Know Thyself
Mentor 1
Trudy Watt
Start Date
1-5-2020 12:00 AM
Description
Know Thyself uses storytelling, architectural representation, and artificial intelligence (AI), to produce a high-fidelity reflection of spatial experience to advance human-centered design more nimbly than previously possible. This data serves as a foundation for testing novel ways of analyzing buildings and inhabitant experience, and for investigating bias in traditional architectural analysis and the emergent field of human-centered AI. This methodology addresses the paradigm shift present in the growing senior population, a large percentage of which are retiring in the next 10 to 20 years. It is crucial to foster preemptive thinking. Developing a fundamentally proactive health support system requires forethought and integrated affordances that directly improve the wellbeing of the aging community. The team is training in the TimeSlips method established by UWM’s Anne Basting. Utilizing her process, we are engaging elderly patients in a narrative-driven conversation which utilizes observation, improvisation, and creativity to help redirect discussions from recollection. These narrative data, along with traditional architectural analysis methods, are the sources for synthesizing an iterative process that cultivates a fingerprint representative of the Luther Manor community. Back at UWM, the group uses these conversations to drive narrative material analysis through exploring new applications of AI that analyze data via a human-centered, machine learning methodology. We will use new applications of unsupervised machine learning, devised by the team, to process these data. Throughout the research process, Know Thyself operates in parallel with the early design efforts of Plunkett Raysich Architects, who have begun the process for renovation and addition to the 19-acre Luther Manor campus in early 2020. During the 2020 spring semester, the group worked on establishing foundational working methodologies through team-oriented research development (i.e., co-design with faculty); and gaining an understanding of the current research taking place in the healthcare and AI communities.
Know Thyself
Know Thyself uses storytelling, architectural representation, and artificial intelligence (AI), to produce a high-fidelity reflection of spatial experience to advance human-centered design more nimbly than previously possible. This data serves as a foundation for testing novel ways of analyzing buildings and inhabitant experience, and for investigating bias in traditional architectural analysis and the emergent field of human-centered AI. This methodology addresses the paradigm shift present in the growing senior population, a large percentage of which are retiring in the next 10 to 20 years. It is crucial to foster preemptive thinking. Developing a fundamentally proactive health support system requires forethought and integrated affordances that directly improve the wellbeing of the aging community. The team is training in the TimeSlips method established by UWM’s Anne Basting. Utilizing her process, we are engaging elderly patients in a narrative-driven conversation which utilizes observation, improvisation, and creativity to help redirect discussions from recollection. These narrative data, along with traditional architectural analysis methods, are the sources for synthesizing an iterative process that cultivates a fingerprint representative of the Luther Manor community. Back at UWM, the group uses these conversations to drive narrative material analysis through exploring new applications of AI that analyze data via a human-centered, machine learning methodology. We will use new applications of unsupervised machine learning, devised by the team, to process these data. Throughout the research process, Know Thyself operates in parallel with the early design efforts of Plunkett Raysich Architects, who have begun the process for renovation and addition to the 19-acre Luther Manor campus in early 2020. During the 2020 spring semester, the group worked on establishing foundational working methodologies through team-oriented research development (i.e., co-design with faculty); and gaining an understanding of the current research taking place in the healthcare and AI communities.
