Phishing Indicator Extraction

Phishing Indicator Extraction is a computational method that identifies and isolates specific artifacts or patterns from digital communications to detect potential phishing attempts by distinguishing malicious intent from benign content.

10

Publications

567

Citations

32

Authors

15

Institutions

Publications per year

2017–2026

6

Authors

32

  1. SA
    Shadi AlZu’bi

    Al-Zaytoonah University of Jordan, Jordan

    1 publications · 142 citations

    1

    h-index

  2. KD
    Keshav Dahal

    University of the West of Scotland, United Kingdom

    1 publications · 67 citations

    1

    h-index

  3. JJ
    Jeremy Jacob

    University of York, United Kingdom

    1 publications · 27 citations

    1

    h-index

  4. BJ
    B. Janet

    National Institute of Technology Tiruchirappalli, India

    1 publications · 25 citations

    1

    h-index

  5. AM
    Ala Mughaid

    Hashemite University, Jordan

    1 publications · 142 citations

    1

    h-index

1–5 of 32

Venues

  1. Computer Speech & Language

    1 publications · 65 citations

    1

    h-index

  2. Cluster Computing

    1 publications · 142 citations

    1

    h-index

  3. Cognitive Computation

    1 publications · 67 citations

    1

    h-index

  4. Human-centric Computing and Information Sciences

    HCIS · 1 publications · 147 citations

    1

    h-index

  5. Peer-to-Peer Networking and Applications

    1 publications · 25 citations

    1

    h-index

1–5