Abstract
The k-nearest neighbor algorithm (kNN) is one of the most well-known techniques in standard supervised learning. It could be adapted to the setting of multiple-instance learning (MIL) by using set-based distance metrics, such as Citation-kNN and Bayesian-kNN. However, kNN suffers from several drawbacks, including high storage requirements, low efficiency in classification response and low noise tolerance. These drawbacks would become particularly significant in MIL since every example here is a set of instances. One of the most promising solutions is dependent on prototype selection, and many prototype selection methods have been proposed in standard supervised learning. In this paper, we propose an efficient Salience-based Prototype Selection (MISPS) method to tackle the above problems in MIL. Then we present two variants of Citation-kNN and Bayesian-kNN based on MISPS, called MISPS-CkNN and MISPS-BkNN. Experimental results on five benchmark data-sets show that MISPS is effective and our MISPS-based algorithms are competitive to the state-of-the-art.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Dietterich, T.G., Lathrop, R.H., Lozano-Pérez, T.: Solving the Multiple Instance Problem with Axis-parallel Rectangles. Artif. Intell. 89, 31–71 (1997)
Viola, P., Platt, J., Zhang, C.: Multiple Instance Boosting for Object Detection. In: NIPS, pp. 1417–1424. MIT Press, Cambridge (2006)
Fung, G., Dundar, M., Krishnapuram, B., Rao, B.R.: Multiple Instance Learning for Computer Aided Diagnosis. In: NIPS, pp. 425–432. MIT Press, Cambridge (2007)
Babenko, B., Yang, M.H., Belongie, S.: Robust Object Tracking with Online Multiple Instance Learning. IEEE Trans. Pattern Anal. Mach. Intell. 33, 1619–1632 (2011)
Zhang, Q., Goldman, S.A., Yu, W., Fritts, J.E.: Content-Based Image Retrieval Using Multiple-Instance Learning. In: ICML, pp. 682–689. Morgan Kaufmann, San Francisco (2002)
Rahmani, R., Goldman, S.A., Zhang, H., Cholleti, S.R., Fritts, J.E.: Localized Content-Based Image Retrieval. IEEE Trans. Pattern Anal. Mach. Intell. 30, 1902–1912 (2008)
Wang, J., Zucker, J.D.: Solving the Multiple-Instance Problem: A Lazy Learning Approach. In: ICML, pp. 1119–1126. Morgan Kaufmann, San Francisco (2000)
Garcia, S., Derrac, J., Cano, J., Herrera, F.: Prototype Selection for Nearest Neighbor Classification: Taxonomy and Empirical Study. IEEE Trans. Pattern Anal. Mach. Intell. 34, 417–435 (2012)
Maron, O., Lozano-Pérez, T.: A Framework for Multiple-Instance Learning. In: NIPS, pp. 570–576. MIT Press, Cambridge (1998)
Zhang, Q., Goldman, S.A.: EM-DD: An Improved Multiple-Instance Learning Technique. In: NIPS, pp. 1073–1080. MIT Press, Cambridge (2001)
Ramon, J., De Raedt, L.: Multi Instance Neural Networks. In: ICML 2000 Workshop on Attribute-Value and Relational Learning (2000)
Andrews, S., Tsochantaridis, I., Hofmann, T.: Support Vector Machines for Multiple-Instance Learning. In: NIPS, pp. 561–568. MIT Press, Cambridge (2003)
Chen, Y., Wang, J.Z.: Image Categorization by Learning and Reasoning with Regions. J. Mach. Learn. Res. 5, 913–939 (2004)
Chen, Y., Bi, J., Wang, J.Z.: MILES: Multiple-Instance Learning via Embedded Instance Selection. IEEE Trans. Pattern Anal. Mach. Intell. 28, 1931–1947 (2006)
Li, W.J., Yeung, D.Y.: MILD: Multiple-Instance Learning via Disambiguation. IEEE Trans. on Knowl. and Data Eng. 22, 76–89 (2010)
Fu, Z., Robles-Kelly, A., Zhou, J.: MILIS: Multiple Instance Learning with Instance Selection. IEEE Trans. Pattern Anal. Mach. Intell. 33, 958–977 (2011)
Erdem, A., Erdem, E.: Multiple-Instance Learning with Instance Selection via Dominant Sets. In: Pelillo, M., Hancock, E.R. (eds.) SIMBAD 2011. LNCS, vol. 7005, pp. 177–191. Springer, Heidelberg (2011)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Yuan, L., Huang, Q., Liu, J., Tang, X. (2013). Salience-Based Prototype Selection for K-Nearest Neighbor Classification in Multiple-Instance Learning. In: Yang, J., Fang, F., Sun, C. (eds) Intelligent Science and Intelligent Data Engineering. IScIDE 2012. Lecture Notes in Computer Science, vol 7751. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-36669-7_6
Download citation
DOI: https://doi.org/10.1007/978-3-642-36669-7_6
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-36668-0
Online ISBN: 978-3-642-36669-7
eBook Packages: Computer ScienceComputer Science (R0)