Social Computing: Supporting Grounding in Communication with IoT-based Knowledge Sharing and Learning

Main Achievement:

KnowledgeKeeper (KK) is a framework to support knowledge transfer for the purpose of work training and task performing.

We conceptualize it with three main parts:

     Capture experts’ workflow and knowledge

     Synthesize knowledge

     Represent knowledge for learning

Now, KK has already captured experts’ task workflow with multi-modal sensors. 

Quantitative Impact: 

We compared the amount of knowledge units being generated when pairing expert-performers with three different partners, another domain expert, a novice learner, or a teaching expert.

Expert-performers who were paired with novice partners can generate more knowledge units than other two pairs. (Experts in Expert-Novice pair generated about 40 knowledge units, while they generated about 30 knowledge units in Expert-Expert and Expert-Teacher condition.

When pairing with another domain expert, expert-performer can have more discussion with their partner and reflect more than other two conditions.

End Goal: 

–     To capture experts’ workflow in multi-modal manner

–     To synthesize content from multiple experts with human computation 

–     To co-construct knowledge with learners when learners are interacting with KK

 (Updated in Jul, 2017)