.cas.cz/separaty/historie/somol-floating%20search%20methods%20in%20feature%20selection.pdf

Floating search methods in feature selection

The overall mean recognition probability (mean accuracy) of a pattern classifier is calculated and numerically plotted as a function of the pattern measurement complexity n and design data set size m . Utilized is the well-known probabilistic model of a two-class, discrete-measurement pattern environment (no Gaussian or statistical independence assumptions are made). The minimum-error recognition rule (Bayes) is used, with the unknown pattern environment probabilities estimated from the data relative frequencies. In calculating the mean accuracy over all such environments, only three parameters remain in the final equation: n, m , and the prior probability p_{c} of either of the pattern classes. With a fixed design pattern sample, recognition accuracy can first increase as the number of measurements made on a pattern increases, but decay with measurement complexity higher than some optimum value. Graphs of the mean accuracy exhibit both an optimal and a maximum acceptable value of n for fixed m and p_{c} . A four-place tabulation of the optimum n and maximum mean accuracy values is given for equally likely classes and m ranging from 2 to 1000 . The penalty exacted for the generality of the analysis is the use of the mean accuracy itself as a recognizer optimality criterion. Namely, one necessarily always has some particular recognition problem at hand whose Bayes accuracy will be higher or lower than the mean over all recognition problems having fixed n, m , and p_{c} .

On the mean accuracy of statistical pattern recognizers

In the type of recognition system under discussion, the physical sample to be recognized is first subjected to a battery of tests; on the basis of the test results, the sample is then assigned to one of a number of prespecified categories. The theory of how test results should be combined to yield an optimal assignment has been discussed in an earlier paper. Here, attention is focused on the tests themselves. At present, we usually measure the effectiveness of a set of tests empirically, i.e., by determining the percentage of correct recognitions made by some recognition device which uses these tests. In this paper, we discuss some of the theoretical problems encountered in trying to determine a more formal measure of the effectiveness of a set of tests; a measure which might be a practical substitute for the empirical evaluation. Specifically, the following question is considered: What constitutes an effective set of tests, and how is this effectiveness dependent on the correlations among, and the properties of, the individual tests in the set? Specific suggestions are considered for the case in which the test results are normally distributed, but arbitrarily correlated. The discussion is supported by the results of experiments dealing with automatic recognition of hand-printed characters.

T. Marill

Papers

On the effectiveness of receptors in recognition systems