Banking chatbot (Rasa framework-based)

Features

• Intent analysis and the ability to extract and save all the entities during the conversation with a user
• Communication on banking topics such as mortgages, deposit accounts and credit cards
• The ability to estimate user creditworthiness by means of database appending with the information that a user provides
• Automatic fullfilment of the official documents that are necessary for mortgage application
• Custom actions that can form queries from user messages and return the documents by index and query using ElasticSearch

1. Download the repository and install everything from requirements.txt (pip install -r requirements.txt)

*For RASA NLU:

  git clone https://github.com/RasaHQ/rasa_nlu.git
  cd rasa_nlu
  pip install -r requirements.txt
  pip install -e .

Spacy:

  pip install rasa_nlu[spacy]
  python -m spacy download en_core_web_md
  python -m spacy link en_core_web_md en

RASA Core:

  pip install rasa_core

2. Use the next command to launch the bot in your console:

  make launch 

*(don't forget to do it from the folder RASA_banking_chatbot)

3. Interactive learning

  make interactive

4. Graph visualization

  make visualize

5. Activation of actions

  make activate_actions

*point your host in the code actions.py to get access to your ElasticSearch server

You are welcome to broaden the bot by adding actions and intents and forming scenarios of communication in stories.md file.