A hidden Markov model of library users classification
Su Qiong · 2010
User classification is a critical problem in modern library. But there has been little research into intelligent methods for improving the accuracy of user classification quality. In this article, we propose a HMM of library user classification (HMMLUC) to recognize the different type of library users. Using user previous behavior, HMMLUC learns a probabilistic model over the types of the user, and then applies this model at every step of the data entry process to improve user service quality. The experiment results proof that the performance of the model is sound. The average precision ratio of the HMMLUC is 91.2, while the recall ratio is 85.3.