engleski

Expert System for Detecting a Child's Gift in Mathematics

According to Johnson (2000) mathematically gifted children have needs that differ from those of other children. In order to pay special attention to gifted children, teachers usually use mathematical competencies as the only criterion for detecting such pupils. However, it is also important to include other components while deciding about the giftedness in mathematics, such as cognitive components of gift, personal components that contribute the gift development, strategies of learning and exercising, as well as some environmental factors. This paper aims to identify five main components of gift in mathematics based on previous research, and to create an intelligent expert system that will support teachers in detecting gifted children in the fourth year of elementary school. The systems is based on decision rules and forward chaining inference engine used to classify each pupil into one of the four output categories: (1) presumably gifted child in mathematics, (2) child with a special interest in mathematics, (3) child with average mathematical competencies, and (4) child with insufficiently developed mathematical competencies. An empirical research is conducted including 247 pupils, age 10, selected from different elementary schools in Osijek, Croatia. Both, teacher and the expert system assessments, are obtained for each of the pupils. The paper also compares the decisions of the expert system with the teachers' assessments of pupils' gift. The results show a significant difference among teacher and system assessments, and that more pupils are identified as potentially gifted by the system. It implies that an intelligent system leads teacher to consider other components of giftedness in mathematics, and therefore can serve as an efficient methodological tool in detecting gifted children as well as in educating teachers.

components of gift ; mathematics ; intelligent expert system ; if-then rules ; estimation of gift ; t-test

Rad je u cjelosti indeksiran u najvećoj svjetskoj digitalnoj bazi pedagoških istraživanja Education Resources Information Center (ERIC). Stoga je pretraživ i dostupan preko Centra za online baze podataka u Hrvatskoj.