metadata
license: mit
task_categories:
- text-classification
language:
- en
pretty_name: sunny thakur
size_categories:
- n<1K
Phishing and Benign Email Dataset
This dataset contains a curated collection of phishing and legitimate (benign) emails for use in cybersecurity training, phishing detection models, and email classification systems. Each entry is structured with subject, body, intent, technique, target, and classification label.
📁 Dataset Format
The dataset is stored in .jsonl (JSON Lines) format. Each line is a standalone JSON object.
Fields:
| Field | Description |
|---|---|
id |
Unique email ID |
subject |
Email subject line |
body |
Full email content |
intent |
Purpose of the email (e.g., credential harvesting, malware delivery) |
technique |
Social engineering or technical trick used |
target |
Who or what is being impersonated |
spoofed_sender |
Forged email address |
label |
"phishing" or "benign" |
🔍 Example Entry
{
"id": "phish-0003",
"subject": "Unusual Activity Detected",
"body": "Suspicious login from Russia. Reset password now: https://g00gle-security.com",
"intent": "Credential Harvesting",
"technique": "Homoglyph Link Spoofing",
"target": "Google",
"spoofed_sender": "no-reply@g00gle.com",
"label": "phishing"
}
✅ Use Cases
Train NLP models to detect phishing content
Classify or filter email threats in spam filters
Educate users and red teams on phishing techniques
Simulate email-based social engineering scenarios
📊 Dataset Summary
🔐 Phishing Emails: ~100 entries
✅ Benign Emails: ~100 entries
⚖️ Balanced and ready for binary classification or LLM pretraining
⚠️ Disclaimer
This dataset is for research and educational purposes only. Do not use these emails in live environments. The authors are not responsible for misuse.
📜 License
This dataset is released under the MIT License.
📫 Contributions
Feel free to submit pull requests to add more phishing or legit emails. The more diverse, the better for training secure systems.