Phishing Website Classification Data

Phishing website dataset for machine learning classification. The dataset's specific size, creator, and update history are not documented. It originates from the Kaggle platform.

Use Cases

• Train a binary classifier to predict the 'phishing' label using website URL and HTML features.
• Perform feature importance analysis on attributes like 'url_length', 'ssl_final_state', or 'having_sub_domain' to identify key phishing indicators.
• Benchmark model performance for phishing detection tasks using the provided labeled examples.
• Develop ensemble models that combine lexical URL features with page content features for improved classification accuracy.

Strengths

• Dataset is specifically curated for a well-defined machine learning classification task.
• Includes labeled examples essential for supervised learning models.

Limitations

• Unknown row count prevents assessment of statistical power for model training.
• Unknown feature set and column names limit understanding of data scope and potential biases.
• Lack of provenance details, such as collection method and time range, reduces reproducibility.

Provenance

Source

Kaggle