基于出行意图的潜在高价值旅客发现概率模型
|
摘要
由于潜在高价值旅客当前乘机历史记录少,较难被航空公司准确发现并关注.对此,提出基于出行意图的潜在高价值旅客发现概率模型.首先建立一个基于统计的潜在高价值旅客发现概率模型,再将旅客出行意图引入概率模型,发现旅客潜在航线需求,优化旅客潜在价值计算,从而通过出行意图发现潜在高价值旅客.实验结果表明,相比于次数法、里程法以及RFM模型等传统的旅客价值度量方法,基于出行意图的潜在高价值旅客发现概率模型能够有效识别潜在高价值旅客.
Potential high-value passengers can not be effectively discovered by airways due to the limited historical booking records of passengers. Aiming at this problem,a probabilistic model for discovering potential high-value passengers based on trip purposes mining is proposed. Firstly,we present a probabilistic model based on statistics to measure the value of passengers. Then,trip purposes are introduced into the model to discover potential airline demands of each passenger and to optimize passenger potential value calculation. Therefore,potential high-value passengers can be discovered through the trip purposes mining. Experiments show that the proposed model can identify the potential high-value passengers more accurately than the traditional passenger value evaluation methods based on the passengers' cumulative number of flight times,passengers' cumulative mileage and recency frequency monetry model.
Potential high-value passengers can not be effectively discovered by airways due to the limited historical booking records of passengers. Aiming at this problem,a probabilistic model for discovering potential high-value passengers based on trip purposes mining is proposed. Firstly,we present a probabilistic model based on statistics to measure the value of passengers. Then,trip purposes are introduced into the model to discover potential airline demands of each passenger and to optimize passenger potential value calculation. Therefore,potential high-value passengers can be discovered through the trip purposes mining. Experiments show that the proposed model can identify the potential high-value passengers more accurately than the traditional passenger value evaluation methods based on the passengers' cumulative number of flight times,passengers' cumulative mileage and recency frequency monetry model.
引文
[1]Lin Youfang,Wan Huaiyu,Jiang Rui,et al. Inferring the travel purposes of passenger groups for better understanding of passengers[J]. IEEE Transactions on Intelligent Transportation Systems,2015,16(1):235-243.
[2]Wang Jingjing,Chen Xi,Chen Zhihong,et al. Cluster algorithm based on LDA model for public transport passengers'trip purpose identification in specific area[C]∥Proceedings of the 2016 IEEE International Conference on Intelligent Transportation Engineering(ICITE). Washington DC:IEEE,2016:186-192.
[3]Feng Xia,Xu Bingyu,Lu Min,et al. Potential high-value passengers discovery by random walk on passengerroute heterogeneous network[J]. Journal of Computational and Theoretical Nanoscience,2015,12(8):2217-2222.
[4]于洪,李俊华.一种解决新项目冷启动问题的推荐算法[J].软件学报,2015,26(06):1395-1408.Yu Hong,Li Junhua. Algorithm to solve the cold-start problem in new item recommendations[J]. Journal of Software,2015,26(06):1395-1408.
[5]曹建平,王晖,夏友清等.基于LDA的双通道在线主题演化模型[J].自动化学报,2014,40(12):2877-2886.Cao Jianping,Wang Hui,Xia Youqing,et al. Bi-path evolution model for onlinetopic model based on LDA[J].ACTA Autmatic Sinica,2014,40(12):2877-2886.
[6]郭蓝天,李扬,慕德俊,等.一种基于LDA主题模型的话题发现方法[J].西北工业大学学报,2016,34(4):698-702.Guo Lantian,Li Yang,Mu Dejun,et al. A LDA model based topic detection method[J]. Journal of Northwestern Polytechnical University,2016,34(4):698-702.
[7]谢昊,江红.一种面向微博主题挖掘的改进LDA模型[J].华东师范大学学报(自然科学版),2013(6):93-101.Xie Hao,Jiang Hong. Improved LDA model for microblog topic mining[J]. Journal of East China Normal University(Natural Science),2013(6):93-101.
[8]Blei D M,Ng A Y,Jordan M I. Latent dirichlet allocation[J]. The Journal of Machine Learning Research,2003,3(3):993-1022.
[9]曹娟,张勇东,李锦涛,等.一种基于密度的自适应最优LDA模型选择方法[J].计算机学报,2008,31(10):1780-1787.Cao Juan,Zhang Yongdong,Li Jintao,et al. A method of adaptively selecting best LDA model based on density[J]. Chinese Journal of Computers,2008,31(10):1780-1787.
[2]Wang Jingjing,Chen Xi,Chen Zhihong,et al. Cluster algorithm based on LDA model for public transport passengers'trip purpose identification in specific area[C]∥Proceedings of the 2016 IEEE International Conference on Intelligent Transportation Engineering(ICITE). Washington DC:IEEE,2016:186-192.
[3]Feng Xia,Xu Bingyu,Lu Min,et al. Potential high-value passengers discovery by random walk on passengerroute heterogeneous network[J]. Journal of Computational and Theoretical Nanoscience,2015,12(8):2217-2222.
[4]于洪,李俊华.一种解决新项目冷启动问题的推荐算法[J].软件学报,2015,26(06):1395-1408.Yu Hong,Li Junhua. Algorithm to solve the cold-start problem in new item recommendations[J]. Journal of Software,2015,26(06):1395-1408.
[5]曹建平,王晖,夏友清等.基于LDA的双通道在线主题演化模型[J].自动化学报,2014,40(12):2877-2886.Cao Jianping,Wang Hui,Xia Youqing,et al. Bi-path evolution model for onlinetopic model based on LDA[J].ACTA Autmatic Sinica,2014,40(12):2877-2886.
[6]郭蓝天,李扬,慕德俊,等.一种基于LDA主题模型的话题发现方法[J].西北工业大学学报,2016,34(4):698-702.Guo Lantian,Li Yang,Mu Dejun,et al. A LDA model based topic detection method[J]. Journal of Northwestern Polytechnical University,2016,34(4):698-702.
[7]谢昊,江红.一种面向微博主题挖掘的改进LDA模型[J].华东师范大学学报(自然科学版),2013(6):93-101.Xie Hao,Jiang Hong. Improved LDA model for microblog topic mining[J]. Journal of East China Normal University(Natural Science),2013(6):93-101.
[8]Blei D M,Ng A Y,Jordan M I. Latent dirichlet allocation[J]. The Journal of Machine Learning Research,2003,3(3):993-1022.
[9]曹娟,张勇东,李锦涛,等.一种基于密度的自适应最优LDA模型选择方法[J].计算机学报,2008,31(10):1780-1787.Cao Juan,Zhang Yongdong,Li Jintao,et al. A method of adaptively selecting best LDA model based on density[J]. Chinese Journal of Computers,2008,31(10):1780-1787.