Fast Mining Method of User Behavior Data based on Bayesian Network

Yuting Jiang

Information Engineering Institute, Jiangsu Maritime Institute, Nanjing, 211100, China

Abstract: The conventional user behavior data mining method can achieve the data mining of user behavior. However, due to the large volume of data filtering, there is a shortcoming of user behavior mining response time. Therefore, a fast mining method based on Bayesian network for user behavior data is proposed. The Bayesian network is introduced, the overall framework of the mining model is built and a fast mining algo-rithm for user behavior data is designed to complete the establishment of a fast mining model of user behavior; Based on the user behavior data selection mechanism, relying on the mining information input sequence and the construction of the output protocol, the user behavior can be quickly mined. The experimental data show that the proposed user behavior data mining method has a higher average response time of 47.23% compared with the conventional user behavior data mining method, and has high data mining accuracy, which is suitable for the rapid mining of user behavior data. 

Keywords: Bayesian network; User behavior; Data mining; Mining method