https://github.com/sukhmancs/recommender_system
This repository contains an implementation of a text document recommender system using Python. The system recommends similar documents based on vector representations and similarity calculations.
https://github.com/sukhmancs/recommender_system
Last synced: 7 months ago
JSON representation
This repository contains an implementation of a text document recommender system using Python. The system recommends similar documents based on vector representations and similarity calculations.
- Host: GitHub
- URL: https://github.com/sukhmancs/recommender_system
- Owner: sukhmancs
- License: mit
- Created: 2024-02-29T05:00:31.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2024-02-29T08:12:51.000Z (over 2 years ago)
- Last Synced: 2024-12-27T13:46:03.277Z (over 1 year ago)
- Language: Python
- Size: 57.1 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
🚀 Recommender System
This repository contains an implementation of a text document recommender system using Python. The system recommends similar documents based on vector representations and similarity calculations.
Overview
Recommender systems such as this are a core application of statistical AI. At the heart of recommender systems is a similarity calculation. In this implementation, we use vector representations of documents and a document similarity calculation to recommend relevant articles to users.
# Features
- Load text documents from various datasets (e.g., BBC news articles, scientific abstracts, Wikipedia articles).
- Preprocess text data to remove stopwords and perform lemmatization.
- Vectorize documents using TF-IDF vectorization.
- Calculate document similarity using cosine similarity.
- Generate recommendations based on the most similar documents to a selected document, while also including some less similar documents to provide diversity.
- Avoid recommending the same document or documents with the same title as the selected document.
# Usage
Clone the repository:
```bash
git clone https://github.com/your_username/text-document-recommender.git
```
Run the recommender system:
```bash
python recommender.py
```
# Dataset
The datasets used in this project include:
- BBC news articles
- Scientific abstracts
- Wikipedia articles
These datasets have been adapted for this task.
# License
This project is licensed under the MIT License. See the [LICENSE file](./LICENSE) for details.