Please wait a minute...
浙江大学学报(工学版)
浙江大学学报(工学版)  2019, Vol. 53 Issue (9): 1768-1778    DOI: 10.3785/j.issn.1008-973X.2019.09.016
计算机科学与人工智能     
城市兴趣点演化规律的可预测性分析
吴奇龙(),於志文*(),路新江,郭斌
西北工业大学 计算机学院，陕西 西安 710129
Analysis on predictability of urban point-of-interest evolution
Qi-long WU(),Zhi-wen YU*(),Xin-jiang LU,Bin GUO
School of Computer Science, Northwestern Polytechnical University, Xi’an 710129, China
 全文: PDF(1719 KB)   HTML
摘要：

建立兴趣点生命周期可预测性模型来刻画城市兴趣点演化规律的可预测性. 该模型针对城市兴趣点在连续时间窗口的变化情况，给出其生命周期长度和生命周期状态的定义，分析兴趣点在演化过程中的可预测性. 为量化城市兴趣点演化规律的可预测性，将城市兴趣点的生命周期长度和生命周期状态的信息熵同Fano不等式相结合，基于信息论中信息的不确定性度量方法给出城市兴趣点生命周期长度和生命周期状态的可预测性计算方法，并结合7个城市的城市兴趣点数据，计算出不同粒度、不同类别的兴趣点的可预测性. 结果表明：城市兴趣点的生命周期是可预测的且不同类别兴趣点的可预测性差异较大；相对于稳定状态和爆发状态的兴趣点，处于消亡状态的兴趣点的可预测性更高.
关键词： 城市兴趣点;  可预测性;  信息熵;  Fano不等式    
Abstract:

A PLCPA model was established to describe the predictability of urban point-of-interest evolution. Based on the change of urban interest points in continuous time window, a definition of the life cycle length and life cycle status was given and the predictability of point-of-interest evolution was analyzed. Fano’s inequality was combined with the life cycle length and status of urban point-of-interest to quantify the predictability of the evolution of urban point-of-interest. And the predictability calculation of life cycle length and status of urban point-of-interest was given based on the information uncertainty measure in the information theory. The predictability of different categories of point-of-interest with different levels of granularity was calculated according to the urban point-of-interest data of seven cities. Results show that the life cycle of urban interest points is predictable; the predictabilities of different categories of point-of-interest are quite different; the predictability of the points-of-interest in the decaying status is higher than that of the stable status and the booming status.
Key words: urban point-of-interest    predictability    information entropy    Fano’s inequality
收稿日期: 2018-12-17 出版日期: 2019-09-12
CLC:  TP 399  
通讯作者: 於志文     E-mail: 2014302721@mail.nwpu.edu.cn;zhiwenyu@nwpu.edu.cn
作者简介: 吴奇龙（1995—），男，硕士生，从事普适计算相关研究. orcid.org/0000-0003-3570-1539. E-mail: 2014302721@mail.nwpu.edu.cn
服务　　
把本文推荐给朋友
加入引用管理器
E-mail Alert
作者相关文章　　
吴奇龙
於志文
路新江
郭斌

引用本文:

吴奇龙,於志文,路新江,郭斌. 城市兴趣点演化规律的可预测性分析[J]. 浙江大学学报(工学版), 2019, 53(9): 1768-1778.

Qi-long WU,Zhi-wen YU,Xin-jiang LU,Bin GUO. Analysis on predictability of urban point-of-interest evolution. Journal of ZheJiang University (Engineering Science), 2019, 53(9): 1768-1778.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2019.09.016    或    http://www.zjujournals.com/eng/CN/Y2019/V53/I9/1768

图 1  兴趣点生命周期可预测性分析模型
图 2  特定地区的地图数据采集过程
类别 amenity building historic leisure shop sport tourism
北京 3 992 24 0 130 1 609 0 747
柏林 23 308 19 0 2 002 18 617 64 1 210
伦敦 24 364 42 0 982 14 905 61 981
纽约 13 191 27 0 1 848 4 167 24 368
巴黎 18 065 28 0 422 12 741 23 1 291
上海 3 092 18 0 279 1 432 12 322
东京 37 485 20 1 1 024 22 665 14 832
表 1  每一个城市的主类别对应的POI数量
图 4  POI生命周期长度的累积分布
图 3  POI数量的比例分布
图 5  纽约市不同类别POI生命长度的累积分布
图 6  生命周期状态分布
图 7  POI生命周期长度的熵值分布
图 8  POI生命周期长度的可预测性
图 9  leisure类别POI的生命周期长度频次分布
图 10  子类POI生命周期长度的可预测性
图 11  伦敦不同类别POI数量的分布
图 12  初级类别POI生命周期长度的可预测性
图 13  北京市不同生命周期状态的POI所占的比例
图 14  北京市不同类别POI所占比例
图 15  伦敦市不同类别、不同状态POI的可预测性
图 16  不同时间窗口对应的POI生命状态的可预测性
1 WANG J, LI C, XIONG Z, et al Survey of data-centric smart city[J]. Journal of Computer Research and De-velopment, 2014, 51 (2): 239- 259
2 YUAN Q, CONG G, MA Z, et al. Time-aware point-of-interest recommendation [C] // Proceedings of the 36th International ACM SIGIR Conference on Research and Development in Information Retrieval. New York: ACM, 2013: 363–372.
3 吕琳媛 复杂网络链路预测[J]. 电子科技大学学报, 2010, 39 (5): 651- 661
LV Lin-yuan Link Prediction on complex networks[J]. Journal of University of Electronic Science and Technology of China, 2010, 39 (5): 651- 661
4 KENNEDY L S, NAAMAN M. Generating diverse and representative image search results for landmarks [C] // Proceedings of the 17th International Conference on World Wide Web. Beijing: ACM, 2008: 297–306.
5 LIAN D, ZHAO C, XIE X, et al. GeoMF: joint geograph-ical modeling and matrix factorization for Point-of-Interest recommendation [C] // ACM Sigkdd International Conference on Knowledge Discovery and Data Mining. New York: ACM, 2014: 831–840.
6 CHENG C, YANG H, LYU M R, et al. Where you like to go next: successive Point-of-Interest recommendation [C] // International Joint Conference on Artificial Intelligence. Beijing: IJCAI. 2013: 2605–2611.
7 ZHAO W, LI Q, LI B Extracting hierarchical landmarks from urban POI data[J]. Yaogan Xuebao: Journal of Remote Sensing, 2011, 15 (5): 973- 988
8 方斌, 张宪, 杨柳 基于 POI 数据的城市边界变化提取研究——以山西运城市城区为例[J]. 现代测绘, 2017, 40 (5): 20- 22
FANG Bin, ZHANG Xian, YANG Liu Study on urban boundary identification based on POI: a case study of Yuncheng City in Shanxi[J]. Modern Surveying and Mapping, 2017, 40 (5): 20- 22
9 禹文豪, 艾廷华, 刘鹏程, 等 设施 POI 分布热点分析的网络核密度估计方法[J]. 测绘学报, 2015, 44 (12): 1378- 1383
YU Wen-hao, AI Ting-hua, LIU Peng-cheng, et al Network kernel density estimation for the analysis of facility POI hotspots[J]. Acta Geodaetica et Cartographica Sinica, 2015, 44 (12): 1378- 1383
10 YUAN J, ZHENG Y, ZHANG L, et al. Where to find my next passenger [C] // Proceedings of the 13th International Conference on Ubiquitous Computing. Beijing: ACM, 2011: 109–118.
11 YU G, YUAN J, LIU Z. Predicting human activities using spatio-temporal structure of interest points [C] // Proceed-ings of the 20th ACM International Conference on Multimedia. Nara: ACM, 2012: 1049–1052.
12 LU X, YU Z, SUN L, et al. Characterizing the life cycle of point of interests using human mobility patterns [C] // Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing. Heidelberg: ACM, 2016: 1052–1063.
13 LU X, YU Z, LIU C, et al. Forecasting the rise and fall of volatile point-of-interests [C] // 2017 IEEE International Conference on Big Data. Boston: IEEE, 2017: 1307–1312.
14 BIALEK W, NEMENMAN I, TISHBY N Predictability, complexity, and learning[J]. Neural Computation, 2001, 13 (11): 2409- 2463
15 ANG A, BEKAERT G Stock return predictability:Is it there?[J]. The Review of Financial Studies, 2006, 20 (3): 651- 707
16 SONG C, QU Z, BLUMM N, et al Limits of predictability in human mobility[J]. Science, 2010, 327 (5968): 1018- 1021
17 胡伟.改革开放40年中国工业经济发展的区域特征[J/OL].区域经济评论, 2019(1):1–15[2019-01-24]. Https://doi.org/10.14017/j.cnki.2095-5766.2019.0017.
HU Wei. Regional characteristics of China's industrial economic development in the 40 years of reform and openin up [J/OL]. Regional Economic Review, 2019(1): 1–15 [2019-01-24]. Https://doi.org/10.14017/j.cnki.2095-5766.2019.0017.
18 QIN S M, VERKASALO H, MOHTASCHEMI M, et al Patterns, entropy, and predictability of human mobility and life[J]. PLoS One, 2012, 7 (12): e51353
19 SINATRA R, SZELL M Entropy and the predictability of online life[J]. Entropy, 2014, 16 (1): 543- 556
20 NAVET N, CHEN S H. On predictability and profitability: would gp induced trading rules be sensitive to the observed entropy of time series? [M] // Natural Computing in Computational Finance. Berlin Heidelberg: Springer, 2008: 197-210.
21 AARONSON J, PARK K K. Predictability, entropy and information of infinite transformations [J]. arXiv preprint arXiv: 0705.2148, 2007.
22 DELSOLE T Predictability and information theory. Part I: measures of predictability[J]. Journal of the Atmospheric Sciences, 2004, 61 (20): 2425- 2440
23 BOLTZMANN L. Lectures on gas theory [M]. Chelmsford: Courier Corporation, 2012: 34.
[1] 王卫东,李俊杰,王京,傅庆湘,康文泓. 基于未确知测度的公路交通效率评价[J]. 浙江大学学报(工学版), 2016, 50(1): 48-54.
[2] 王宇良,郑海俊,周业丰,黄正梁,王靖岱,阳永荣. 多股进料搅拌釜停留时间分布的信息熵[J]. 浙江大学学报(工学版), 2015, 49(3): 590-597.
[3] 金伟良,李志远,许晨. 基于相对信息熵的混凝土结构寿命预测方法[J]. J4, 2012, 46(11): 1991-1997.
[4] 邓军,余忠华,吴昭同. 基于信息熵的概念设计的质量评价[J]. J4, 2009, 43(8): 1480-1484.
[5] 俞婷婷, 王卫标, 尤爱菊. 钱塘江下沙海塘越顶安全性分析[J]. J4, 2009, 43(8): 1543-1548.
[6] 黄鹏, 卜佳俊, 陈纯, 等. 利用加权特征模型改进问句分类[J]. J4, 2009, 43(6): 994-998.
[7] 束志恒 陈德钊. 用于连续属性离散化的RSE-Chi2方法[J]. J4, 2005, 39(6): 849-852.