Intelligent link categorization based on site's content.
NLP Link Saver is a simple web application designed to categorize web links. Built on a modern tech stack, it leverages a fine-tune of DeBERTa-v3-base from Microsoft, DeBERTa-v3-base-mnli-fever-anli. It serves as a proof of concept for a real world application of NLP text classification.
- Frontend: Next.js
- Backend: Prisma, SQLite
- Text Classification Model: DeBERTa-v3-base-mnli-fever-anli via Hugging Face
-
Clone the Repository
git clone https://github.com/your-username/nlp-link-saver.git cd nlp-link-saver -
Install Dependencies
npm install
-
Set Up Environment Variables
Create a
.envfile in the project root and add the necessary environment variables.DATABASE_URL="file:./dev.db" HUGGINGFACE_BEARER_TOKEN="your_hugging_face_bearer_token" -
Run the Application
npm run dev
Visit
http://localhost:3000in your browser.
- Add new links via the input field.
- View categorized links on the main page.
Laurer, Moritz, Wouter van Atteveldt, Andreu Salleras Casas, and Kasper Welbers. 2022. ‘Less Annotating, More Classifying – Addressing the Data Scarcity Issue of Supervised Machine Learning with Deep Transfer Learning and BERT - NLI’. Preprint, June. Open Science Framework. https://osf.io/74b8k.