https://github.com/hsm207/movielens-weaviate
How to use vector search to build a content-based recommender system
https://github.com/hsm207/movielens-weaviate
large-language-models natural-language-processing neural-search recommender-system recsys vector-search
Last synced: 6 months ago
JSON representation
How to use vector search to build a content-based recommender system
- Host: GitHub
- URL: https://github.com/hsm207/movielens-weaviate
- Owner: hsm207
- Created: 2023-01-21T11:39:14.000Z (almost 3 years ago)
- Default Branch: master
- Last Pushed: 2023-02-06T02:34:00.000Z (almost 3 years ago)
- Last Synced: 2025-04-11T14:25:43.138Z (8 months ago)
- Topics: large-language-models, natural-language-processing, neural-search, recommender-system, recsys, vector-search
- Language: Jupyter Notebook
- Homepage:
- Size: 14.2 MB
- Stars: 6
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Introduction
Code to accompany the [How To Use Vector Search To Quickly Build A Content-Based Filtering Recommender System](https://medium.com/@_init_/how-to-quickly-build-a-content-based-filtering-recommender-system-using-a-vector-database-f6c52d444c94) blog post on Medium
# Prerequisites
1. VS Code
2. Docker
# Usage
1. Clone the repo
2. Open the folder in VS Code inside a dev container when prompted
3. Run `make download-data` to download the datasets
4. Run the [01_metadata.ipynb](./notebooks/01_metadata.ipynb) notebook to prepare to scrape the movie posters and stuff
5. Run `make scrape-movie-metadata` to scrape the movie posters and stuff
6. Read the rest of the notebooks in the [notebooks](./notebooks/) folder in sequence