Springe direkt zu Inhalt

Mandana Saberdest:

Redesigning Wikidata's Recoin for Algorithm Awareness

Requirements

  • Good Knowledge of JavaScript, HTML, CSS
  • Proficiency in German & English
Academic Advisor
Discipline
interaction design, user-centered design, front-end
Degree
Bachelor of Science (B.Sc.)

Contents

Recently, the topic of algorithm awareness [1] has received some attention within the disciplines of Human Computer Interaction, Human Factors and Science and Technology Studies. Hamilton et al., in introducing the term, outline that it delineates the extent to which a user is aware of the existence and function of algorithms in a specific context of use. Subsequent researchers have formulated possible methodological characteristics of designing for algorithm awareness [2,3]. One example is Motahhare Eslami's redesign of the Facebook newsfeed [2].

In this BSc, we aim to build on this by redesigning an existing tool in the Wikidata system, called Recoin: https://www.wikidata.org/wiki/Wikidata:Recoin We want to understand how this tool, which currently provides a visual icon that indicates the relative completeness of a given information, can be made more understandable.

A possible procedure could consist in a detailed description of Eslami's experiment, which is then used to conduct similar user tests, i.e.:

  1. Contextual research on the Recoin tool and the Wikidata Community
  2. Conceiving and creating three prototypical designs for the Recoin tool
  3. Conducting comparative user tests (f.e. UEQ)
  4. Evaluate the results as to the effectiveness in terms of algorithm awareness (e.g., trust, transparency, system comprehension)


References

  1. Hamilton, K. et al. 2014. A Path to Understanding the Effects of Algorithm Awareness. CHI ’14 Extended Abstracts, 631–642.
  2. Motahhare Eslami. 2017. Understanding and Designing Around Users’ Interaction with Hidden Algorithms in Sociotechnical Systems. Comp. of the 2017 ACM CSCW Conference, 57–60.
  3. Kizilcec, R.F. 2016. How Much Information?: Effects of Transparency on Trust in an Algorithmic Interface. Proc. of the 2016 CHI Conference, 2390–2395.
  4. Balaraman, V., Razniewski, S., & Nutt, W. 2018. Recoin: Relative Completeness in Wikidata. http://wikiworkshop.org/2018/papers/wikiworkshop2018_paper_2.pdf