How to setup corpus collection if not using the data collected:
- Set-up SearchX backend, instructions can be found here: https://github.com/searchx-framework/searchx-backend
- Add a BingApi key to the .env file that you created in the previous step as follows:
BING_ACCESS_KEY=<your_api_Key>
. This is required because the vertical selection relies on the BingApi to retrieve results. - Startup the SearchX Backend, after the server is up you should see some messages about queries and verticals being imported, and their results retrieved
How to import collected corpus data:
- Download and extract the data, it can be found here: https://drive.google.com/uc?id=1utMND5IPfTkNGDlEyMjm-aZJ9rpQt4t1&export=download
- Use the command
mongorestore --host <your_db_host> --port <your_db_port> --db aggregated-search dump/aggregated-search
to import the data into the database.
To run the evaluation script:
- Install Python 3.7 locally using a virtual or a system environment.
- Install the requirements listed in the file
requirements.txt
using the commandpip install -r requirements.txt
- Run the
main.py
located in thepython
directory using the command:python python/main.py
. If using a virtual environment, make sure that you are running the script in it.