In this paper, a novel approach to feature extraction for classification is proposed based directly on the decision boundaries. We note that feature extraction is equivalent to retaining informative features or eliminating redundant features; thus, the terms “discriminantly information feature” and “discriminantly redundant feature” are first defined relative to feature extraction for classification. Next, it is shown how discriminantly redundant features and discriminantly informative features are related to decision boundaries. A novel characteristic of the proposed method arises by noting that usually only a portion of the decision boundary is effective in discriminating between classes, and the concept of the effective decision boundary is therefore introduced. Next, a procedure to extract discriminantly informative features based on a decision boundary is proposed. The proposed feature extraction algorithm has several desirable properties: 1) It predicts the minimum number of features necessary to achieve the same classification accuracy as in the original space for a given pattern recognition problem; 2) it finds the necessary feature vectors. The proposed algorithm does not deteriorate under the circumstances of equal class means or equal class covariances as some previous algorithms do. Experiments show that the performance of the proposed algorithm compares favorably with those of previous algorithms.
|Number of pages||13|
|Journal||IEEE transactions on pattern analysis and machine intelligence|
|Publication status||Published - 1993 Apr|
Bibliographical noteFunding Information:
Manuscript received February 4, 1991; revised June 30, 1992. This work was supported by NASA under grant NAGW-925. Recommended for acceptance by Editor-in-Chief A. K. Jain. The authors are with the School of Electrical Engineering, Purdue University, West Lafayette, IN 47907. IEEE Log Number 9206555.
All Science Journal Classification (ASJC) codes
- Computer Vision and Pattern Recognition
- Computational Theory and Mathematics
- Artificial Intelligence
- Applied Mathematics