李一野;邓浩江;

• 1:中国科学院声学研究所国家网络新媒体工程技术研究中心
• 2:中国科学院大学

摘要(Abstract)：

结合多个模型集成学习可以提升单模型预测算法的性能,本文提出一种基于多特征融合的视频点击率预测方法,将哈希降维的特征和GBDT组合特征进行拼接作为输入特征,采用随机梯度下降法对逻辑回归、因子分解机和场感知因子分解机的输出值进行线性加权的迭代调整。实验结果表明该算法的预测效果优于基于单模型算法,也优于基于套袋方法的随机森林算法和基于平均法的其他集成算法,可以提高视频点击率预测精度。

关键词(KeyWords)： 点击率预测;多特征融合;集成学习

Abstract:

Keywords:

基金项目(Foundation): 先导专项课题：SEANET技术标准化研究与系统研制(编号：XDC02010701)

作者(Authors): 李一野;邓浩江;

参考文献(References)：

• [1] He X,Pan J,Jin O,et al.Practical Lessons from Predicting Clicks on Ads at Facebook[C]//international workshop on data mining for online advertising,2014:1-9.
• [2] Rendle S.Factorization Machines[C]//international conference on data mining,2010:995-1000.
• [3] Juan Y,Zhuang Y,Chin W,et al.Field-aware Factorization Machines for CTR Prediction[C]//conference on recommender systems,2016:43-50.
• [4] Río S D,López V,Benítez J M,et al.On the use of MapReduce for imbalanced big data using Random Forest[J].Information Sciences An International Journal,2014,285(3):112-137.
• [5] 匡俊，唐卫红，陈雷慧，等.基于特征工程的视频点击率预测算法[J].华东师范大学学报(自然科学版),2018,2018(3):77-87.
• [6] Juan Y,Lefortier D,Chapelle O,et al.Field-aware Factorization Machines in a Real-world Online Advertising System[C]//the web conference,2017:680-688.
• [7] Miyoshi S,Uezu T,Okada M.Statistical Mechanics of Time Domain Ensemble Learning[J].Journal of the Physical Society of Japan,2006,75(8):2652-2674.
• [8] Berk R A.An Introduction to Ensemble Methods for Data Analysis[J].Sociological Methods & Research,2006,34(3):263-295.
• [9] Johnson R W.An Introduction to the Bootstrap[J].Teaching Statistics,2001,23(2):49-54.
• [10] Prasad A M,Iverson L R,Liaw A,et al.Newer Classification and Regression Tree Techniques:Bagging and Random Forests for Ecological Prediction[J].Ecosystems,2006,9(2):181-199.
• [11] 吴桂平，侯晓琴，王冰，周军，张艳，颜永红.基于艺人画像的歌曲点播量预测[J].网络新媒体技术，2017,6(03):20-26.
• [12] Schapire R E.A brief introduction to boosting[C]//international joint conference on artificial intelligence,1999:1401-1406.
• [13] Quinlan J R.Induction of Decision Trees[J].Machine Learning,1986,1(1):81-106.
• [14] Ye J,Chow J,Chen J,et al.Stochastic gradient boosted distributed decision trees[C]//conference on information and knowledge management,2009:2061-2064.
• [15] Davis J,Goadrich M.The relationship between Precision-Recall and ROC curves[C]//international conference on machine learning,2006:233-240.
• [16] 刘会河，徐维超，刘舜.基于SVM的降维方法在三类ROC分析中的应用[J].计算机与现代化，2016(7):49-54
• [17] Bradley A P.The use of the area under the ROC curve in the evaluation of machine learning algorithms[J].Pattern Recognition,1997,30(7):1145-1159.
• [18] Meng L,Shi J,Wang H,et al.SVM with improved grid search and its application to wind power prediction[C]//international conference on machine learning and cybernetics,2013:603-609.
• [19] Collobert R,Weston J,Bottou L,et al.Natural Language Processing (Almost) from Scratch[J].Journal of Machine Learning Research,2011,12(1):2493-2537.
• [20] 孙娅楠，林文斌.梯度下降法在机器学习中的应用[J].苏州科技大学学报(自然科学版),2018,35(02):26-31.