计算机科学 ›› 2016, Vol. 43 ›› Issue (Z11): 557-563.doi: 10.11896/j.issn.1002-137X.2016.11A.126
朱奕健,张正卿,黄一清,白瑞瑞,严建峰
ZHU Yi-jian, ZHANG Zheng-qing, HUANG Yi-qing, BAI Rui-rui and YAN Jian-feng
摘要： 在电信运营商领域,外呼推荐是一种重要的推荐产品和服务的途径。实现了一种基于运营商大数据的自动外呼推荐系统,该系统能够挖掘用户的行为特征并且使用机器学习的方法预测用户对于被推荐产品的接受可能性。传统推荐系统使用的模型算法为矩阵分解、大规模稀疏特征分类、神经网络等。采用随机森林算法的主要原因是随机森林具有并行化程度高、训练速度快、生成的决策树可解释等诸多优点,适合于基于电信业数据的推荐系统。该外呼推荐系统基于Hadoop、Impala和Spark等大数据处理平台及工具,使用随机森林分类器作为核心算法,将用户最近的行为特征回归为接受外呼推荐产品的可能性。在线测试表明使用该系统与当前部署的人工随机外呼相比,能够提升约41%的用户接受率；同时,根据模型算法输出特征的重要性,进一步给出了两类用户的特征分析。
| [1] Sarwar B,Karypis G,Konstan J,et al.Application of dimensionality reduction in recommender system-a case study[R].Minnesota Univ Minneapolis Dept of Computer Science,2000 [2] Koren Y,Bell R,Volinsky C.Matrix factorization techniques for recommender systems[J].Computer,2009 (8):30-37 [3] Huang Yi-qing,Zhu Fang-zhou,Yuan Ming-xuan,et al.Telcochurn prediction with big data[M].SIGMOD,2015 [4] Yuan Ming-xuan,Deng Ke,Zeng Jia,et al.OceanST:A distributed analytic system for large-scale spatiotemporal mobile broadband data[C]∥VLDB (Demo).2014:1561-1564 [5] Page L,Brin S,Motwani R,et al.The PageRank Citation Ranking:Bringing Order to the Web[R].Stanford InfoLab,1999 [6] Zhu X,Ghahramani Z.Learning from labeled andunlabeled data with label propagation[R].Technical Report CMU-CALD-02-107,Carnegie MellonUniversity,2002 [7] Zeng J,Cheung W K,Liu J.Learning topic modelsby belief propagation[J].IEEE Trans.Pattern Anal.Mach.Intell.,2013,35(5):1121-1134 [8] Rendle S.Scaling factorization machines to relational data[C]∥PVLDB.2013:337-348 [9] Neslin S,Gupta S,Kamakura W A,et al.Defection Detection:Measuring and Understanding the Predictive Accuracy of Customer Churn Models[J].Social Science Electronic Publishing,2006,43(2):204-211 [10] Hadden J,Tiwari A,Roy R,et al.Computer assisted customerchurn management:State-of-the-art and future trends[J].Computers & Operations Research,2007,34(10):2902-2917 [11] Lima E.Domain knowledge integration in data mining using decision tables:case studies in churn prediction[J].Journal of the Operational Research Society,2009,60(8):1096-1106(11) [12] Verbeke W,Martens D,Mues C,et al.Building comprehensible customer churn prediction models with advanced rule induction techniques.[J].Expert Systems with Applications,2011,38(3):2354-2364 [13] Jinbo S,Xiu L,Wenhuang L.The Application ofAdaBoost inCustomer Churn Prediction[C]∥2007 International Conference on Service Systems and Service Management.IEEE,2007:1-6 [14] Lemmens A,Croux C.Bagging and boosting classification trees to predict churn[J].Journal of Marketing Research,2006,43(2):276-286 [15] Datta P,Masand B R,Mani D,et al.Automated Cellular Modeling and Prediction on a Large Scale[J].Artificial Intelligence Review,2000,14(6):485-502 [16] Hung S,Yen D C,Wang H.Applying data mining to telecomchurn management.[J].Expert Systems with Applications,2006,31:515-524 [17] Burez J,Van den Poel D.Handling class imbalance in customer churn prediction[J].Dirk Van den Poel,2008,36(3):4626-4636 [18] Davis J,Goadrich M.The Relationship Between Precision-Recall and ROC Curves[C]∥ICML ’06:Proceedings of the 23rd International Conference on Machine Learning.2006 [19] 方匡南,吴见彬,朱建平,等.随机森林方法研究综述[J].统计与信息论坛,2011,26(3):32-38 |
| No related articles found! |
|
||
首页