Applying the Design Sprint to Interactive Machine Learning Experience Design: A Case Study from Aveni

Abstract

AbstractDesign Thinking was applied to the Aveni product in the format of a Design Sprint. The sprint challenge was to identify & design the first Interactive Machine Learning user experience for the product. Design Thinking was applied to the project to (a) lower the risk of user rejection of the Interactive Machine Learning interaction and (b) decide which data to collect from users first. The outcome of the sprint was an experience-based roadmap towards the selected Interactive Machine Learning interaction. After the sprint, participants structured Human-in-the-loop designs by user workflow where previously they were structured by model or data type. This case study provides an example of the application of Design Thinking, through a Design Sprint, to design an Interactive Machine Learning Human Computer Interaction, in order to lower risk, which might be employed by researchers or other industry professionals. Our main contribution is to present the Design Sprint as an approach for defining which aspects of a machine learning solution are target for user-Interactive Machine Learning, and successfully designing interactions to capture the user input.KeywordsDesign SprintDesign thinking processInteractive machine learningProduct design and development