Swahili-English Code-Mixed SMS Scam Dataset: A Multilingual Resource for Fraud Detection in East Africa
# π¨ Kenya Scam Database Collection Platform
A community-driven platform for collecting and documenting scam messages to protect Kenyan citizens from fraud.
## π― About
This platform allows users to anonymously submit scam messages they receive via SMS, WhatsApp, email, or other channels. The data collected helps build a comprehensive database for research and public awareness.
### ποΈ Research Institution
**Daystar University - Institute of Data Science, AI, and Cybersecurity**
### π Live Platform
Visit:
https://[your-username].
github.io
## π― Purpose
- π€ Build datasets for scam detection AI models
- π‘οΈ Protect communities through awareness
- π¬ Support cybersecurity research in Kenya
- π Document scam patterns targeting African populations
- π Improve tokenization for African languages in AI
## π How to Use
1. Visit the platform
2. Copy-paste the scam message you received
3. Select the type of scam (SMS, WhatsApp, etc.)
4. Add optional context (location, sender info)
5. Submit anonymously - your data helps protect others!
## π Data Privacy & Security
- β
No personal information collected
- β
All submissions anonymized
- β
Data used only for research and public safety
- β
Compliant with data protection regulations
- β
Real-time Firebase security rules
## π Current Statistics
- **Total Reports**: Updated in real-time
- **Languages**: English, Kiswahili, Mixed
- **Categories**: Lottery, Financial, Loan, Job, Romance scams
## π οΈ Technical Stack
- **Frontend**: HTML5, CSS3, JavaScript (ES6+)
- **Database**: Firebase Firestore
- **Hosting**: GitHub Pages
- **Real-time**: Firebase real-time listeners
## π€ Contributing
- Report bugs through GitHub issues
- Suggest improvements
- Share with your community
- Partner with us for research
## π§ Contact
**Daystar University - Institute of Data Science, AI, and Cybersecurity**
- Email: jmursi@daystar.ac.ke
- Location: Nairobi, Kenya
## π Research Impact
This platform supports:
- PhD/Masters research on β¦