Feature condensing algorithm for feature selection
Pavel Křížek, Josef Kittler, Václav Hlaváč · Proceedings - International Conference on Pattern Recognition/Proceedings/International Conference on Pattern Recognition · 2008
A new unsupervised filter-based feature selection method is introduced. Its principle consists in merging similar features into clusters using a distance measure derived from the correlation coefficient. Subsequently, only one representative feature is selected from each cluster. In experiments with real-world data, we show that the proposed method is benefical as a pre-filtering step for more sophisticated feature selection techniques.