Please note: This master’s thesis presentation will take place online.
Ehsan Jahangirzadeh Soure, Master’s candidate
David R. Cheriton School of Computer Science
Supervisor: Professor Jian Zhao
Complex problems and interprofessional work require more resources to be involved, which has been possible through collaboration. Collaborative work is evolving from physical collaboration to more virtual forms through digital media over time. The evolution of collaboration is not just about how and where it is conducted. With the improvements in artificial intelligence, computers are moving towards playing a collaborative role instead of just supporting human-human collaboration. As conversations are essential in collaborative tasks to achieve common objectives, in this thesis, we investigate human-human and human-AI collaborative interfaces to support effective discussion.
First, we introduce a collaborative visual analytics tool, CoUX, to facilitate UX evaluators collectively reviewing and discussing think-aloud usability test videos of digital interfaces. We designed CoUX based on a formative study with two UX experts and insights from the literature. CoUX enables collaboration amongst UX evaluators for logging, commenting, and consolidating the discovered problems with a chatbox-like user interface in order to reduce bias and errors. Furthermore, CoUX is the first step in this thesis to integrate synchronous and asynchronous collaboration in the think-aloud review workflow.
As the next step, we implement IVA, an intelligent virtual assistant that automatically provides tools and services to meeting attendees based on their conversations to facilitate collaboration and multitasking amongst them. To design IVA, we conducted a formative study to investigate users’ current practices and concerns for online meetings and how an AI assistant can facilitate collaboration and multitasking. IVA features a set of components, such as a customizable infinite environment where users can add and manipulate any applications besides the main meeting session. To validate our assumptions and design, we evaluated our design by mocking collaborative meetings with our participants using a Wizard-of-Oz implementation. In general, participants found IVA less distracting and more efficient compared to existing online meeting platforms.