We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
An Unsupervised Approach for Selection of Candidate Feature Set Using Filter Based Techniques.
- Authors
POTHARAJU, Sai Prasad; SREEDEVI, Marriboyina
- Abstract
High dimensionality is the one of the important issue in preprocessing stage of data mining. Initial feature space may have irrelevant or redundant features. These properties of features decrease the performance of classifier, and also require more memory and high computing power. This issue can be addressed by selecting the best feature subset for improving the classification performance. In this research, we have proposed an unsupervised approach using filter based feature selection methods and K-Means clustering technique to derive the candidate subset. Score of each feature is calculated using traditional filter based methods. Then Min-Max technique is applied to normalize the dataset. K-Means algorithm is employed on the dataset to form the clusters of features. To decide the strong subset, Multi-Layer Perceptron(MLP) is applied on each cluster. Best cluster is selected based on the minimum Root Mean Square (RMS) error rate given by MLP. This framework is compared with traditional methods over six well known datasets having the total features in between 34 and 90 using various classification algorithms. The proposed method recorded 75% competitive rate than Information Gain(IG), 71% success rate than Gain Ratio Attribute Evaluator(GR) and Chi Square Attribute Evaluator(Chi), 83% competitive rate than ReliefF(Rel) traditional methods. Jrip classifier performed 55%, J48 recorded 66%, Naive Bayes displayed 88%, IBK (Instance Based) displayed 80% success rate over all the datasets.
- Publication
Gazi University Journal of Science, 2018, Vol 31, Issue 3, p789
- ISSN
1303-9709
- Publication type
Academic Journal