A new cross-training approach by using labeled data

ABSTRACT

We propose a new cross-training based learning algorithm in this paper. This algorithm generates three classifiers based on the three subsets of original labeled and unlabeled training set. The proposed algorithm is evaluated using data from the UCI repository by the experiment. Experimental results show that our algorithm can improve classification accuracy compared to those of other algorithms.

REFERENCES

• 1: X. Zhu, "Semi-Supervised Learning Literature Survey," Computer Science, University of Wisconsin-Madison, 2007.
• 2: Avrim Blum , Tom Mitchell, Combining labeled and unlabeled data with co-training, Proceedings of the eleventh annual conference on Computational learning theory, p.92-100, July 24-26, 1998, Madison, Wisconsin, United States [doi>10.1145/279943.279962]
• 3: Nuanwan Soonthornphisaj , Boonserm Kijsirikul, Iterative cross-training: An algorithm for learning from unlabeled Web pages, International Journal of Intelligent Systems, v.19 n.1-2, p.131-147, January 2004 [doi>10.1002/int.v19:1/2]
• 4: C. L. Blake and C. J. Merz, "UCI repository of machine learning databases," 1998.
• 5: J. Ross Quinlan, C4.5: programs for machine learning, Morgan Kaufmann Publishers Inc., San Francisco, CA, 1993
• 6: M. J. L. Orr, "Introduction to radial basis function networks," Center for Cognitive Science, University of Edinburgh, Scotland, 1996.

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE
      I.2.6 Learning
          Subjects: Knowledge acquisition

Additional Classification:
  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE
      I.2.6 Learning
          Subjects: Parameter learning
  I.5 PATTERN RECOGNITION
      I.5.2 Design Methodology
          Subjects: Classifier design and evaluation

General Terms:
Algorithms, Experimentation, Theory

Keywords:
co-training, cross-training, semi-supervised learning