Abstract—In China, the prediction of the examination
achievement is important in the field of online learning, for both
the learners and the service providers. On the other hand, the
learners' online learning time, e.g. the time of watching the
teaching video, can be acquired and recorded easily on most
learning platforms. In this paper, we try to establish an EKF
model on the exam/test achievement and the learning time, and
use the model to predict the achievement in the next exam. The
experiment on a two-semester network education course proves,
the method can achieve the prediction precision, which is not
inferior to the classic methods, and meanwhile own some
advantages such as, the capability of noise resistance, uniform
convergence, and controllable complexity etc.
Index Terms—Achievement prediction, extended Kalman
filter, online learning.
Jun Xiao is with Shanghai Open University, Shanghai, China (e-mail:
xiaoj@shtvu.edu.cn).
Hongliang Gu was with Shanghai Jiao Tong University, Shanghai, China
(e-mail: hlgu@sjtu.edu.cn).
[PDF]
Cite: Jun Xiao and Hongliang Gu, "An Achievement Prediction Method on the Video Learning Time Based on EKF," International Journal of Future Computer and Communication vol. 6, no. 3, pp. 102-105, 2017.
