Detecting and Categorizing Indices in Lecture Video Using Supervised Machine Learning

  • Christopher Brooks
  • G. Scott Johnston
  • Craig Thompson
  • Jim Greer
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7884)

Abstract

This work reports on the evaluation of detecting scene transitions in lecture video through supervised machine learning. It expands on previous work by gathering training data from multiple human raters. We include a robust evaluation that compares predictions against the entire set of expert classifications in disagreement. Finally, we explore some of the issues around constructing training data from multiple human experts, specifically emphasizing that evaluation strategies should be carefully considered when using aggregated training data.

References

  1. 1.
    Dickson, P., Adrion, W., Hanson, A.: Automatic Capture of Significant Points in a Computer Based Presentation. In: Eighth IEEE International Symposium on Multimedia (ISM 2006), pp. 921–926 (2006)Google Scholar
  2. 2.
    Brooks, C., Amundson, K.: Detecting Significant Events in Lecture Video using Supervised Machine Learning. In: 2009 Conference on Artificial Intelligence in Education (2009)Google Scholar
  3. 3.
    Fleiss, J.L.: Measuring nominal scale agreement among many raters. Psychological Bulletin 76(5), 378–382 (1971)CrossRefGoogle Scholar
  4. 4.
    Landis, J.R., Koch, G.G.: The Measurement of Observer Agreement for Categorical Data. Biometrics 33(1), 159–174 (1977)MathSciNetMATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Christopher Brooks
    • 1
  • G. Scott Johnston
    • 1
  • Craig Thompson
    • 1
  • Jim Greer
    • 1
  1. 1.Department of Computer ScienceUniversity of SaskatchewanSaskatoonCanada

Personalised recommendations