计算机科学 ›› 2017, Vol. 44 ›› Issue (7): 191-196.doi: 10.11896/j.issn.1002-137X.2017.07.034
周先亭,黄文明,邓珍荣
ZHOU Xian-ting, HUANG Wen-ming and DENG Zhen-rong
摘要： 针对目前微博转发行为预测具有的特征选择任意性、准确率不高的问题,提出了融合异常检测与随机森林的微博转发行为预测方法。首先,提取用户基本特征、博文基本特征、博文内容主题特征,并基于相对熵计算用户活跃度、博文影响力；其次,通过结合过滤式与封装式特征选择方法筛选出关键特征组；最后,融合异常检测与随机森林算法,依据筛选后的关键特征组进行微博转发行为预测,并利用袋外数据误差估计设置随机森林中的决策树和特征数。在真实新浪微博数据集上与基于逻辑回归、决策树、朴素贝叶斯、随机森林等算法的微博转发行为预测方法进行实验对比,结果表明所提方法的预测准确率(90.5％) 高于基准方法中最优的随机森林方法的预测准确率,同时验证了特征筛选方法的有效性。
| [1] CAO J X,WU J L,SHI W,et al.Sina microblog information diffusion analysis and prediction[J].Chinese Journal of Compu-ters,2014,37(4):779-790.(in Chinese) 曹玖新,吴江林,石伟,等.新浪微博网信息传播分析与预测[J].计算机学报,2014,37(4):779-790. [2] 中国互联网络信息中心.第37次中国互联网络发展状况统计报告[R].北京:中国互联网络信息中心,2016. [3] PETROVIC S,OSBORNE M,LAVRENKO V.RT to win! Predicting message propagation in Twitter[C]∥ Proceedings of the Fifth International Conference on Weblogs and Social Media.Barcelonia,Spain,2011. [4] MORCHID M,DUFOUR R,BOUSQUET P M,et al.Feature selection using Principal Component Analysis for massive retweet detection[J].Pattern Recognition Letters,2014,49:33-39. [5] YANG Z,GUO J,CAI K,et al.Understanding retweeting be-haviors in social networks[C]∥Proceedings of the 19th ACM Conference on Information and Knowledge Management(CIKM 2010).Toronto,Ontario,Canada,2010:1633-1636. [6] ROMERO D M,MEEDER B,KLEINBERG J.Differences in themechanics of information diffusion across topics:idioms,political hashtags,and complex contagion on twitter[C]∥Proceedings of the 20th International Conference on World Wide Web(WWW 2011).Hyderabad,India,2011:695-704. [7] ZHANG Y,LU R,YANG Q.Predicting retweeting in microblogs[J].Journal of Chinese Information Processing,2012,26(4):109-114.(in Chinese) 张旸,路荣,杨青.微博客中转发行为的预测研究[J].中文信息学报,2012,26(4):109-114. [8] LI Y L,YU H T,LIU L X.Predict algorithm of micro-blog retweet scale based on SVM[J].Application Research of Computers,2013,30(9):2594-2597.(in Chinese) 李英乐,于洪涛,刘力雄.基于SVM的微博转发规模预测方法[J].计算机应用研究,2013,30(9):2594-2597. [9] ZHAO Y,SHAO B L,BIAN G Q,et al.Prediction of retweeting behavior for imbalanced dataset in microblogs[J].Journal of Computer Applications,2015,35(7):1959-1964.(in Chinese) 赵煜,邵必林,边根庆,等.面向不平衡微博数据集的转发行为预测方法[J].计算机应用,2015,35(7):1959-1964. [10] SUH B,HONG L,PIROLLI P,et al.Want to be Retweeted? Large scale analytics on factors impacting retweet in Twitter network[C]∥2010 IEEE International Conference on Social Computing / IEEE International Conference on Privacy,Security,Risk and Trust.IEEE,2010:177-184. [11] HU W.Real-time Twitter sentiment toward midterm exams[J].Sociology Mind,2012,2(2):177-184. [12] WU J H,ZUO K Z,JIE B,et al.New discriminative feature selection method[J].Journal of Computer Applications,2015,35(10):2752-2756.(in Chinese) 吴锦华,左开中,接标,等.新颖的判别性特征选择方法[J].计算机应用,2015,35(10):2752-2756. [13] KAR M,NUNES S,RIBEIRO C.Summarization of changes in dynamic text collections using Latent Dirichlet Allocation model[J].Information Processing & Management,2015,51(6):809-833. [14] LIU S P,YIN J,OUYANG J,et al.Topic mining from microblogs based on MB-HDP model[J].Chinese Journal of Compu-ters,2015(7):1408-1419.(in Chinese) 刘少鹏,印鉴,欧阳佳,等.基于MB-HDP 模型的微博主题挖掘[J].计算机学报,2015(7):1408-1419. |
| No related articles found! |
|
||
首页