Computational Analysis of Situational Judgement Tests with Human Values

  by   Adam Wyner






Departments Computer Science, Zienkiewicz Institute for Modelling, Data and AI
DescriptionHuman values such as 'security' or 'benevolence' (so-called Schwartz values) are relevant to how people reason and behave in situations. While rather abstract, we can consider whether or note they are understood to hold in situations. We look at situation judgement tests, where users are given descriptions of situations and asked to judge what would be done and why; they are widely deployed in employment evaluates, e.g., health care, security, or goverment services. The situations we consider involve human values. The aim of the project is to address the question "Can a model accurately infer and predict the 'operative' human value in a situation?". The objective is to exercise rule-based, machine learning, or generative AI approaches on a corpus of situation jugement questions and answers that involve values.
PreparationSchwartz (2012) An Overview of the Schwartz Theory of Basic Values https://scholarworks.gvsu.edu/cgi/viewcontent.cgi?article=1116&context=orpc Situational Judgement Tests https://en.wikipedia.org/wiki/Situational_judgement_test
Project Categories Artificial Intelligence (AI), Human Computer Interaction (HCI), January Cohort, Law
Project Keywords


Level of Studies

Level 6 (Undergraduate Year 3) yes
Level 7 (Masters) yes
Level 8 (PhD) yes