Reducing Dimensionality in Multiple Instance Learning with a Filter Method
In this article, we describe a feature selection algorithm which can automatically find relevant features for multiple instance learning. Multiple instance learning is considered an extension of traditional supervised learning where each example is made up of several instances and there is no specific information about particular instance labels. In this scenario, traditional supervised learning can not be applied directly and it is necessary to design new techniques. Our approach is based on principles of the well-known Relief-F algorithm which is extended to select features in this new learning paradigm by modifying the distance, the difference function and computation of the weight of the features. Four different variants of this algorithm are proposed to evaluate their performance in this new learning framework. Experiment results using a representative number of different algorithms show that predictive accuracy improves significantly when a multiple instance learning classifier is learnt on the reduced data set.
KeywordsFeature Selection Feature Selection Method Multiple Instance Feature Selection Algorithm Multiple Instance Learn
Unable to display preview. Download preview PDF.
- 4.Yuan, X., Hua, X.S., Wang, M., Qi, G.J., Wu, X.Q.: A novel multiple instance learning approach for image retrieval based on adaboost feature selection. In: ICME 2007: Proceedings of the IEEE International Conference on Multimedia and Expo., Beijing, China, pp. 1491–1494. IEEE, Los Alamitos (2007)CrossRefGoogle Scholar
- 6.Herman, G., Ye, G., Xu, J., Zhang, B.: Region-based image categorization with reduced feature set. In: Proceedings of the 10th IEEE Workshop on Multimedia Signal Processing, Cairns, Qld, pp. 586–591 (2008)Google Scholar
- 7.Kononenko, I.: Estimating attributes: analysis and extension of relief. In: Bergadano, F., De Raedt, L. (eds.) ECML 1994. LNCS, vol. 784, pp. 171–182. Springer, Heidelberg (1994)Google Scholar
- 8.Chevaleyre, Y.Z., Zucker, J.D.: Solving multiple-instance and multiple-part learning problems with decision trees and decision rules. Application to the mutagenesis problem. In: Stroulia, E., Matwin, S. (eds.) Canadian AI 2001. LNCS (LNAI), vol. 2056, pp. 204–214. Springer, Heidelberg (2001)CrossRefGoogle Scholar
- 9.Zhang, D., Wang, F., Si, L., Li, T.: M3IC: Maximum margin multiple instance clustering, pp. 1339–1344 (2009)Google Scholar
- 11.Edgar, G.: Measure, topology, and fractal geometry, 3rd edn. Springer, Heidelberg (1995)Google Scholar
- 12.Cohen, H.: Image restoration via n-nearest neighbour classification. In: ICIP 1996: Proceedings of the International Conference on Image Processing, pp. 1005–1007 (1996)Google Scholar
- 13.Yang, C., Lozano-Perez, T.: Image database retrieval with multiple-instance learning techniques. In: ICDE 2000: Proceedings of the 16th International Conference on Data Engineering, Washington, DC, USA, pp. 233–243. IEEE Computer Society, Los Alamitos (2000)Google Scholar