Effective Linear-Time Feature Selection
Pradhananga, N. (2007). Effective Linear-Time Feature Selection (Thesis, Master of Science (MSc)). The University of Waikato, Hamilton, New Zealand. Retrieved from http://hdl.handle.net/10289/2315
Permanent Research Commons link: http://hdl.handle.net/10289/2315
The classification learning task requires selection of a subset of features to represent patternsto be classified. This is because the performance of the classifier and the cost ofclassification are sensitive to the choice of the features used to construct the classifier.Exhaustive search is impractical since it searches every possible combination of features.The runtime of heuristic and random searches are better but the problem still persistswhen dealing with high-dimensional datasets.We investigate a heuristic, forward, wrapper-based approach, called Linear SequentialSelection, which limits the search space at each iteration of the feature selection process.We introduce randomization in the search space. The algorithm is called RandomizedLinear Sequential Selection. Our experiments demonstrate that both methods are faster,find smaller subsets and can even increase the classification accuracy.We also explore the idea of ensemble learning. We have proposed two ensemble creationmethods, Feature Selection Ensemble and Random Feature Ensemble. Both methods applya feature selection algorithm to create individual classifiers of the ensemble. Ourexperiments have shown that both methods work well with high-dimensional data.
The University of Waikato
All items in Research Commons are provided for private study and research purposes and are protected by copyright with all rights reserved unless otherwise indicated.
- Masters Degree Theses