An Accelerator of Feature Selection Applying a General Fuzzy Rough Model
Feature selection, also known as variable selection or attribute reduction, is to select a subset relevant features to speedup learning/mining and to improve the learning/mining quality. In the big data era, some feature selection methods have to face the running time problem led by the large-scale data. As a result, in this paper, we try to narrow this gap by proposing a feature selection accelerator. Considering fuzzy rough techniques need no extra expert knowledge, we design the feature selection accelerator based on fuzzy rough reduction techniques. First, we proposed a fuzzy rough accelerator by deleting the learned/discernible instances in the process of feature selection, which decreases the computation and accelerates feature selection. Second, we design a fuzzy rough based feature selection accelerated algorithm. Finally, the numerical experiments demonstrate that the proposed accelerated algorithm could obtain the same reduction results and save much more time, especially on the large-scale datasets.
KeywordsFeature selection Fuzzy rough techniques Accelerator Fuzzy positive region
This work is supported by National Key Research & Develop Plan (No. 2016YFB1000702), National Key R&D Program of China (2017YFB1400700), and NSFC under the grant No. 61732006, 61532021, 61772536, 61772537, 61702522 and NSSFC (No. 12\&ZD220), and the Fundamental Research Funds for the Central Universities, and the Research Funds of Renmin University of China (15XNLQ06). It was partially done when the authors worked in SA Center for Big Data Research in RUC. This Center is funded by a Chinese National 111 Project Attracting. This work is also supported by the Macao Science and Technology Development Fund (081/2015/A3).
- 2.Kira, K., Rendell, L.A.: A practical approach to feature selection. In: Proceedings of the 9th International Conference on Machine Learning, pp. 249–256. Morgan Kaufmann, Los Altos (1992)Google Scholar
- 18.Kryszkiewicz, M., Lasek, P.: FUN: fast discovery of minimal sets of attributes functionally determining a decision attribute. Trans. Rough Sets 9, 76–95 (2008)Google Scholar
- 21.Chen, D.G., Tsang, E.C.C., Zhao, S.Y.: Attributes reduction with fuzzy rough sets. In: IEEE International Conference on Systems, Man, and Cybernetics, Vol. 1, pp. 486–491 (2007)Google Scholar