Download the dataset on Kaggle - Book Recommendation Dataset
There are several csv on that dataset:
- Book's lists dataset
- Book's users dataset
- Book's Ratings dataset
Total of the dataset if you merge them is 1.149.780 rows and 12 columns but there are also several missing values and several duplicate book's title, therefore I handle it by remove them and I got 205.170 rows and 12 columns but In this case I only used 10.000 dataset because my computer can not handle more dataset, but you can used dataset as much as you want.
- download the clean books dataset on books_data.csv
- download cosine similarity on cosine_similarity.csv
- download requirements.txt with pip install -r requirements.txt
- run python3 app.py
- Test on your postman with method POST and endpoint /books_recommendations
- Use key on your body as book_title and type your favorite's book title as value