The purpose of this research is to provide a methodology to detect and mitigate destructive patterns that are generated in crowdsourcing systems and result in a negative consequence through contributors’ interactions.
Vote manipulation is a form of co-destruction that can result from intentionally biasing a rating. In Wikipedia, co-destruction can happen for example as a result of blanking pages or inserting unrelated content to the page by a group of people.
Our ultimate goal is to provide guidance when setting up crowdsourcing systems that prevent co- destruction or detect it before it becomes problematic.
As a VRES student you will assist the research team by extracting real-life data sets.
As a research student (Honours, Masters) you may develop algorithms that can provide a detection /mitigation mechanism for a particular co-destruction pattern in crowdsourcing systems such as Wikipedia and GitHub.
As a result of finishing this project, you will need to:
- extract a relevant information from a crowdsourcing platform using API that would help detect co-destruction patterns
- provide an algorithm for detection/mitigation of a particular co-destruction pattern occurring in crowdsourcing systems.
You may be eligible to apply for a research scholarship.
Contact the supervisor for more information.