多源异构数据融合的智能商业选址推荐算法
|
摘要
随着社交媒体和基于位置网络服务的快速发展,基于海量用户生成数据进行智能推荐成为研究热点。然而,已有工作主要面向在线产品推荐,在如何利用物理空间的多维异构数据进行推荐方面研究较少。文中以城市商业选址为背景,利用社交媒体上的用户签到数据、小区房价数据以及各种POI数据等多源城市数据,在数据预处理基础上进行多侧面商业特征和地理特征提取,提出基于随机森林的商业选址推荐方法。使用北京地区的多源城市数据建立模型,通过排序评价指标NDCG对实验结果进行评估。结果表明所提出的方法具有较好的推荐性能。
With rapid development of social media and location-based network services,the intelligent recommendation based on the user-generated mass data has thus become a hot research hotspot. However,the existing studies mainly focus on online item recommendation,and there are few studies that utilize multi-dimensional heterogeneous data in the physical environments for recommendation. The multi-source urban data such as users′ check-in data,housing price data,and various POI data in social media are used in this paper to extract multi-sided commercial features and spatial features on the basis of data preprocessing under background of urban commercial site selection. The commercial site selection recommendation method based on the random forest is proposed. The multi-source urban data in Beijing city are adopted to build the model to assess the experimental results with the ranking evaluation indicator NDCG. The experiment results indicate that the proposed method has better performance.
With rapid development of social media and location-based network services,the intelligent recommendation based on the user-generated mass data has thus become a hot research hotspot. However,the existing studies mainly focus on online item recommendation,and there are few studies that utilize multi-dimensional heterogeneous data in the physical environments for recommendation. The multi-source urban data such as users′ check-in data,housing price data,and various POI data in social media are used in this paper to extract multi-sided commercial features and spatial features on the basis of data preprocessing under background of urban commercial site selection. The commercial site selection recommendation method based on the random forest is proposed. The multi-source urban data in Beijing city are adopted to build the model to assess the experimental results with the ranking evaluation indicator NDCG. The experiment results indicate that the proposed method has better performance.
引文
[1]屈弘扬,於志文,田苗,等.基于LBSN的商业选址推荐系统的研究与实现[J].计算机科学,2015,42(9):33-36.QU Hongyang,YU Zhiwen,TIAN Miao. Research and implementation of business location recommendation system based on LBSN[J]. Computer Science,2015,42(9):33-36.
[2]张义,陈虞君,杜博文,等.智慧城市多模式数据融合模型[J].北京航空航天大学学报,2016,42(12):2683-2690.ZHANG Yi,CHEN Yujun,DU Bowen. Intelligent city multimode data fusion model[J]. Journal of Beijing University of Aeronautics and Astronautics,2016,42(12):2683-2690.
[3] KARAMSHUK D,NOULAS A,SCELLATO S,et al. Geo-spotting:mining online location-based services for optimal retail store placement[C]//Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining.[S.l.]:ACM,2013:793-801.
[4] LI Y,ZHENG Y,JI S,et al. Location selection for ambulance stations:a data-driven approach[C]//Proceedings of the 23rd SIGSPATIAL International Conference on Advances in Geographic Information Systems.[S.l.]:ACM,2015:85-91.
[5] FU Y,GE Y,ZHENG Y,et al. Sparse real estate ranking with online user reviews and offline moving behaviors[C]//Proceedings of 2014 IEEE International Conference on Data Mining(ICDM).[S.l.]:IEEE,2014:120-129.
[6] JENSEN P. Network-based predictions of retail store commercial categories and optimal locations[J]. Physical review E,2006,74(3):035101.
[7] LI J,GUO B,WANG Z,et al. Where to place the next outlet? harnessing cross-space urban data for multi-scale chain store recommendation[C]//Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing:Adjunct.[S.l.]:ACM,2016:149-152.
[8] WANG Y,WANG L,LI Y,et al. A theoretical analysis of NDCG ranking measures[C]//Proceedings of the 26th Annual Conference on Learning Theory(COLT 2013). 2013:8-13.
[9] SUáREZ-VEGA R, SANTOS-PE?ATE D R, DORTAGONZáLEZ P. Location models and GIS tools for retail site location[J]. Applied geography,2012,35(1/2):12-22.
[10] SUáREZ-VEGA R, SANTOS-PE?ATE D R, DORTAGONZáLEZ P,et al. A multi-criteria GIS based procedure to solve a network competitive location problem[J]. Applied geography,2011,31(1):282-291.
[11]瓦普尼克.统计学习理论的本质[M].北京:清华大学出版社,2000.VAPNIK V N. The essence of Statistical Learning Theory[M].Beijing:Tsinghua University Press,2000.
[12]张学工.关于统计学习理论与向量机[J].自动化学报,2000,26(1):32-42.ZHANG Xuegong. On statistical learning theory and vector machine[J]. Journal of automation,2000,26(1):32-42.
[13]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.FANG Kuangnan,WU Jianbin,ZHU Jianping,et al. Review of research on stochastic forest methods[J]. Statistics and information forum,2011,26(3):32-38.
[2]张义,陈虞君,杜博文,等.智慧城市多模式数据融合模型[J].北京航空航天大学学报,2016,42(12):2683-2690.ZHANG Yi,CHEN Yujun,DU Bowen. Intelligent city multimode data fusion model[J]. Journal of Beijing University of Aeronautics and Astronautics,2016,42(12):2683-2690.
[3] KARAMSHUK D,NOULAS A,SCELLATO S,et al. Geo-spotting:mining online location-based services for optimal retail store placement[C]//Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining.[S.l.]:ACM,2013:793-801.
[4] LI Y,ZHENG Y,JI S,et al. Location selection for ambulance stations:a data-driven approach[C]//Proceedings of the 23rd SIGSPATIAL International Conference on Advances in Geographic Information Systems.[S.l.]:ACM,2015:85-91.
[5] FU Y,GE Y,ZHENG Y,et al. Sparse real estate ranking with online user reviews and offline moving behaviors[C]//Proceedings of 2014 IEEE International Conference on Data Mining(ICDM).[S.l.]:IEEE,2014:120-129.
[6] JENSEN P. Network-based predictions of retail store commercial categories and optimal locations[J]. Physical review E,2006,74(3):035101.
[7] LI J,GUO B,WANG Z,et al. Where to place the next outlet? harnessing cross-space urban data for multi-scale chain store recommendation[C]//Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing:Adjunct.[S.l.]:ACM,2016:149-152.
[8] WANG Y,WANG L,LI Y,et al. A theoretical analysis of NDCG ranking measures[C]//Proceedings of the 26th Annual Conference on Learning Theory(COLT 2013). 2013:8-13.
[9] SUáREZ-VEGA R, SANTOS-PE?ATE D R, DORTAGONZáLEZ P. Location models and GIS tools for retail site location[J]. Applied geography,2012,35(1/2):12-22.
[10] SUáREZ-VEGA R, SANTOS-PE?ATE D R, DORTAGONZáLEZ P,et al. A multi-criteria GIS based procedure to solve a network competitive location problem[J]. Applied geography,2011,31(1):282-291.
[11]瓦普尼克.统计学习理论的本质[M].北京:清华大学出版社,2000.VAPNIK V N. The essence of Statistical Learning Theory[M].Beijing:Tsinghua University Press,2000.
[12]张学工.关于统计学习理论与向量机[J].自动化学报,2000,26(1):32-42.ZHANG Xuegong. On statistical learning theory and vector machine[J]. Journal of automation,2000,26(1):32-42.
[13]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.FANG Kuangnan,WU Jianbin,ZHU Jianping,et al. Review of research on stochastic forest methods[J]. Statistics and information forum,2011,26(3):32-38.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |