Utilizing the correlation between decision trees to facilitate mutual understanding

Proposes to detect the correlation between between the nodes in decision trees to facilitate mutual understanding in collaboration. In accordance with the need for dealing with large-scale and complex problems, it is required to support collaborative works among people through the interaction between people and machines. We hypothesize that mutual understanding in collaboration can be established when the way of thinking by others can be explained with his/her “own words” to each other. Mutual classification of diagnosis cases is utilized in our approach to find out the correlation between the nodes in the decision trees without destroying the original characteristics of each decision tree. An experiment on motor diagnosis cases with artificially encoded conceptual difference suggested the effectiveness of our approach toward facilitating mutual understanding