 |
 |
Managing efficiently and effectively the huge amount of information continuously produced by most of nowadays socio-technical systems (STS) is indeed a challenging task, mostly due to peculiar features such as volume, variety, and velocity of data. Since they are heavily interaction-centred and feature “humans in the loop”, STS call for approaches dealing with the dimension of interaction since their very foundation, in order to, e.g., deal with unpredictability of user behaviour. In this paper we describe the Molecules of Knowledge model (MoK for short), which addresses the above issues by integrating a chemical-inspired computational model with a socio-cognitive model of interaction in order to promote self-organisation of information as well as the adaptiveness of the coordination process. MoK promotes a paradigm shift for information sys- tems, where information items spontaneously aggregate into novel information chunks and flow toward the (potentially) interested prosumers. Accordingly, we thoroughly describe the MoK computational and inter- action models, then discuss its architecture and implementation as a Java-based technology, and finally evaluate the MoK approach through both simulations and test deployments.
keywords
Molecules of Knowledge, anticipatory coordination, behavioural implicit communication, self-organisation, adaptiveness, socio-technical systems, knowledge-intensive environments
journal or series
ACM Transactions on Autonomous and Adaptive Systems
(TAAS)