Hot topic trend prediction of topic based on markov chain and dynamic backtracking
点击次数:
所属单位:计算机科学与技术学院/人工智能学院/软件学院
发表刊物:Lect. Notes Comput. Sci.
摘要:Predicting topic trend in social networks can provide good reference value for public opinion guidance and commercial marketing. In this paper, we discuss the hot topic evaluation methods, and then present a method for evaluating the topic popularity of microblog based on multiple factors, which comprehending four factors (the number of micro blog, number of forwarding, number of comments, and number of praise) and using relative ranking method to define the value of micro blog popularity. In order to improve the prediction accuracy of hot topics, we present a prediction algorithm based on Markov chain and dynamic backtracking, which is based our evaluation method. In the algorithm, we use the simulated annealing method to find the optimal parameters and improve the accuracy of the prediction algorithm based on the Markov chain by historical backtracking. Analysis and simulation results demonstrate that the proposed algorithm is more accurate than some conventional methods. © Springer International Publishing AG, part of Springer Nature 2018.
ISSN号:0302-9743
是否译文:否
发表时间:2018-01-01
合写作者:Liu, Jue,He, Ying,Hou, Yating
通讯作者:许峰