Unsupervised Feature Selection

Unsupervised Feature Selection is a feature selection paradigm that identifies and retains the most relevant subset of features from high-dimensional data without relying on labeled outcomes, typically by leveraging intrinsic data structures, statistical dependencies, or geometric properties to preserve maximal information or discriminative capacity inherent in the unlabeled data.

1.4K

Publications

99.3K

Citations

3.9K

Authors

1.4K

Institutions

Publications per year

2017–2026

777

Authors

3.9K

  1. FN
    Feiping Nie

    Northwestern Polytechnical University, China

    34 publications · 3.2K citations

    27

    h-index

  2. XL
    Xuelong Li

    Northwestern Polytechnical University, China

    26 publications · 1.3K citations

    20

    h-index

  3. YT
    Y‐h. Taguchi

    Chuo University, Japan

    17 publications · 466 citations

    14

    h-index

  4. TL
    Tianrui Li

    Southwest Jiaotong University, China

    12 publications · 731 citations

    13

    h-index

  5. HL
    Huan Liu

    University of Wollongong, Australia

    14 publications · 2.8K citations

    13

    h-index

1–5 of 3.9K

Venues

  1. Expert Systems with Applications

    58 publications · 4.9K citations

    36

    h-index

  2. Pattern Recognition

    46 publications · 4.2K citations

    36

    h-index

  3. Knowledge-Based Systems

    45 publications · 2.9K citations

    30

    h-index

  4. Neurocomputing

    45 publications · 4K citations

    29

    h-index

  5. IEEE Transactions on Knowledge and Data Engineering

    TKDE · 27 publications · 2.4K citations

    25

    h-index

1–5