Claudia Müller-Birn

Jesse Josua Benjamin, Christoph Kinkeldey, Claudia Müller-Birn et al.

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

Katrin Glinka, Patrick Tobias Fischer, Claudia Müller-Birn et al.

We present an exploratory case study describing the design and realisation of a ''pure mixed reality'' application in a museum setting, where we investigate the potential of using Microsoft's HoloLens for object-centred museum mediation. Our prototype supports non-expert visitors observing a sculpture by offering interpretation that is linked to visual properties of the museum object. The design and development of our research prototype is based on a two-stage visitor observation study and a formative study we conducted prior to the design of the application. We present a summary of our findings from these studies and explain how they have influenced our user-centred content creation and the interaction design of our prototype. We are specifically interested in investigating to what extent different constructs of initiative influence the learning and user experience. Thus, we detail three modes of activity that we realised in our prototype. Our case study is informed by research in the area of human-computer interaction, the humanities and museum practice. Accordingly, we discuss core concepts, such as gaze-based interaction, object-centred learning, presence, and modes of activity and guidance with a transdisciplinary perspective.

Claudia Müller-Birn, Katrin Glinka, Peter Sörries et al.

Researchers and practitioners increasingly consider a human-centered perspective in the design of machine learning-based applications, especially in the context of Explainable Artificial Intelligence (XAI). However, clear methodological guidance in this context is still missing because each new situation seems to require a new setup, which also creates different methodological challenges. Existing case study collections in XAI inspired us; therefore, we propose a similar collection of case studies for human-centered XAI that can provide methodological guidance or inspiration for others. We want to showcase our idea in this workshop by describing three case studies from our research. These case studies are selected to highlight how apparently small differences require a different set of methods and considerations. With this workshop contribution, we would like to engage in a discussion on how such a collection of case studies can provide a methodological guidance and critical reflection.

Jesse Josua Benjamin, Claudia Müller-Birn, Simon Razniewski

The global infrastructure of the Web, designed as an open and transparent system, has a significant impact on our society. However, algorithmic systems of corporate entities that neglect those principles increasingly populated the Web. Typical representatives of these algorithmic systems are recommender systems that influence our society both on a scale of global politics and during mundane shopping decisions. Recently, such recommender systems have come under critique for how they may strengthen existing or even generate new kinds of biases. To this end, designers and engineers are increasingly urged to make the functioning and purpose of recommender systems more transparent. Our research relates to the discourse of algorithm awareness, that reconsiders the role of algorithm visibility in interface design. We conducted online experiments with 105 participants using MTurk for the recommender system Recoin, a gadget for Wikidata. In these experiments, we presented users with one of a set of three different designs of Recoin's user interface, each of them exhibiting a varying degree of explainability and interactivity. Our findings include a positive correlation between comprehension of and trust in an algorithmic system in our interactive redesign. However, our results are not conclusive yet, and suggest that the measures of comprehension, fairness, accuracy and trust are not yet exhaustive for the empirical study of algorithm awareness. Our qualitative insights provide a first indication for further measures. Our study participants, for example, were less concerned with the details of understanding an algorithmic calculation than with who or what is judging the result of the algorithm.

Maximilian Mackeprang, Kim Kern, Thomas Hadler et al.

People react differently to inspirations shown to them during brainstorming. Existing research on large-scale ideation systems has investigated this phenomenon through aspects of timing, inspiration similarity and inspiration integration. However, these approaches do not address people's individual preferences. In the research presented, we aim to address this lack with regards to inspirations. In a first step, we conducted a co-located brainstorming study with 15 participants, which allowed us to differentiate two types of ideators: Inspiration seekers and inspiration avoiders. These insights informed the study design of the second step, where we propose a user model for classifying people depending on their ideator types, which was translated into a rule-based and a random forest-based classifier. We evaluated the validity of our user model by conducting an online experiment with 380 participants. The results confirmed our proposed ideator types, showing that, while seekers benefit from the availability of inspiration, avoiders were influenced negatively. The random forest classifier enabled us to differentiate people with a 73 \% accuracy after only three minutes of ideation. These insights show that the proposed ideator types are a promising user model for large-scale ideation. In future work, this distinction may help to design more personalized large-scale ideation systems that recommend inspirations adaptively.

Maximilian Mackeprang, Claudia Müller-Birn, Maximilian Timo Stauss

Interactive intelligent systems, i.e., interactive systems that employ AI technologies, are currently present in many parts of our social, public and political life. An issue reoccurring often in the development of these systems is the question regarding the level of appropriate human and computer contributions. Engineers and designers lack a way of systematically defining and delimiting possible options for designing such systems in terms of levels of automation. In this paper, we propose, apply and reflect on a method for human-computer configuration design. It supports the systematic investigation of the design space for developing an interactive intelligent system. We illustrate our method with a use case in the context of collaborative ideation. Here, we developed a tool for information extraction from idea content. A challenge was to find the right level of algorithmic support, whereby the quality of the information extraction should be as high as possible, but, at the same time, the human effort should be low. Such contradicting goals are often an issue in system development; thus, our method proposed helped us to conceptualize and explore the design space. Based on a critical reflection on our method application, we want to offer a complementary perspective to the value-centered design of interactive intelligent systems. Our overarching goal is to contribute to the design of so-called hybrid systems where humans and computers are partners.

Christoph Kinkeldey, Claudia Müller-Birn, Tom Gülenman et al.

Machine learning systems are ubiquitous in various kinds of digital applications and have a huge impact on our everyday life. But a lack of explainability and interpretability of such systems hinders meaningful participation by people, especially by those without a technical background. Interactive visual interfaces (e.g., providing means for manipulating parameters in the user interface) can help tackle this challenge. In this paper we present PreCall, an interactive visual interface for ORES, a machine learning-based web service for Wikimedia projects such as Wikipedia. While ORES can be used for a number of settings, it can be challenging to translate requirements from the application domain into formal parameter sets needed to configure the ORES models. Assisting Wikipedia editors in finding damaging edits, for example, can be realized at various stages of automatization, which might impact the precision of the applied model. Our prototype PreCall attempts to close this translation gap by interactively visualizing the relationship between major model metrics (recall, precision, false positive rate) and a parameter (the threshold between valuable and damaging edits). Furthermore, PreCall visualizes the probable results for the current model configuration to improve the human's understanding of the relationship between metrics and outcome when using ORES. We describe PreCall's components and present a use case that highlights the benefits of our approach. Finally, we pose further research questions we would like to discuss during the workshop.