TARDIS AI Chatbot

The TARDIS AI Chatbot is a simple web-based chatbot application that uses natural language processing (NLP) to provide answers based on a set of predefined documents. The chat history is displayed on the web page, and users can rate the chatbot's answers.

Chatbot design

Design_white Design_black

Features

  • User-friendly web interface.
  • Chat history display with user inputs and chatbot responses.
  • Ability to rate chatbot answers.
  • Ability to create Jira Ticket automatically.
  • API Documentation in form of Swagger

Parts of solution

  • Initialization
  • Reading documents
  • Text preprocessing (Spacy and Porter Stemmer)
  • Building a knowledge base
  • TF-IDF Vectorization
  • Similarity Extraction
  • Information, Entity and Intent Extraction
  • Result as output

Prerequisites

Before running the application, make sure you have the following installed:

  • Python 3.x
  • Flask (pip install Flask)
  • spaCy (pip install spacy)
  • scikit-learn (pip install scikit-learn)
  • en_core_web_md (python -m spacy download en_core_web_md)

Structure

main.py: The main Flask application file. docs/: Folder containing the documents used for chatbot responses. static/: Folder containing static assets such as images. templates/: Folder containing HTML templates for rendering web pages.

Starting app - via terminal

cd "2-CORECHI" ##changing directory to current project souce venv/bin/activate ##starting virtual python enviroment to run flask python3 main.py ##starting app

Closing app and virtual enviroment

CTRL + C deactivate

Localhost address

http://127.0.0.1:5000