An AI Agent whose primary language is swahili trained on KOT tweets.
# π KOT-AI - Kenyan Digital Linguistics Explorer
KOT-AI is an innovative digital linguistics research tool that explores and documents the evolution of Sheng - Kenya's dynamic urban language that blends Swahili, English, and local dialects. Using advanced AI and natural language processing, the project creates a living archive of contemporary East African urban linguistics.
## π― Project Overview
KOT-AI continuously analyzes trending Kenyan tweets to:
- Track emerging Sheng expressions and their meanings
- Document usage patterns and cultural context
- Create a searchable database of contemporary urban language
- Generate creative content that demonstrates language evolution
## β¨ Key Features
- **Real-time Language Analysis**: Monitors and processes trending Kenyan tweets
- **Pattern Recognition**: Identifies new Sheng words and expressions
- **Context Mapping**: Documents usage patterns and cultural significance
- **Engagement Tracking**: Measures the popularity and spread of expressions
- **Creative Generation**: Produces culturally-aware, creative content
- **Research Database**: Maintains a structured repository of findings
## π οΈ Tech Stack
- **Runtime**: Node.js
- **AI Processing**: Claude 3.5 (Anthropic)
- **Data Source**: Twitter API v2
- **Database**: MongoDB
- **Logging**: Winston
- **API Layer**: Express.js
## π Getting Started
### Prerequisites
```bash
Node.js >= 18.0.0
MongoDB >= 4.4
npm >= 8.0.0
```
### Environment Setup
1. Clone the repository:
```bash
git clone
github.com
cd KOT-AI
```
2. Create `.env` file:
```env
ANTHROPIC_API_KEY=your_anthropic_key
TWITTER_API_KEY=your_twitter_key
TWITTER_API_SECRET=your_twitter_secret
TWITTER_ACCESS_TOKEN=your_access_token
TWITTER_ACCESS_SECRET=your_access_secret
MONGODB_USER=your_mongodb_username
MONGODB_PASS=your_mongodb_password
```
3. Install dependencies:
```bash
npm install
```
4. Start the application:
```bash
npm run dev # Development mode
npm start # Produ β¦