Multi-label Classification Using Rough Sets
In multi-label classification, each instance may be associated with multiple labels simultaneously which is different from the traditional single-label classification where an instance is only associated with a single label. In this paper, we propose two types of approaches to deal with multi-label classification problem based on rough sets. The first type of approach is to transform the multi-label problem into one or more single-label problems and then use the classical rough set model to make decisions. The second type of approach is to extend the classical rough set model in order to handle multi-label dataset directly, where the new model considers the correlations among labels. The effectiveness of multi-label rough set model is presented by a series of experiments completed for two multi-label datasets.
Keywordsrough sets multi-label classification correlation
Unable to display preview. Download preview PDF.
- 5.Wang, G.: Extension of rough set under incomplete information systems. Journal of Computer Research and Development 10, 1–9 (2002)Google Scholar
- 11.De Comit, F., Gilleron, R., Tommasi, M.: Learning multi-label alternating decision trees from texts and data. In: Perner, P., Rosenfeld, A. (eds.) MLDM 2003. LNCS, vol. 2734, pp. 251–274. Springer, Heidelberg (2003)Google Scholar
- 15.Yao, Y.Y., Wong, S.K.M., Lingras, P.: A decision-theoretic rough set model, Methodologies for Intelligent Systems. In: Ras, Z.W., Zemankova, M., Emrichm, M.L. (eds.) Methodologies for Intelligent Systems, vol. 5, pp. 17–25. North-Holland, New York (1990)Google Scholar