Information enclosing knowledge networks: A study of social relations

Abstract

We design and test a multiagent-based knowledge-exchange network, the results of which will guide corporate decision-making in this field. Its novelty resides in the application of multiagents to the field of knowledge management. We selected this methodology because it is valid for the purpose of the study, because it generates realistic and practical conclusions and, moreover, because it serves to negotiate the difficulties that these aspects of a qualitative and social nature can in reality imply for the firm. We designed the knowledge network through simulation with multiagents, studying some of their most important parameters. In doing so, we arrived at a set of conclusions, prominent among which are that by using the network, we may easily identify experts, and people that either share or conceal knowledge. The network also helps us locate people with leadership capabilities and those that remain isolated, as well as the formation of subgroups that segregate themselves from the original network and that can set up their own independent relational structures, unsupportive of knowledge exchange.