Jun Arima

Design and Evaluation a Knowledge Management System by Using Mathematical Model of Knowledge Transfer

Based on our experiences, we have proposed a mathematical model for knowledge transfer in order to make knowledge management mechanism or system take root in the organization and to obtain guidelines to make it work. We are developing a know-how sharing system designed based on the insight obtained from the proposed model. We derived and applied the two ideas as design guidelines based on the analysis using the proposed model: one is mutual reviewing to increase the sense of participation, and another is establishment of the criteria to evaluate the background information about the knowledge to be shared. In this paper, we explain the proposed mathematical model and the system design based on the model. And we describe the evaluation on the prototype system. It shows that the mathematical model could derive guidelines to make the KM system work well.

Inductive Logic Programming Beyond Logical Implication

This paper discusses the generalization of definite Horn programs beyond the ordering of logical implication. Since the seminal paper on generalization of clauses based on θ subsumption, there are various extensions in this area. Especially in inductive logic programming(ILP), people are using various methods that approximate logical implication, such as inverse resolution(IR), relative least general generalization(RLGG), and inverse implication(II), to generalize clauses. However, a program is more general than another program does not necessarily mean that the former should logically imply the latter. At least in the context of inductive synthesis of logic programs, we observe that the set inclusion ordering on the success set of logic programs is a more useful notion of generalization.

Reformulation of Explanation by Linear Logic: Toward Logic for Explanation

The use of the concept of “explanation” spreads extensively over fields of Artificial Intelligence: EBG, analogy, abduction, natural language understanding, diagnosis, etc. Their formalisms, however, suffer inconveniences from the nature of the logic underlying them — classical logic. This paper explores one of the crucial inconveniences stemming from classical logic and attempts newly to construct an adequate logic for “explanation” based on linear logic.