Image Retrieval for Arguments

Title	Image Retrieval for Arguments
Summary	Given a controversial topic, the task is to retrieve images (from web pages) for each stance (pro/con) that show support for that stance.
Keywords	NLP, ML
TimeFrame
References	https://touche.webis.de/clef22/touche22-web/image-retrieval-for-arguments.html
Prerequisites	ML, Python
Author
Supervisor	Pablo Picazo
Level	Master
Status	Draft

A full description of the project can be found: https://touche.webis.de/clef22/touche22-web/image-retrieval-for-arguments.html

Synopsis:

Given a controversial topic, the task is to retrieve images (from web pages) for each stance (pro/con) that show support for that stance. [forum]

Data:

This task uses a focused crawl of about 20,000 images (and associated web pages) as document collection. See the collection's README for more information on its contents and file formats.

Evaluation

Systems are evaluated on Touché topics 1–50 by the ratio of images among the 20 retrieved images for each topic (10 images for each stance) that are all three: relevant to the topic, argumentative, and have the associated stance. The file format is explained in the README).