ICS-SVM: A user retweet prediction method for hot topics based on improved SVM

• DOI: 10.1016/j.dcan.2021.07.003
• Publication Date: 2021-07-15T14:21:03Z
• Authors: Tianji Dai; Yunpeng Xiao; Xia Liang; Qian Li; Tun Li
• Keywords: Cuckoo search

In social networks, many complex factors affect the prediction of user forwarding behavior. This paper proposes an improved SVM method for behavior hot topics to improve accuracy. Firstly, we consider that Cuckoo Search algorithm can select optimal penalty parameters and kernel function optimize thus predict user's Secondly, this considers comprehensively from own external factors. Finally, based on characteristics changing over time, time-slicing is used trend topics. Experiments show accurately sense