COMBINING TWO LAZY LEARNING METHODS FOR CLASSIFICATION AND KNOWLEDGE DISCOVERY - A Case Study for Malignant Melanoma Diagnosis

Eva Armengol, Susana Puig

2011

Abstract

The goal of this paper is to construct a classifier for diagnosing malignant melanoma. We experimented with two lazy learning methods, k-NN and LID, and compared their results with the ones produced by decision trees. We performed this comparison because we are also interested on building a domain model that can serve as basis to dermatologists to propose a good characterization of early melanomas. We shown that lazy learning methods have a better performance than decision trees in terms of sensitivity and specificity. We have seen that both lazy learning methods produce complementary results (k-NN has high specificity and LID has high sensitivity) suggesting that a combination of both could be a good classifier. We report experiments confirming this point. Concerning the construction of a domain model, we propose to use the explanations provided by the lazy learning methods, and we see that the resulting theory is as predictive and useful as the one obtained from decision trees.

References

  1. Argenziano, G., Zalaudek, I., Ferrara, G., HofmannWellenhof, R., and Soyer, H. (2007). Proposal of a new classification system for melanocytic naevi. Br J Dermatol, 157(2):217-227.
  2. Armengol, E. (2008). Building partial domain theories from explanations. Knowledge Intelligence, 2(8):19-24.
  3. Armengol, E. and Plaza, E. (2001). Lazy induction of descriptions for relational case-based learning. In Reaedt, L. D. and Flach, P., editors, ECML-2001., number 2167 in Lecture Notes in Artificial Intelligence, pages 13-24. Springer.
  4. Bareiss, E. R., Porter, B. W., and Wier, C. C. (1988). PROTOS: an examplar-based learning apprentice. Int. J. Man-Mach. Stud., 29(5):549-561.
  5. Fawcett, T. (2006). An introduction to ROC analysis. Pattern Recogn. Lett., 27:861-874.
  6. Helma, C. and Kramer, S. (2003). A survey of the predictive toxicology challenge 2000-2001. Bioinformatics, pages 1179-1200.
  7. Hofmann-Wellenhof, R., Blum, A., Wolf, I., Zalaudek, I., Piccolo, D., Kerl, H., Garbe, C., and Soyer, H. (2002). Dermoscopic classification of Clark's nevi (atypical melanocytic nevi). Clin Dermatol, 20(3):255-258.
  8. López de Mántaras, R. (1991). A distance-based attribute selection measure for decision tree induction. Machine Learning, 6:81-92.
  9. Mitchell, T. (1997). Machine Learning. McGraw-Hill International Editions. Computer Science Series.
  10. Mitchell, T., Keller, R., and Kedar-Cabelli, S. (1986). Explanation-based learning: A unifying view. Machine Learning, 1(1):47-80.
  11. Plaza, E., Armengol, E., and Ontan˜ón, S. (2005). The explanatory power of symbolic similarity in case-based reasoning. Artificial Intelligence Review. Special Issue on Explanation in Case-based Reasoning, 24:145- 161.
  12. Prodromidis, A., Chan, P., and Stolfo, S. (2000). Metalearning in distributed data mining systems: Issues and approaches. In Book on Advances of Distributed Data Mining, editors Hillol Kargupta and Philip Chan, AAAI press, 2000.
  13. Puig, S., Argenziano, G., Zalaudek, I., Ferrara, G., Palou, J., Massi, D., Hofmann-Wellenhof, R., Soyer, H., and Malvehy, J. (2007). Melanomas that failed dermoscopic detection: a combined clinicodermoscopic approach for not missing melanoma. Dermatol Surg, 33(10):1262-1273.
  14. Roth-Berghofer, T. R. (2004). Explanations and casebased reasoning: Foundational issues. In Funk, P. and Calero, P. A. G., editors, Advances in Case-Based Reasoning, pages 389-403. Springer-Verlag.
  15. Vestergaard, M. and Menzies, S. (2008). Automated diagnostic instruments for cutaneous melanoma. Semin Cutan Med Surg, 27(1):32-6.
  16. Witten, I., Frank, E., Trigg, L., Hall, M., Holmes, G., and Cunningham, S. (1999). Weka: Practical machine learning tools and techniques with Java implementations.
Download


Paper Citation


in Harvard Style

Armengol E. and Puig S. (2011). COMBINING TWO LAZY LEARNING METHODS FOR CLASSIFICATION AND KNOWLEDGE DISCOVERY - A Case Study for Malignant Melanoma Diagnosis . In Proceedings of the International Conference on Knowledge Discovery and Information Retrieval - Volume 1: KDIR, (IC3K 2011) ISBN 978-989-8425-79-9, pages 192-199. DOI: 10.5220/0003652202000207


in Bibtex Style

@conference{kdir11,
author={Eva Armengol and Susana Puig},
title={COMBINING TWO LAZY LEARNING METHODS FOR CLASSIFICATION AND KNOWLEDGE DISCOVERY - A Case Study for Malignant Melanoma Diagnosis},
booktitle={Proceedings of the International Conference on Knowledge Discovery and Information Retrieval - Volume 1: KDIR, (IC3K 2011)},
year={2011},
pages={192-199},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003652202000207},
isbn={978-989-8425-79-9},
}


in EndNote Style

TY - CONF
JO - Proceedings of the International Conference on Knowledge Discovery and Information Retrieval - Volume 1: KDIR, (IC3K 2011)
TI - COMBINING TWO LAZY LEARNING METHODS FOR CLASSIFICATION AND KNOWLEDGE DISCOVERY - A Case Study for Malignant Melanoma Diagnosis
SN - 978-989-8425-79-9
AU - Armengol E.
AU - Puig S.
PY - 2011
SP - 192
EP - 199
DO - 10.5220/0003652202000207