Website Phishing Data for Tabular ML Classification, 2014

A tabular dataset curated for evaluating predictive models on independent and identically distributed data, with the intended task of classification. The data originates from a 2014 study on phishing detection using associative classification data mining and is licensed under CC BY 4.0.

Use Cases

• Benchmarking classification algorithms based on the dataset's stated focus for IID tabular data evaluation.
• Training phishing detection models based on the dataset's domain of web security and cybersecurity.
• Studying the impact of data duplication on model performance based on the noted anomaly of many duplicates.

Strengths

• Explicitly curated for evaluating ML models on IID tabular classification tasks.
• Licensed under the permissive CC BY 4.0 license.
• Original data source and reference are provided for citation and verification.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file formats, and sample data are unknown, which may limit suitability assessment.
• The data contains an anomaly of many duplicates (~50%), which could affect model training.

Provenance

Source

https://doi.org/10.24432/C5B301; Abdelhamid, Neda, Aladdin Ayesh, and Fadi Thabtah. 'Phishing detection based associative classification data mining.' Expert Systems with Applications 41.13 (2014): 5948-5959.

Collection Method

Curated by the TabArena team from the original source for a tabular ML study.

Time Range

2014

Freshness

Dataset year is 2014; last update date is unknown.

Geography

null