Chatbot Web Application

Overview

This project is a web-based chatbot built using Flask, MongoDB, and TF-IDF based similarity matching.
The chatbot answers user queries by finding the most similar question in the database.
Queries that cannot be answered are logged for future improvements.

---

Features

• Flask-based web API for chatbot interaction
• Stores questions and answers in MongoDB
• Uses TF-IDF vectorization and cosine similarity for query matching
• Handles unanswered queries automatically
• CORS enabled for front-end integration

---

Project Structure

• app.py — Main Flask application
• mongodb.py — Script to insert QA pairs into MongoDB
• requirements.txt — Python dependencies
• README.md — Project documentation

---

Installation

Prerequisites

• Python 3.8+
• MongoDB Atlas account or local MongoDB instance
• pip package manager

Clone the repository

Open your terminal and run:

git clone https://github.com/experimentalsolution/chatbot-experimental-solution.git 
cd my-project

Install dependencies

pip install -r requirements.txt

Setup MongoDB

• Create a database named chatbot_db.
• Create collections:
• qa_pairs → standard QA pairs
• qa_pairs_experimental_solution → experimental QA pairs
• qa_pairs_not_answered → unanswered queries
• Update the MongoDB URI in app.py and mongodb.py:
• MONGO_URI = "your_mongodb_connection_string"

Insert initial QA pairs

python mongodb.py

Running the Application

Start the Flask server:

python app.py

Default server: http://127.0.0.1:5000/ POST request to /chatbot with JSON payload:

{ "query": "Hello" }

Response example:

{ "response": "Morning, Afternoon, Evening to whatever time zone you are! How can I help you?" }

How It Works

• Fetches all questions and responses from MongoDB.
• Converts questions into vectors using TF-IDF.
• Compares the user query to stored questions with cosine similarity.
• Returns the matched response if similarity > 50%.
• Logs queries that cannot be answered to qa_pairs_not_answered.

Future Improvements

• Add machine learning models for better query understanding
• Implement intent classification
• Build a frontend UI for live interaction
• Enable real-time learning from unanswered queries

Dependencies

• Flask
• Flask-CORS
• pymongo
• scikit-learn

Install dependencies:

pip install flask flask-cors pymongo scikit-learn

Demo

Live demo is created here on a angular front end .It is deployed . Please check it out and share your feedback .

License

MIT License