Explanation Strategies as an Empirical-Analytical Lens for Socio-Technical Contextualization of Machine Learning Interpretability

During a research project in which we developed machine learning (ML) driven visualization system for non-ML experts, reflected on interpretability ML, computer-supported collaborative work and human-computer interaction. We found that while there are manifold technical approaches, these often focus ML experts evaluated decontextualized empirical studies. hypothesized participatory design may support the understanding of stakeholders' situated sense-making our project, yet, guidance regarding inexhaustive. Building philosophy technology, formulated explanation strategies as an empirical-analytical lens explicating how explanations mediate contextual preferences concerning people's interpretations. In this paper, contribute report proof-of-concept use to analyze co-design workshop with methodological implications research, suggest further investigation technological mediation theories space.