Introducing assistive analytical agent for exploratory and decision making in analytical multi-user sessions

Official URL: http://risc01.sabanciuniv.edu/record=b1817056 (Table of Contents)

In multi-user analytical sessions, participants analyze data and collaborate toward a shared vision. These decision-making processes are challenging and generally prolonged. In this thesis, we introduce a system for facilitating decision-making in multi-user analytical sessions. Our system is comprised of a tabletop setup and an assistive analytical agent. The tabletop, as a medium for data visualization, makes data tangible, hence more comprehensible and natural to operate with. Simultaneously, while the participants interact with the data on the table, the assistant agent understands their ongoing conversations and offers extra information about the data. In addition, the agent answers the participants’ questions either regarding the data or open-domain ones, supports or rejects their ideas by providing its reasoning, and preserves the productivity and the efficiency of the session by confirming that the participants do not deviate from the session’s goal. Overall, by combining data visualization with the analytical agent our system accelerates the analytical process, fosters exploration through it, and guides next sessions using previous sessions’ summarizations.