The Impact of Reliability Evaluation on a Semi-supervised Learning Approach
In self-training methods, unlabeled samples are first assigned a provisional label by the classifier, and then used to extend the training set of the classifier itself. For this latter step it is important to choose only the samples whose classification is likely to be correct, according to a suitably defined reliability measure.
In this paper we want to study to what extent the choice of a particular technique for evaluating the classification reliability can affect the learning performance. To this aim, we have compared five different reliability evaluators on four publicly available datasets, analyzing and discussing the obtained results.
KeywordsNear Neighbor Reliability Evaluation Unlabeled Data Reliability Estimator Unlabeled Sample
- 2.Chapelle, O., Zien, A., Schölkopf, B. (eds.): Semi-supervised learning. MIT Press, Cambridge (2006)Google Scholar
- 5.Foggia, P., Percannella, G., Sansone, C., Vento, M.: Evaluating Classification Reliability for Combining Classifiers. In: Proceedings of the 14th International Conference on Image Analysis and Processing, Modena (IT), September 10-14, pp. 711–716 (2007)Google Scholar
- 6.Gargiulo, F., Mazzariello, C., Sansone, C.: A Self-training Approach for Automatically Labeling IP Traffic Traces. In: Kurzynski, M., Puchala, E., Wozniak, M., Zolnierek, A. (eds.) Computer Recognition Systems, vol. 2, pp. 705–717. Springer, Heidelberg (2008)Google Scholar
- 10.Rosenberg, C., Hebert, M., Schneiderman, H.: Semi-supervised self-training of object detection models. In: Proceedings of the Seventh IEEE Workshop on Applications of Computer Vision, Breckenridge, USA, January 5-7 (2005)Google Scholar
- 11.Zhu, X.: Semi-supervised learning literature survey. Technical Report 1530, Computer Sciences, University of Wisconsin-Madison (2006)Google Scholar