CCO - 729 - 7 - Topics in Artificial Intelligence: Learning in Data Streams

Total of Credits: 8
Hours for Theoretical Classes: 60
Hours for Exercises or Seminars: 60

Objective

To offer the student the conditions to:

  • Acquire basic concepts about conventional and data stream Machine Learning with emphasis on the differences between these two types of learning;
  • Know different machine learning algorithms for data streams and be able to identify the most appropriate class of algorithms for each situation, especially with regard to supervision;
  • Understand the role of Computational Intelligence methodologies in the context of data streams.

Catalog Description

  • Machine Learning: types of supervision (supervised, unsupervised, semi-supervised), paradigms (classic, evolutive, neural, statistical)
  • Data Streams: basic concepts
  • Tree-based algorithms for data streams
  • Clustering algorithms for data streams
  • Computational Intelligence Methodologies in the context of data streams: evolutionary computation, fuzzy systems, neural computation

Main Bibliography

  1. THEODORIDIS, Sergios, 1951; KOUTROMBAS, Konstantinos. Pattern recognition. 4 ed. Burlington: Elsevier, c2009. 961 p. ISBN 978-1-59749-272-0.
  2. GAMA, J. Knowledge Discovery from Data Streams. [S.l.]: Chapman and Hall, 2010. 255 p.
  3. WITTEN, I., FRANK, E., HALL, M. Data Mining – Practical Machine Learning Tools and Techniques. Morgan Kaufmman, 2011.
  4. AGGARWAL, C. C. An Introduction to Data Streams. In: AGGARWAL, C. C. (Ed.). Data Streams. Boston, MA: Springer US, 2007. p. 1–8. 4. THEODORIDIS, Sergios, 1951; KOUTROMBAS, Konstantinos. Pattern recognition. 4 ed. Burlington: Elsevier, c2009. 961 p. ISBN 978-1-59749-272-0.
  5. DOMINGOS, P.; HULTEN, G. Mining high-speed data streams. In: Proceedings of the sixth ACM SIGKDD international conference on Knowledge discovery and data mining - KDD ’00, 2000. p. 71–80. 6. SILVA, J. A. et al. Data stream clustering: A survey. ACM Computing Surveys, v. 46, n. 1, p. 1–31, oct 2013.