A guidance of data stream characterization for meta-learning.

Item request has been placed! ×
Item request cannot be made. ×
  Processing Request
  • Additional Information
    • Author-Supplied Keywords:
      algorithm selection
      data streams
      Feature extraction
    • Abstract:
      The problem of selecting learning algorithms has been studied by the meta-learning community for more than two decades. One of the most important task for the success of a meta-learning system is gathering data about the learning process. This data is used to induce a (meta) model able to map characteristics extracted from different data sets to the performance of learning algorithms on these data sets. These systems are built under the assumption that the data are generated by a stationary distribution, i.e., a learning algorithm will perform similarly for new data from the same problem. However, many applications generate data whose characteristics can change over time. Therefore, a suitable bias at a given time may become inappropriate at another time. Although meta-learning has been used to continuously select a learning algorithm in data streams, data characterization has received less attention in this context. In this study, we provide a set of guidelines to support the proposal of characteristics able to describe non-stationary data over time. This guidance considers both the order of arrival of the examples and the type of variables involved in the base-level learning. In addition, we analyze the influence of characteristics regarding their dependence on data morphology. Experimental results using real data streams showed the effectiveness of the proposed data characterization general scheme to support algorithm selection by meta-learning systems. Moreover, the dependent metafeatures provided crucial information for the success of some meta-models. [ABSTRACT FROM AUTHOR]
    • Abstract:
      Copyright of Intelligent Data Analysis is the property of IOS Press and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
    • Author Affiliations:
      1Universidade Estadual Paulista (UNESP), Campus de Itapeva, São Paulo, Brazil
      2Universidade Federal do Maranhão, Campus de São Luís, Maranhão, Brazil
      3INESC TEC, Faculdade de Engenharia da Universidade do Porto, Universidade do Porto, Porto, Portugal
      4Instituto de Ciências Matemáticas e de Computação, Universidade de São Paulo, São Carlos, Brazil
    • ISSN:
      1088-467X
    • Accession Number:
      10.3233/IDA-160083
    • Accession Number:
      124805302
  • Citations
    • ABNT:
      ROSSI, A. L. D. et al. A guidance of data stream characterization for meta-learning. Intelligent Data Analysis, [s. l.], v. 21, n. 4, p. 1015–1035, 2017. DOI 10.3233/IDA-160083. Disponível em: http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=asn&AN=124805302&custid=s8280428. Acesso em: 10 dez. 2019.
    • AMA:
      Rossi ALD, de Souza BF, Soares C, de Leon Ferreira de Carvalho ACP. A guidance of data stream characterization for meta-learning. Intelligent Data Analysis. 2017;21(4):1015-1035. doi:10.3233/IDA-160083.
    • APA:
      Rossi, A. L. D., de Souza, B. F., Soares, C., & de Leon Ferreira de Carvalho, A. C. P. (2017). A guidance of data stream characterization for meta-learning. Intelligent Data Analysis, 21(4), 1015–1035. https://doi.org/10.3233/IDA-160083
    • Chicago/Turabian: Author-Date:
      Rossi, André Luis Debiaso, Bruno Feres de Souza, Carlos Soares, and André Carlos Ponce de Leon Ferreira de Carvalho. 2017. “A Guidance of Data Stream Characterization for Meta-Learning.” Intelligent Data Analysis 21 (4): 1015–35. doi:10.3233/IDA-160083.
    • Harvard:
      Rossi, A. L. D. et al. (2017) ‘A guidance of data stream characterization for meta-learning’, Intelligent Data Analysis, 21(4), pp. 1015–1035. doi: 10.3233/IDA-160083.
    • Harvard: Australian:
      Rossi, ALD, de Souza, BF, Soares, C & de Leon Ferreira de Carvalho, ACP 2017, ‘A guidance of data stream characterization for meta-learning’, Intelligent Data Analysis, vol. 21, no. 4, pp. 1015–1035, viewed 10 December 2019, .
    • MLA:
      Rossi, André Luis Debiaso, et al. “A Guidance of Data Stream Characterization for Meta-Learning.” Intelligent Data Analysis, vol. 21, no. 4, July 2017, pp. 1015–1035. EBSCOhost, doi:10.3233/IDA-160083.
    • Chicago/Turabian: Humanities:
      Rossi, André Luis Debiaso, Bruno Feres de Souza, Carlos Soares, and André Carlos Ponce de Leon Ferreira de Carvalho. “A Guidance of Data Stream Characterization for Meta-Learning.” Intelligent Data Analysis 21, no. 4 (July 2017): 1015–35. doi:10.3233/IDA-160083.
    • Vancouver/ICMJE:
      Rossi ALD, de Souza BF, Soares C, de Leon Ferreira de Carvalho ACP. A guidance of data stream characterization for meta-learning. Intelligent Data Analysis [Internet]. 2017 Jul [cited 2019 Dec 10];21(4):1015–35. Available from: http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=asn&AN=124805302&custid=s8280428