Applyxmag aids inclusive design with high precision and low cost
Apply-<x>Mag: One Tool to Support Many Inclusive Design Methods
Human-Computer Interaction
Summary
Making products that everyone can use can be hard and expensive, which makes some companies skip it and leave some users behind. The authors created Apply-<x>Mag, a tool powered by large language models that helps designers include many types of users by checking their designs against inclusive design guidelines. They tested it on several products and found it was quite accurate and didn’t use many computer resources. This tool helps designers make their products better for diverse users without spending too much time or effort.
What this means in practice
- •For product design teams: Check design products automatically against inclusive design principles to reduce user experience barriers for underserved populations.
- •For usability testing groups: Use the tool to validate inclusive design methods quickly across multiple products with measurable precision and recall.
Authors
Sadia Afroz, Rudrajit Choudhuri, Fatima A. Moussaoui, Amreeta Chatterjee, Margaret Burnett, Anita Sarma
Abstract
Doing inclusive design in HCI practice can be labor-intensive, a costly barrier that some companies and HCI practitioners may be unwilling or unable to overcome. Yet, not doing inclusive design is costly too, in the form of UX barriers that disproportionately disadvantage under-served user populations. To address this problem, we introduce Apply-<x>Mag, an LLM-powered tool to support HCI practitioners' work to design their products inclusively to wide ranges of users. Apply-<x>Mag is general, supporting any inclusive design method that can be expressed as <x>Mags (i.e., using attribute ranges and heuristics). It is also effective: Empirical results with researcher and practitioner teams using various combinations of two <x>Mags on 7 products showed Apply-<x>Mag precision averaging 90-99% and recall averaging 82-89%. Further, its environmental costs were reasonable, costing about the same resources as 2-4 ordinary Google searches.