Knowledge Affordances for Hybrid Human-AI Information Seeking

A recent preprint on arXiv presents the idea of knowledge affordances (KAs) to organize how agents recognize significant opportunities for information retrieval in mixed human-AI settings. The study defines KAs as explicit, semantically based descriptions detailing what a knowledge source can provide, the types of inquiries it can address, and the contextual attributes involved. It posits that KAs are relational, potentially arising from the interaction of the agent's objectives, preferences, and contextual elements. This research links various fields, including affordances, semantic web services, and knowledge engineering.