Skip to main content
Erschienen in:
Buchtitelbild

2016 | OriginalPaper | Buchkapitel

Algorithm Selection Using Performance and Run Time Behavior

Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.

search-config
loading …

Abstract

In data mining, an important early decision for a user to make is to choose an appropriate technique for analyzing the dataset at hand so that generalizations can be learned. Intuitively, a trial-and-error approach becomes impractical when the number of data mining algorithms is large while experts’ advice to choose among them is not always available and affordable. Our approach is based on meta-learning, a way to learn from prior learning experience. We propose a new approach using regression to obtain a ranked list of algorithms based on data characteristics and past performance of algorithms in classification tasks. We consider both accuracy and time in generating the final ranked result for classification, although our approach can be extended to regression problems.

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

  • über 102.000 Bücher
  • über 537 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Maschinenbau + Werkstoffe
  • Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Springer Professional "Technik"

Online-Abonnement

Mit Springer Professional "Technik" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 390 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Maschinenbau + Werkstoffe




 

Jetzt Wissensvorsprung sichern!

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 340 Zeitschriften

aus folgenden Fachgebieten:

  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Versicherung + Risiko




Jetzt Wissensvorsprung sichern!

Literatur
1.
Zurück zum Zitat Abdulrahman, M., Brazdil, P.: Measures for combining accuracy and time for meta-learning. In: ECAI, p. 49 (2014) Abdulrahman, M., Brazdil, P.: Measures for combining accuracy and time for meta-learning. In: ECAI, p. 49 (2014)
2.
Zurück zum Zitat Bensusan, H., Kalousis, A.: Estimating the predictive accuracy of a classifier. In: Flach, P.A., De Raedt, L. (eds.) ECML 2001. LNCS (LNAI), vol. 2167, pp. 25–36. Springer, Heidelberg (2001)CrossRef Bensusan, H., Kalousis, A.: Estimating the predictive accuracy of a classifier. In: Flach, P.A., De Raedt, L. (eds.) ECML 2001. LNCS (LNAI), vol. 2167, pp. 25–36. Springer, Heidelberg (2001)CrossRef
3.
Zurück zum Zitat Brazdil, P.B., Soares, C., Da Costa, J.P.: Ranking learning algorithms: using IBL and meta-learning on accuracy and time results. Mach. Learn. 50(3), 251–277 (2003)CrossRefMATH Brazdil, P.B., Soares, C., Da Costa, J.P.: Ranking learning algorithms: using IBL and meta-learning on accuracy and time results. Mach. Learn. 50(3), 251–277 (2003)CrossRefMATH
4.
Zurück zum Zitat Caruana, R., Niculescu-Mizil, A.: Data mining in metric space: an empirical analysis of supervised learning performance criteria. In: Proceedings of the Tenth ACM SIGKDD. ACM (2004) Caruana, R., Niculescu-Mizil, A.: Data mining in metric space: an empirical analysis of supervised learning performance criteria. In: Proceedings of the Tenth ACM SIGKDD. ACM (2004)
5.
Zurück zum Zitat Cutler, A., Zhao, G.: Fast classification using perfect random trees. Utah State University (1999) Cutler, A., Zhao, G.: Fast classification using perfect random trees. Utah State University (1999)
6.
Zurück zum Zitat Dinu, L.P., Iuga, I.: The naive bayes classifier in opinion mining: in search of the best feature set. In: Gelbukh, A. (ed.) CICLing 2012, Part I. LNCS, vol. 7181, pp. 556–567. Springer, Heidelberg (2012)CrossRef Dinu, L.P., Iuga, I.: The naive bayes classifier in opinion mining: in search of the best feature set. In: Gelbukh, A. (ed.) CICLing 2012, Part I. LNCS, vol. 7181, pp. 556–567. Springer, Heidelberg (2012)CrossRef
7.
8.
Zurück zum Zitat Feurer, M., Springenberg, J.T., Hutter, F.: Using meta-learning to initialize bayesian optimization of hyperparameters. In: ECAI Workshop (MetaSel) (2014) Feurer, M., Springenberg, J.T., Hutter, F.: Using meta-learning to initialize bayesian optimization of hyperparameters. In: ECAI Workshop (MetaSel) (2014)
9.
Zurück zum Zitat Frank, E., Hall, M., Pfahringer, B.: Locally weighted naive bayes. In: Proceedings of the Nineteenth Conference on Uncertainty in Artificial Intelligence, pp. 249–256. Morgan Kaufmann Publishers Inc., Burlington (2002) Frank, E., Hall, M., Pfahringer, B.: Locally weighted naive bayes. In: Proceedings of the Nineteenth Conference on Uncertainty in Artificial Intelligence, pp. 249–256. Morgan Kaufmann Publishers Inc., Burlington (2002)
11.
Zurück zum Zitat Gama, J., Brazdil, P.: Characterization of classification algorithms. In: Pinto-Ferreira, C., Mamede, N.J. (eds.) EPIA 1995. LNCS, vol. 990, pp. 189–200. Springer, Heidelberg (1995)CrossRef Gama, J., Brazdil, P.: Characterization of classification algorithms. In: Pinto-Ferreira, C., Mamede, N.J. (eds.) EPIA 1995. LNCS, vol. 990, pp. 189–200. Springer, Heidelberg (1995)CrossRef
12.
Zurück zum Zitat Hall, M., Frank, E.: The WEKA data mining software: an update. ACM SIGKDD Explor. Newslett. 11(1), 10–18 (2009)MathSciNetCrossRef Hall, M., Frank, E.: The WEKA data mining software: an update. ACM SIGKDD Explor. Newslett. 11(1), 10–18 (2009)MathSciNetCrossRef
13.
14.
15.
Zurück zum Zitat Leite, R., Brazdil, P., Vanschoren, J.: Selecting classification algorithms with active testing. In: Perner, P. (ed.) MLDM 2012. LNCS, vol. 7376, pp. 117–131. Springer, Heidelberg (2012)CrossRef Leite, R., Brazdil, P., Vanschoren, J.: Selecting classification algorithms with active testing. In: Perner, P. (ed.) MLDM 2012. LNCS, vol. 7376, pp. 117–131. Springer, Heidelberg (2012)CrossRef
16.
Zurück zum Zitat Maas, A.L., Daly, R.E., Pham, P.T., Huang, D., Ng, A.Y., Potts, C.: Learning word vectors for sentiment analysis. In: 49th ACL, pp. 142–150 (2011) Maas, A.L., Daly, R.E., Pham, P.T., Huang, D., Ng, A.Y., Potts, C.: Learning word vectors for sentiment analysis. In: 49th ACL, pp. 142–150 (2011)
17.
Zurück zum Zitat Prudêncio, R.B.C., de Souto, M.C.P., Ludermir, T.B.: Selecting machine learning algorithms using the ranking meta-learning approach. In: Jankowski, N., Duch, W., Gra̧bczewski, K. (eds.) Meta-Learning in Computational Intelligence. SCI, vol. 358, pp. 225–243. Springer, Heidelberg (2011)CrossRef Prudêncio, R.B.C., de Souto, M.C.P., Ludermir, T.B.: Selecting machine learning algorithms using the ranking meta-learning approach. In: Jankowski, N., Duch, W., Gra̧bczewski, K. (eds.) Meta-Learning in Computational Intelligence. SCI, vol. 358, pp. 225–243. Springer, Heidelberg (2011)CrossRef
18.
Zurück zum Zitat Quinlan, J.R.: Combining instance-based and model-based learning. In: Proceedings of the Tenth International Conference on Machine Learning (1993) Quinlan, J.R.: Combining instance-based and model-based learning. In: Proceedings of the Tenth International Conference on Machine Learning (1993)
19.
Zurück zum Zitat Smola, A.J., et al.: Regression estimation with support vector learning machines. Master’s thesis, Technische Universit at München (1996) Smola, A.J., et al.: Regression estimation with support vector learning machines. Master’s thesis, Technische Universit at München (1996)
20.
Zurück zum Zitat Sorzano, C.O.S., Vargas, J., Montano, A.P.: A survey of dimensionality reduction techniques. arXiv preprint (2014). arXiv:1403.2877 Sorzano, C.O.S., Vargas, J., Montano, A.P.: A survey of dimensionality reduction techniques. arXiv preprint (2014). arXiv:​1403.​2877
21.
Zurück zum Zitat Thornton, C., Hutter, F., Hoos, H.H., Leyton-Brown, K.: Auto-WEKA: combined selection and hyperparameter optimization of classification algorithms. In: 19th SIGKDD. ACM (2013) Thornton, C., Hutter, F., Hoos, H.H., Leyton-Brown, K.: Auto-WEKA: combined selection and hyperparameter optimization of classification algorithms. In: 19th SIGKDD. ACM (2013)
22.
Zurück zum Zitat Wolpert, D., Macready, W.: No free lunch theorems for optimization. IEEE Trans. Evol. Comput. 1(1), 67–82 (1997)CrossRef Wolpert, D., Macready, W.: No free lunch theorems for optimization. IEEE Trans. Evol. Comput. 1(1), 67–82 (1997)CrossRef
23.
Zurück zum Zitat You, Z.H., Lei, Y.K., Zhu, L., Xia, J., Wang, B.: Prediction of protein-protein interactions from amino acid sequences with ensemble extreme learning machines and principal component analysis. BMC Bioinform. 14(8), 1 (2013) You, Z.H., Lei, Y.K., Zhu, L., Xia, J., Wang, B.: Prediction of protein-protein interactions from amino acid sequences with ensemble extreme learning machines and principal component analysis. BMC Bioinform. 14(8), 1 (2013)
Metadaten
Titel
Algorithm Selection Using Performance and Run Time Behavior
verfasst von
Tri Doan
Jugal Kalita
Copyright-Jahr
2016
DOI
https://doi.org/10.1007/978-3-319-44748-3_1

Premium Partner