Semi-Supervised Learning Algorithm

• A Semi-Supervised Learning Algorithm is a Supervised Learning Algorithm that is applicable to a Semi-Supervised Machine Learning Task.
  • Context:
    • Can make use of [[Training_Cases?]] that lack [[Target?]] data
    • Can achieve higher Accuracy than Supervised Learning Algorithm.
  • Examples:
    • [[EM?]] with generative mixture models
    • [[Self-Training_Learning_Algorithm?]]
    • [[Co-training_Learning_Algorithm?]]
    • [[Transductive_Support_Vector_Machines?]]
    • [[Graph-based_Learning_Algorithm?]]
    • Bootstrapping
  • See: Unsupervised Learning Algorithm, Bootstrapping

---

References

• M. Seeger. (2001). Learning with labeled and unlabeled data. Technical Report, University of Edinburgh.
• O. Chapelle, A. Zien, and B. Scholkopf. 2006. Semi-supervised learning. Editors. MIT Press.
• X. Zhu. (2006). Semi-Supervised Learning Literature Survey. Technical Report - University of Wisconsin – Madison. (http://www.cs.wisc.edu/~jerryzhu/pub/ssl_survey.pdf">paper.pdf)