Hall M.A. Correlation-based feature selection for machine learning. PhD thesis, Department of Computer Science, University of Waikato, Hamilton, New Zealand (1998).  Home/Search   Document Details and Download   Summary   Related Articles   Check

....evaluation function. In order to reduce the problem of higher order feature content feature feature correlation, causally complexly interacting features are identified through Bayesian network d separation analysis and combined into joint features. 4. 1 Correlation Based Feature Set Evaluation In [13] a correlation based feature set evaluation function, the CFS function, is introduced within the area of off line feature selection in machine learning. The CFSfunction measures the prediction capability of a feature set S = 1 , F n with respect to a hypothesis H , based on the first order ....

....the MIG function when it comes to representing the prediction capability of causally complexly interacting features. Despite this limitation, feature selection based on the CFS function scores comparable to other state of the art feature selection approaches on several feature selection benchmarks [13] (as long as higherorder correlation is limited) Due to its low computational complexity and due to its state of the art performance on feature selection benchmarks, the CFS function seems to be a good candidate for evaluating the prediction capability of a feature set in real time hypothesis ....

M. A. Hall, Correlation-based Feature Selection for Machine Learning, Ph.D. thesis, Department of Computer Science, Waikato University, New Zealand, 1999.

....ones. Our motivation to investigate filter and wrapper in this work can be summarized as follows. In [Das01] the boosting based were compared to alternative wrapper algorithms and it was concluded they perform equivalently. However, there were no comparisons with other filter methods (e.g. Hal99] Therefore, we investigate how the boosting based method of [TV00] compares with filter methods based on support vector machines (SVM) and information gain [Hal99] In terms of wrapper methods, the simulations in [Ng98] used only synthetic data. The ordered fs algorithm was used with real life ....

....algorithms and it was concluded they perform equivalently. However, there were no comparisons with other filter methods (e.g. Hal99] Therefore, we investigate how the boosting based method of [TV00] compares with filter methods based on support vector machines (SVM) and information gain [Hal99] In terms of wrapper methods, the simulations in [Ng98] used only synthetic data. The ordered fs algorithm was used with real life (microarray) data in [XJK01] but no comparisons of ordered fs with other methods were presented. Our experimental framework seems appropriate to evaluate some ....

[Article contains additional citation context not shown here]

M. Hall. Correlation-based feature selection for machine learning. PhD thesis, University of Waikato, 1999.

....also partially eliminated by data preprocessing; mainly by detecting and removing irrelevant features, or by selecting most relevant ones [25, 26] This method is called Feature Reduction. There are many methods for selecting relevant attributes, for example methods using PAC [23] or correlation [24]. Example of such data can be find in section 4, where we try on SPA data to predict capacity required for therapeutic utilities. 2.4 Learning The term learning usually corresponds to fitting designed model. Through the process of learning, we are improving model prediction accuracy as fast ....

M. Hall. Correlation-based Feature Selection for Machine Learning. PhD thesis, Waikato University, Department of Computer Science, Hamilton, NZ, 1998.

....learners, naive Bayes, multi layer perceptrons etc. and basic evaluation methods like cross validation and bootstrapping [1] WEKA has some pre processing algorithms for the manipulation of attributes as well as three basic feature selection schemes, namely the feature correlation based approach [2], a wrapper approach [3] and a filter approach. Additionaly WEKA provides meta classifiers like bagging and boosting. MLC is a library of C classes for supervised machine learning. It provides a number of learning schemes similar to those used in WEKA. Additionally wrappers around these basic ....

M.A. Hall. Correlation-based feature selection for machine learning. Dissertation, Department of Computer Science, University of Waikato, Hamilton, New Zealand, 1999.

No context found.

M. A. Hall, Correlation-based feature selection for machine learning, Ph.D. thesis, Department of Computer Science, University of Waikato, Hamilton, New Zealand, 1998.

....of principal components handles k valued discrete attributes by converting them to k binary attributes. This has the disadvantage of increasing the dimensionality of the original space when multi valued discrete attributes are present. 2. 4 CFS CFS (Correlation based Feature Selection) [5] is the rst of the methods that evaluate subsets of attributes rather than individual attributes. At the heart of the algorithm is a subset evaluation heuristic that takes into account the usefulness of individual features for predicting the class along with the level of intercorrelation among ....

M. A. Hall. Correlation-based feature selection for machine learning. PhD thesis, Department of Computer Science, University of Waikato, Hamilton, New Zealand, 1998.

....class. In some cases however, there may be subsidiary features that are locally predictive in a small area of the instance space. Some machine learning algorithms are able to make use of locally predictive features and in these situations CFS has been shown to degrade their performance somewhat [7]. The version of CFS used in the experiments described in this paper includes a heuristic to include locally predictive features and avoid the re introduction of redundancy. After the feature subset space has been searched, the remaining unselected features are examined one by one to determine ....

Hall, M. A. 1998. Correlation-based Feature Selection for Machine Learning. Ph.D diss. Hamilton, NZ: Waikato University, Department of Computer Science.

No context found.

Hall M.A. Correlation-based feature selection for machine learning. PhD thesis, Department of Computer Science, University of Waikato, Hamilton, New Zealand (1998).

No context found.

Hall M.A. Correlation-based feature selection for machine learning. PhD thesis, Department of Computer Science, University of Waikato, Hamilton, New Zealand (1998).

No context found.

Hall, M.A. (1999). Correlation based Feature Selection for Machine Learning. Doctoral dissertation, Department of Computer Science, The University of Waikato, Hamilton, New Zealand.

No context found.

M.A. Hall, Correlation based feature selection for machine learning. PhD thesis, Dept. of Comp. Science, Univ. of Waikato, Hamilton, New Zealand (1998)

No context found.

Mark A. Hall. Correlation-based Feature Selection for Machine Learning. PhD thesis, Waikato University, Hamilton, NZ, 1998.

No context found.

M.A. Hall. Correlation-based feature selection machine learning. PhD thesis, Department of Computer Science, University of Waikato, New Zealand, 1998.

No context found.

M. A. Hall, Correlation-based Feature Selection for Machine Learning, PhD thesis, Waikato University, New Zealand, 1999.

No context found.

Mark A. Hall. Correlation-Based Feature Selection for Machine Learning. PhD thesis, Department of Computer Science, University of Waikato, Hamilton, New Zealand, Apr. 1999. 44

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC