This study aims to establish a system based on knowledge structure and Bayesian network and a digital individual teaching material which is adaptable to Bayesian Network based Asaptive Test（BNAT）,and then compare the effectiveness between group and individual instruction models proposed. The focus of this study is on the unit of “Cuboid and Cube” . After analyzing the content, the structures of the content were constructed, and items were designed according to them as well. After the pre-test, ordering theory is used to decide the students' knowledge structure and those parameters were used in the item bank. Additionally, the instruction is maked up by referring to the remedial knowledge structure of experts and students. Accordingly the digital individual instruction material is developed. The material was provided for experimental teaching. This experimental teaching includes two kinds of instruction models as following: one teacher teaches two students and one teacher teaches a group of students. Some findings are briefly outlined as follows: 1. The number of items administrated by BNAT is 17.77. 14.23 items are omitted and prediction accuracy can reach 93.16％. The prediction accuracy of bugs and skills can reach 95.28％. 2. This digital individual teaching materials making by knowledge structure and Bayesian network is adaptable to two kinds of instruction models. It shows that this digital individual instruction material is valuable for using. 3. After taking the digital instruction, the effectiveness of individual instruction model is better than that of group model. 4. In point of the students for the average mark group and low mark group, the instruction model of one v.s. two is better than one v.s. a group. In point of the students for the high mark group, the two kinds of instruction models have no significance difference. 5. In point of the schoolboys, the instruction model of one v.s. two is better than one v.s. a group. In point of the schoolgirls, the two kinds of teaching models have no significance difference.
