Detection algorithm for conceptual difference among different persons-as an example of emergence in the human interaction
Taketoshi Yoshida, T. Kondo, Naoki Saiwaki, S. Nishida · 2002
We consider emergence in the human interaction by taking the discovery of conceptual difference as an example. Conceptual difference is a serious problem in group knowledge acquisition systems, especially when different people with different background participate in a group. We deal with the conceptual difference, in which different symbols are used to express the same meaning and/or the same symbol is used to express different meanings. We propose a method to detect such types of conceptual difference by utilizing decision trees which are constructed with ID3 algorithm and GA. Our method is discussed in terms of the realization of emergence in the human interaction. We tackle the "vague" concept of emergence in the human interaction by constructing "concrete" structures to express conceptual difference. Our method ensures the variety in the conceptual structures by utilizing GA to explore various possibilities. The various structures constructed by the system play the role of pointing out conceptual difference from another viewpoint, which is different from those of the members in the group. This function enables the members to understand each other as well as their own concepts, which is expected to contribute to hitting new ideas as a group and to promoting the activity of the group.