adityapradhan202 / Grimoire-Guide

Grimoire Guide is a book recommendation system which leverages statistical machine learning and NLP concepts to recommend the best books to the users. Just provide a plot or description, and it will do the rest of the job for you.

Repository from Github https://github.comadityapradhan202/Grimoire-GuideRepository from Github https://github.comadityapradhan202/Grimoire-Guide

Grimoire-Guide

Grimoire Guide is an AI powered book recommendation system which leverages statistical machine learning, data science and NLPconcepts to recommend the best books to the users. Just provide a plot or description, and it will do the rest of the job for you.

Guide for running this on local machine:

1. Clone the repository:
Use the command given below in the git bash terminal:
git clone https://github.com/adityapradhan202/Grimoire-Guide.git

2. Install the dependencies:
Open a new terminal in the project folder and type this command:
pip install -r requirements.txt

3. Run the app:
In the terminal of the project folder type this command to run the app on local host:
streamlit run app.py

How it works:

Grimoire-Guide uses a simple concept of 'cosine similarity', which is available in scikit-learn. We can convert two separate texts into vectors or numbers and then we can use the concept of cosine similarity to see how similar the two texts are.

Further improvements:

This is more like a practice project to understand and implement the concept of cosine similarity. Further, we can code neural networks on bigger dataset and create more robust ML application.

App images:

img img2

About contribution:

It would be great if you could contribute to and improve 'Grimoire-Guide'. Please create a separate feature branch with your own code and submit a pull request. If your code adds value to the project, your pull request will be reviewed and accepted. 😊

About

Grimoire Guide is a book recommendation system which leverages statistical machine learning and NLP concepts to recommend the best books to the users. Just provide a plot or description, and it will do the rest of the job for you.

License:MIT License


Languages

Language:Jupyter Notebook 92.9%Language:Python 7.1%