Wikimedia CEE Meeting 2024/Submissions/Research to support multilingual knowledge on Wikipedia
| ID : 106 | Research to support multilingual knowledge on Wikipedia | ||
|---|---|---|---|
| Author(s): Pablo Aragón | Username(s): Pablo (WMF) | Type of submission: Lecture | |
| Affiliation: Wikimedia Foundation | Theme(s): Research, technology | ||
| Abstract:
The mission of the Wikimedia Foundation Research team is to develop models and insights utilizing scientific methods, and strengthen Wikimedia research communities in order to support technology and policy needs, and advance the understanding of the Wikimedia projects. In this lecture I will give an overview of recent research led by the team to address knowledge gaps and improve knowledge integrity on Wikipedia and sister projects. In particular, I will show novel machine learning models based on language-agnostic and multilingual approaches so that they can be used by researchers, developers and volunteers from any language community. |
Slides: Slide deck | ||
| Level of advancement: Advanced | |||
| Special requirements: None | |||
| How will this session be beneficial for the communities in the region of CEE?
Attendees will have an up-to-date overview of the work conducted by the Wikimedia Foundation Research team. In particular, they will learn how they can leverage existing machine learning models for tasks of interest such as inferring the topic of a Wikipedia article, estimating the quality of a revision, or predicting its likelihood of being reverted. To illustrate the practical usefulness of these computational techniques, findings will be presented using data from popular projects among communities in the region of CEE. In addition, the final discussion will allow attendees to provide feedback to improve future work. | |||
Interested participants
edit(register below and ask your questions now to the session organiser)
- ...