Fuzzy logic approach on cognition diagnosis with application on number concept for pupils

The purpose of this research is to provide an integrated methodology of fuzzy logic approach on cognition diagnosis. Based on the item concept matrix of testing data, the integrated methodology could provide individualized knowledge structure and clustering of students. The algorithm of this integrated methodology consists of fuzzy logic model of perception (FLMP), interpretive structural modeling (ISM) and fuzzy clustering. The individualized knowledge structure will clearly display features of knowledge structure for each student. Clustering of all students will illustrate the segment of total sample so that students within the same cluster own similar concept structures. Remedial instruction will also become feasible according to the analytic results. An empirical data on number concept for pupils is analyzed and it shows that features of knowledge structures vary with testing score and response patterns.