In this workshop, we present the state of research and technology development to address knowledge gaps and we seek input on the direction of research and brainstorm ideas on what new directions to consider. More specifically:
Present - We give a series of short talks about knowledge gaps, what contributes to them, how we can measure, prioritize and build machine learning systems and socio-technical frameworks to help us address knowledge gaps. We divide the study of knowledge gaps into three main areas: content gaps , readership gaps , and contributorship gaps . We will present research on how to detect what content is missing from an already existing article (sections, citations, images) and bring more content to more languages, the state of readership across 14 Wikipedia languages, as well as the gaps in contributorship and the result of an experiment to improve the retention of women in Wikipedia by building frameworks that can increase confidence early on when they join the projects.
Future - We will discuss the challenges and opportunities for each of the three directions. We will brainstorm with the audience on what new areas of research to consider and how to improve the current approaches we are undertaking.
Test and feedback - We will showcase some of the technologies built in the past year and the audience are encouraged to work with them and provide feedback on how well the technology is working, the applications it can have in their project or language, etc.
The attendees of the session will:
Arrive at a shared understanding and language around knowledge gaps. Knowledge gaps is /not/ content gaps only, it is a combination of content, readership, and contributorship gaps. We need to address all three to address knowledge gaps.
Learn about research that has been done in the area of knowledge gaps in the past year. More specifically, identifying gaps in images, citations, article sections and content.
Learn about the technology (APIs, tools, etc.) built in the past year to help us address knowledge gaps together.
Provide direct feedback for the technology that is built.
Participate in brainstorming conversations on how to use the research outputs and technology built to address Wikimedia’s knowledge gaps.
Brainstorm and ideate on the next steps of the research and important directions we should focus on for further research and technology development.
The audience of this session are editors, editathon organizers, developers, affiliates, representatives of User Groups and Chapters, and of course, researchers. :)