Dataset · zefang-liu
Email database for phishing detection
zefang-liu/phishing-email-dataset
English email dataset for training and testing phishing filters.
What's inside
Contains English email texts for classification into phishing and legitimate categories. The corpus size ranges from 10,000 to 100,000 records. The dataset is a direct copy of the Phishing Email Detection dataset from Kaggle by user Cyber Cop and is distributed under the GNU LGPL 3.0 license.
What you can do with it
- Training a classifier for filtering phishing campaigns
- Testing corporate email security systems
- Fine-tuning compact text models for fraud detection
- Tasks
- Text classification
- Languages
- English
- Size
- 10K–100K rows
- Downloads this month
- 2K
- Likes
- 40
Models trained on it
- CrabInHoney/urlbert-tiny-v5Embeddings 1.4K
- GerritPot/COS720-Project-PhishingText classification 891
- ElSlay/BERT-Phishing-Email-ModelText classification 857
- mradermacher/jellyphish-bert-base-mail-GGUF 569
- IntegralBender/distilbert-spam-detectorText classification 153
- Aaqib099/distilbert-phishing-detectionText classification 70
- Aaqib099/gpt2-phishing-detectionText classification 64
- vete-speis/gemma-4-E2B-it-phishingText generation 31
How to load it
from datasets import load_dataset
ds = load_dataset("zefang-liu/phishing-email-dataset")