Developed for PRI (Information Processing and Retrieval) @ FEUP - MEIC
Uses Solr' capabilities to create the search engine. Uses both keyword matching and embeddings.
Final Classification: 17.3/20
Report can be found here for in-depth analysis.
In order to access Solr capabilities you need to have Solr 9.0 and Docker installed on your system
Then you need to run pip install -r requirements.txt in your terminal inside backend/project to install dependencies
Since embeddings are a huge file that we couldn't upload to GitHub, you need to create them yourself by running get_embeddings.py and placing the file data_embeddings.json inside data
Now to start Solr run startup.sh inside a bash terminal
The frontend is built using React.
To run it, you need Node.js installed in your system.
After that, run npm install to install the node package manager in your system.
To run the React App, go into the frontend folder and run npm start.
The backend was setup using Django.
You need Django installed in your system.
To do that, run pip install django in your terminal.
To start the server, go into the backend folder and run python3 manage.py runserver or python manage.py runserver.
João Lourenço, João Cardoso, Tiago Cruz, Tomás Xavier