Anastasia Zhukova, a graduate of the University of Konstanz’s Master’s Programme Computer and Information Science, has been awarded the Women’s STEM Award 2019. | Copyright: Mikhail A. Zhukov
Anastasia Zhukova, a graduate of the University of Konstanz’s Master’s Programme Computer and Information Science, has been awarded the Women’s STEM Award 2019. | Copyright: Mikhail A. Zhukov

Women’s STEM Award

Anastasia Zhukova, a graduate of the University of Konstanz’s Master’s Programme Computer and Information Science, has won the Women’s STEM Award 2019 for developing an AI application that detects media bias

Anastasia Zhukova, a graduate of the University of Konstanz’s Master’s Programme Computer and Information Science, has been awarded the Women’s STEM Award 2019 for her master’s thesis on “Automated Identification of Framing by Word Choice and Labeling to Reveal Media Bias in News Articles”, which she wrote as a student in Professor Bela Gipp’s Information Science Group. It was part of a larger three-year media bias project entitled “Fake News and Collective Decision Making: Rapid Automated Assessment of Media Bias”, which is funded by the Heidelberg Academy of Sciences and Humanities and is jointly conducted by Professor Bela Gipp and Professor Karsten Donnay, head of the University of Konstanz’s Computational Social Science group. The award ceremony took place at the German Telekom headquarters in Bonn at the beginning of July.

The Women’s STEM Award, which is endowed with EUR 3000 in prize money, has been awarded by Deutsche Telekom, the student magazine “audimax” and the national “MINT Zukunft schaffen” (creating a STEM future) initiative since 2014. The competition is open to female STEM (Science, Technology, Engineering and Mathematics) students from around the world, who may submit either their bachelor’s or master’s theses in a relevant academic discipline.

Anastasia Zhukova’s master’s thesis was selected by an international jury of experts as the most outstanding submission in the area of Artificial Intelligence and the best overall thesis in the competition, which, besides Artificial Intelligence, also included the following categories: Cloud, Cyber Security, Internet of Things and Networks of the Future. Combining social science and computer science approaches and applying them to the field of media analytics, she developed a new AI-based method for automatically recognizing differently phrased and, thus, biased mentions of semantic concepts in news articles. It allows her to detect and resolve phrases referring to persons, groups of individuals, entities, and other concepts that previous automated natural language processing methods were unable to identify. To refer to semantic concepts, journalists and politicians can use different word choice and thus quite often exploit the underlying shifts in language to achieve the desired framing effect. The term “framing” refers to the presentation of events in a way that elicits a specific perception thereof in the readers.

To improve on the existing state-of-the-art methods, Anastasia Zhukova collaborated with experts for frame analysis from the University of Konstanz’s Department of Politics and Public Administration as well as with experts from the University of Konstanz’s Information Science Group and the National Institute of Informatics (NII) in Tokyo, Japan, where she completed a six-month research stay in preparation for her master’s thesis. Having earned her master’s degree from the University of Konstanz, Anastasia Zhukova continues her work on revealing biased news coverage as a doctoral researcher at the University of Wuppertal.

Facts:

  • Anastasia Zhukova, a graduate of the University of Konstanz’s Master’s Programme Computer and Information Science, has been recognised with the Women’s STEM Award 2019
  • Prize money: EUR 3000
  • Her master’s thesis on “Automated Identification of Framing by Word Choice and Labeling to Reveal Media Bias in News Articles” was selected as the most outstanding submission in the area of Artificial Intelligence and best overall thesis
  • It introduces a new AI method for automatically recognising biased media articles, improving on previous automated natural language processing tools
  • Part of a three-year media bias project funded by the Heidelberg Academy of Sciences and Humanities entitled “Fake News and Collective Decision Making: Rapid Automated Assessment of Media Bias”, conducted by Professor Bela Gipp from the University of Konstanz’s Department of Computer and Information Science (now University of Wuppertal) and Professor Karsten Donnay from the Department of Politics and Public Administration
  • Women’s STEM Award 2019 awarded by Deutsche Telekom, the student magazine “audimax” and the “MINT Zukunft schaffen” (creating a STEM future) initiative under the patronage of German Chancellor Angela Merkel