计算机应用研究2016，Vol.33Issue(4)：979-982,4.DOI:10.3969/j.issn.1001-3695.2016.04.004
基于LDA的互联网广告点击率预测研究
Research on click-through rate prediction of Internet advertising based on LDA
摘要
Abstract
Advertisement click-through rate is essential for Internet advertising.Therefore,estimating click-through rate pre-cisely makes significant influence in the efficiency of advertising on the Internet.During the training of predicting models, many problems will arise such as the massive scale of advertisements and users,and the sparseness of training set,which usu-ally lead to a low accuracy of the predictive click-through rate.In order to solve these problems,this paper proposed an algo-rithm named LDA-FMs,which was a kind of predicting click rate algorithm based on LDA.Specifically,LDA-FMs partitioned the original training sets according to different topics,and then built click-through rate prediction models respectively upon dif-ferent topics using partitioned sub-training sets.On this basis,it calculated the advertisement click-through rate by using the probability of advertisement belonged to different topics and the combined with prediction result of every prediction model.The experiment based on real data sets from KDD Cup 2012-Track2,proves the feasibility and validity of this method.关键词
计算广告/点击率/主题模型/因子分解机Key words
computational advertising/click-through rate/topic model/factorization machines分类
信息技术与安全科学引用本文复制引用
朱志北,李斌,刘学军,胡平..基于LDA的互联网广告点击率预测研究[J].计算机应用研究,2016,33(4):979-982,4.基金项目
国家公益性科研专项基金资助项目（201310162）；连云港科技支撑计划资助项目 （）
