Book Recommendation system

View the notebook in kaggle

This is a book recommender system which uses the unsupervised learning algorithm - Nearest Neighnours to predict next book to read

---

Deploying it locally

Steps to run it locally

1. git clone the repository

https://github.com/pratham-saraf/book-recommender.git

2. Download the Data Files
3. Move the downloaded Data files inside data folder which should be created inside model folder
4. Folder structure would look like

.
├── app.py
├── model
│   ├── data
│   │   ├── book_id_map.csv
│   │   ├── recommedation_model.sav
│   │   ├── search_final.json
│   │   └── sparse_matrix.npz
│   ├── recommender.py
│   └── search.py
├── README.md
├── requirements.txt
├── run
├── static
│   ├── dashboard.js
│   ├── find.css
│   ├── find.js
│   ├── images
│   │   ├── dislike.svg
│   │   └── like.svg
│   ├── nav.css
│   ├── recommendation.js
│   ├── signin.css
│   └── starter-template.css
└── templates
    ├── dashboard.html
    ├── find.html
    ├── index.html
    ├── layout.html
    ├── login.html
    ├── recommend.html
    └── signup.html

5. Run the command to create the virtual env and source it

cd book-recommender
python3 -m venv env
source env/bin/activate

6. To install the python dependencies

pip install -r requirements.txt

7. Create a .env file which contains

SECRET_KEY="A random password string for the flask"
MONGO_DB = "Mongo-DB-connection-URI"

8. The mongodb cluster should have two collections named

└── Cluster
    ├── user-book-data
    └── users

9. Give run permissions to the run file and execute it

chmod +x run
./run

---