Corpus ID: 76667408

DEDPUL: Method for Positive-Unlabeled Learning based on Density Estimation

@inproceedings{Ivanov2019DEDPULMF,
  title={DEDPUL: Method for Positive-Unlabeled Learning based on Density Estimation},
  author={Dmitry Ivanov},
  year={2019}
}
  • Dmitry Ivanov
  • Published 2019
  • Mathematics
  • Positive-Unlabeled Classification is an analog of binary classification for the case when the Negative (N) sample in the training set is contaminated with latent instances of the Positive (P) class and hence is Unlabeled (U). We develop DEDPUL, a novel method that simultaneously solves two problems concerning U: estimates the proportions of the mixing components (P and N) in U and classifies U. We conduct experiments on synthetic and real-world data and show that DEDPUL outperforms current… CONTINUE READING

    Figures and Tables from this paper

    References

    SHOWING 1-10 OF 35 REFERENCES
    Class-prior estimation for learning from positive and unlabeled data
    • 82
    • PDF
    Estimating the class prior and posterior from noisy positives and unlabeled data
    • 63
    • Highly Influential
    • PDF
    Analysis of Learning from Positive and Unlabeled Data
    • 157
    • PDF
    Positive-Unlabeled Classification under Class Prior Shift and Asymmetric Error
    • 5
    • PDF
    Convex Formulation for Learning from Positive and Unlabeled Data
    • 129
    • PDF