With the development of internet online education system, knowledge tracking (KT) is becoming more and more widely used. This technology can accurately model students’ learning process, so as to provide students with a personalized knowledge push. However, most previous KT methods don’t take into account the learning ability (LA) and forgetting behavior (FB) of student when assessing the state of knowledge. In fact, each student has their own LA and FB, which plays an important role in the KT process. However, at present, the personalized LA and FB of different students are not given in advance, which makes it more challenging to predict the learning status of each student. To address students’ challenges in modeling KT, we design augmenting knowledge tracing (AKT), which first uses concept-wised percent correct (CPC) to describe students’ overall mastery of knowledge, and builds an individualized forgetting rate (IFR) to describe the degree of forgetting during student learning process. The relationship between student history learning situation and time is considered, moreover, the knowledge mastery and FB are measured from the interaction with the topic during student learning process. The final experimental results show that the performance of proposed model is better than that of traditional methods.