In flexible production environments, technically supported knowledge becomes more and more a key factor for success. Therefore, this paper presents the development of a learning retrieval agent for knowledge extraction from the Active Semantic Network with respect to user-requests. Based on a reinforcement learning approach, the agent learns to interpret the user’s intention and learn his mental models. Especially, the learning algorithm focuses on the retrieval of complex long distant relations. Increasing its learnt knowledge with every request-result-evaluation sequence, the agent enhances his capability in finding the intended information.
I like the approach for using Reinforcement Learning to study the users view of an ASN. While it is clear that to understand the problem clean room, in this case a user, studies are nessesary, I'm curious if you have done any studies to apply the same method for collaborative information filtering and retrieval. Maybe feeding the user's view model back into the ASN.