A chat bot microservice app that can be trained on different datasets.

• python 3.x
  • Rasa NLU
  • MITIE
  • scikit-learn
  • Flask
  • Flask-cors
• virtualenv
• gunicorn
• nginx
• Rasa NLU Trainer

1. Install python 3.6 or above
2. Create a virtual environment to install dependencies $ virtualenv venv
3. Enter the virtual environment $ source venv/bin/activate
4. Install dependencies $ pip install -r requirements.txt

1. Use Rasa NLU Trainer to generate training.json.
2. Place it at ./data/training.json
3. run ./chatai.sh train
4. Create ./data/answers.json.
5. Use intents in the training data as key and value as answers.

• Run ./chatBot start