https://github.com/blackiq/qandai

QandAI is a chatbot that uses artificial intelligence (AI) to answer user questions. It is built using Python, Flask, and scikit-learn, and can be easily customized to suit your needs.
https://github.com/blackiq/qandai

api chatbot flask machine-learning ml python question-answering

Last synced: 17 days ago
JSON representation

QandAI is a chatbot that uses artificial intelligence (AI) to answer user questions. It is built using Python, Flask, and scikit-learn, and can be easily customized to suit your needs.

• Host: GitHub
• URL: https://github.com/blackiq/qandai
• Owner: BlackIQ
• License: mit
• Created: 2023-03-18T16:56:19.000Z (almost 2 years ago)
• Default Branch: main
• Last Pushed: 2023-05-04T19:11:26.000Z (over 1 year ago)
• Last Synced: 2024-11-13T00:31:50.657Z (3 months ago)
• Topics: api, chatbot, flask, machine-learning, ml, python, question-answering
• Language: Python
• Homepage: https://chat.amirhossein.info/api
• Size: 433 KB
• Stars: 2
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# QandAI - Your Question and Answer AI Chatbot

## Introduction

QandAI is a chatbot that uses artificial intelligence (AI) to answer user questions. It is built using Python, Flask, and scikit-learn, and can be easily customized to suit your needs.

## Features

QandAI has the following features:

- Simple and intuitive interface for asking questions

- Uses a Multinomial Naive Bayes model to predict the most likely answer to a given question

- Can be easily trained on new data to improve its accuracy

- Returns the top 3 most likely answers, ranked by probability

- API endpoint for programmatic access to the chatbot

- Includes sample data for testing and training

## Getting Started

### Requirements

To use QandAI, you will need:

- Python 3.x

- Flask

- scikit-learn

- pandas

- joblib

- pipenv

### Installation

To install QandAI, follow these steps:

1. Clone the repository from GitHub: `git clone https://github.com/BlackIQ/QandAI.git`.

2. Navigate to the project directory: `cd QandAI`.

3. Install the required dependencies: `pipenv install`.

### Usage

To use QandAI, follow these steps:

1. Copy `.env.example` to `.env` and change `API_PORT` to your custome port.

2. Train the model on your own data or use the included sample data: `python3 app/core/core.py`

3. Start the Flask server: `python3 wsgi.py`

4. Send a POST request to the `/api/predict` endpoint with a JSON payload containing a `question` key and the user's question as the value.

5. The server will return a JSON response containing the top 3 most likely answers, ranked by probability.

### Customization

To customize QandAI, you can modify the following:

- `data/faq_data.json`: Add or remove questions and answers to train the model on your own data

- `app/core/core.py`: Modify the code for preprocessing and vectorizing the input data

- `models/faq_model.joblib`: Train and save a new model with different hyperparameters or a different algorithm

## License

This project is licensed under the MIT License - see the LICENSE file for details.

## Acknowledgments

QandAI was inspired by the many open-source chatbot projects available online. Thank you to the developers and contributors of scikit-learn, Flask, and joblib for their excellent libraries.