For our final project, we have decided to use this Goodreads dataset to build a book recommender! We realized that recommenders such as Netflix, Spotify, and UberEats exist for movies, music, food, etc. and we wanted to further explore text information systems as taught in this course. Books seem like unexplored territory so we wanted to take advantage of the available data and build a book recommendation system! There’s not really a system/app in place that allows people to search a phrase or get custom recommendations for books to read - Goodreads is more to rate books and save books to read later; not really subscription based or using other user data to suggest books to users.
In this project, we implemented 2 tasks:
- Task 1: recommending books based on user inputted phrase
- Task 2: building a content-based and collaborative-filtering recommender
Our software documentation is found here
The demo recording for our project is found here
The link to download the data used in our use cases is found here
Presentation slides are available here
Our progress report is available here
Our initial proposal for the project is found here
- Goodreads Dataset. https://sites.google.com/eng.ucsd.edu/ucsdbookgraph/home
- Mengting Wan, Julian McAuley, "Item Recommendation on Monotonic Behavior Chains", in RecSys'18.
- Mengting Wan, Rishabh Misra, Ndapa Nakashole, Julian McAuley, "Fine-Grained Spoiler Detection from Large-Scale Review Corpora", in ACL'19