To provide learning suggestions to individual students, the error ratio (ER) for each student to answer the test items related to each concept needs to be analyzed; therefore, it is necessary to establish a Test Item Relationship Table (TIRT), which represents the degree of association between test item Qi and concept Ck (Hwang, 2003a). An illustrative example of a TIRT comprising ten concepts and twelve test items is listed in Table 1, where the TIRT (Qi, Ck) is a value ranging from 0 to 1; “1” represents “high relevance” and “0” represents “no