Author name cluster

Masamichi Shimura

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

3 papers

1 author row

IJCAI Conference 1997 Conference Paper

Using Data and Theory in Multistrategy (Mis)Concept(ion) Discovery

Raymond Sison
Masayuki Numao
Masamichi Shimura

Most conceptual clustering systems rely solely on data to form concepts without supervision; the few that exploit causalities in the background knowledge do so only after the completion of a similarity-based learning phase. In this paper, we describe a multistrategy misconception discovery system, MMD, that utilizes data and theory in a more tightly coupled way. The integration of similarity- and causality-based learning in MMD is shown to be essential for the automatic construction of accurate and meaningful misconceptions that account for errors in novice behavior.

PDF

AAAI Conference 1986 Conference Paper

Learning Arithmetic Problem Solver

Masamichi Shimura

In this paper we describe a problem solving system with a learning mechanism (Learning Arithmetic Problem Solver, LAPS), which can solve arithmetic problems written in natural languages. Since LAPS has knowledge about arithmetic problems in the form of rules, it can solve many different problems without alteration of the program. When LAPS cannot solve a given problem because of a shortage of knowledge, it asks the user how to solve the problem. According to the user’s advice LAPS acquires knowledge and rules. Using these rules, LAPS can solve problems. Furthermore, LAPS can improve its performance at problem solving by synthesizing rules that are applied.

PDF Details

AIJ Journal 1973 Journal Article

Rule-oriented methods in problem solving

Masamichi Shimura
Frank H. George

A heuristic, in the usual sense and certainly in the sense that this term is used in this paper, is a rough guide or principle, or underlying knowledge which is not expressed explicitly. Heuristic methods for solving problems can involve the incorporation of such knowledge into algorithms. This paper describes two methods for finding the solution to problems using heuristic methods. A solution to a problem is represented by a sequence of operators which transpose one problem situation to another. Problems are categorized into four types: reversible regular, irreversible regular, reversible irregular and irreversible irregular problems. The rule-oriented method described were is useful particularly for the reversible regular problems to which heuristic knowledge is applicable. Some examples of the application of heuristics are shown in solving the Missionary-and-Cannibal problem and the Tower-of-Hanoi problem.

Details DOI