This is the recommendation algorithm used for the Boka - a book reading application.
- Instructor: Nguyen Tan Toan
├── archive
│ ├── crawled_dataset
│ ├── processed_dataset
│ └── raw_dataset
├── evaluate-precision
│ ├── item-based
│ └── user-based
├── preprocessing
│ ├── book_dataset
│ ├── rating_dataset
│ └── user_dataset
├── src
│ ├── content-based
│ ├── item-based
│ └── user-based
├── test
└── traditional-approach
- archive: Contains dataset collections, including CSV files. This directory holds the original data, data after crawling, and data after preprocessing.
- preprocessing: Contains functions for processing each dataset.
- traditional-approach: Includes a notebook documenting the entire preprocessing process and the recommendation algorithm using classical methods on the original dataset (without additional crawled fields).
- src: Contains models applying three popular algorithms, including content-based and collaborative filtering.
- evaluate-precision: Contains models modifying the input and output of algorithms in the
src
directory to evaluate the accuracy of each algorithm.
You can find the Google Colab notebook and dataset for this project here
Kieu Ba Duong Mobile developer ML researcher |
Do Thanh Dat Backend developer Project manager |