QuestionComb: A Gamification Approach for the Visual Explanation of Linguistic Phenomena through Interactive Labeling

Linguistic insight in the form of high-level relationships and rules text builds basis our understanding language. However, data-driven generation such structures often lacks labeled resources that can be used as training data for supervised machine learning. The creation ground-truth is a time-consuming process requires domain expertise to resolve ambiguities characterize linguistic phenomena. Furthermore, refinement learning models challenging linguists are complex, in-transparent, difficult understand. To tackle these challenges, we present visual analytics technique interactive labeling applies concepts from gamification explainable Artificial Intelligence (XAI) support complex classification tasks. visual-interactive interface promotes effective data. Visual explanations learned unveil decisions model iterative optimization. gamification-inspired design guides user through provides feedback on performance. As an instance proposed technique, QuestionComb , workspace tailored task question (i.e., information-seeking vs. non-information-seeking questions). Our evaluation studies confirm beneficial engage users continuous feedback, offering when combined with active XAI.