2018/10/01 by Aysenur Bilgin, Bilgin, Aysenur, Laura Hollink +11
Computer Science · #Authorship Attribution and Profiling #Computation and Language (cs.CL) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Natural Language Processing Techniques #Topic Modeling
paper · pdf · doi:10.48550/arxiv.1810.00968
openalex publication_date 2018/10/01 · openalex created_date 2022/10/01 · openalex updated_date 2026/07/28
With the growing abundance of unlabeled data in real-world tasks, researchers\nhave to rely on the predictions given by black-boxed computational models.\nHowever, it is an often neglected fact that these models may be scoring high on\naccuracy for the wrong reasons. In this paper, we present a practical impact\nanalysis of enabling model transparency by various presentation forms. For this\npurpose, we developed an environment that empowers non-computer scientists to\nbecome practicing data scientists in their own research field. We demonstrate\nthe gradually increasing understanding of journalism historians through a\nreal-world use case study on automatic genre classification of newspaper\narticles. This study is a first step towards trusted usage of machine learning\npipelines in a responsible way.\n