Information Retrieval Query Generation

Annotator toolkit for creating manual queries from information retrieval experiments / systems.

Uses AnjularJS, Django

Pre-requirements

You should have at least the following packages installed on your machine: django==1.8 and djangorestframework (worked with version 3.3.3) You can install them using pip:

pip install djangorestframework django==1.8

Setup

All the data was provided to in CSV format. The process to setup is as follows:

• Run ./manage.py syncdb to create the db. Follow the instructions.
• Run python parse_html_to_create_topics.py selection-50-queries/ to create a file called topics_clef.json
• Run the python import_patients.py db.sqlite3 topics_clef.json to import the set of queries.
• Start the webserver with ./manage.py runserver
• Go to http://localhost:8000/<n>, where n is the nth query you want to start annotating.

Getting data out

See http://localhost:8000/queries for the REST get for all queries and corresponding keywords.

Known Bugs

• The current version does not work with django==1.9 yet.