BASSUM: A Bayesian semi-supervised method for classification feature selection

@article{Cai2011BASSUMAB,
  title={BASSUM: A Bayesian semi-supervised method for classification feature selection},
  author={Ruichu Cai and Z. Zhang and Z. Hao},
  journal={Pattern Recognit.},
  year={2011},
  volume={44},
  pages={811-820}
}
  • Ruichu Cai, Z. Zhang, Z. Hao
  • Published 2011
  • Mathematics, Computer Science
  • Pattern Recognit.
  • Feature selection is an important preprocessing step for building efficient, generalizable and interpretable classifiers on high dimensional data sets. Given the assumption on the sufficient labelled samples, the Markov Blanket provides a complete and sound solution to the selection of optimal features, by exploring the conditional independence relationships among the features. In real-world applications, unfortunately, it is usually easy to get unlabelled samples, but expensive to obtain the… CONTINUE READING

    Figures, Tables, and Topics from this paper.

    Hypothesis Testing and Feature Selection in Semi-Supervised Data
    • 3
    Simple strategies for semi-supervised feature selection
    • 15
    • Highly Influenced
    • Open Access
    Markov Blanket Discovery in Positive-Unlabelled and Semi-supervised Data
    • 8
    • Highly Influenced
    • Open Access
    Non-parametric feature selection for machine learning in complex settings
    • 2
    BAMB: A Balanced Markov Blanket Discovery Approach to Feature Selection
    • 6
    A graph Laplacian based approach to semi-supervised feature selection for regression problems
    • 47
    • Open Access
    Causality-based Feature Selection: Methods and Evaluations
    • 4
    • Highly Influenced
    • Open Access
    SADA: A General Framework to Support Robust Causation Discovery
    • 29
    • Open Access

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 20 REFERENCES
    Semi-supervised Feature Selection via Spectral Analysis
    • 245
    • Highly Influential
    • Open Access
    Feature selection with dynamic mutual information
    • 307
    Wrappers for Feature Subset Selection
    • 7,464
    • Open Access
    Toward Optimal Feature Selection
    • 1,588
    • Highly Influential
    • Open Access
    Spectral feature selection for supervised and unsupervised learning
    • 633
    • Open Access
    LS Bound based gene selection for DNA microarray data
    • 86
    • Open Access
    HITON: A Novel Markov Blanket Algorithm for Optimal Variable Selection
    • 249
    • Highly Influential
    • Open Access
    Time and sample efficient discovery of Markov blankets and direct causal relations
    • 234
    • Highly Influential
    • Open Access
    Semi-supervised feature selection via multiobjective optimization
    • 32
    • Highly Influential