登录 | 注册 | 退出 | 公司首页 | 繁体中文 | 满意度调查
综合馆
基于小数据的在线用户兴趣长程演化研究
  • 摘要

    网络大数据中与Web用户行为相关的数据,例如在线点击数据和通讯记录等,为人们深度挖掘和定量分析人类兴趣动力学带来了机遇,这些在线行为数据被称为大数据时代的“小数据”,有助于揭示许多复杂的人类社会与经济现象.Web用户行为建模时常见的前提假设就是人的行为符合M arkov过程,用户下一行为仅依赖于当前行为,与过去的历史行为无关.然而,在线用户行为是一个复杂过程,常常依赖于人的兴趣,对于人类兴趣动力学的本质规律目前知之甚少.利用中国互联网络信息中心提供的30000多名在线用户行为记录数据,基于块熵理论对在线用户行为进行分类研究,通过信息论分析方法,结合熵增曲线的离散导数和积分理论,分析在线用户点击行为的随机性和记忆性特征.研究表明,与常见的假设不同,Web用户的行为并不是一个简单的M arkov过程,而是一个符合幂率的非周期无限长程记忆过程;进一步还发现,用户在线连续点击7个兴趣点,其行为的平均预测增益就可达到95.3%以上,可为大数据时代在线用户兴趣精准预测提供理论指导.

  • 作者

    李勇  孟小峰  刘继  王常青  Li Yong  Meng Xiaofeng  Liu Ji  Wang Changqing 

  • 作者单位

    中国人民大学信息学院 北京 100872; 西北师范大学计算机科学与工程学院 兰州 730070/中国人民大学信息学院 北京 100872/新疆财经大学统计与信息学院 乌鲁木齐 830012/中国互联网络信息中心互联网基础技术开放实验室 北京 100190

  • 刊期

    2015年4期 ISTIC EI PKU

  • 关键词

    小数据  块熵  超熵  兴趣演化  预测增益  small data  block entropy  excess entropy  evolution of interest  predictability gain 

参考文献
  • [1] 朱夏,宋爱波,东方,罗军舟. 云计算环境下基于协同过滤的个性化推荐机制. 计算机研究与发展, 2014,10
  • [2] 孟小峰,慈祥. 大数据管理:概念、技术与挑战. 计算机研究与发展, 2013,1
  • [3] 孟小峰,李勇,祝建华. 社会计算:大数据时代的机遇与挑战. 计算机研究与发展, 2013,12
  • [4] Estrin D. Small data,where n= me. {H}Communications of the ACM, 2014,04
  • [5] Brockmann D;Hufnagel L;Geisel T. The scaling laws of human travel. {H}NATURE, 2006,7075
  • [6] Malmgren R D;Stouffer D B;Motter A E. A poissonian explanation for heavy tails in e-mail communication. Proceedings of the National Academy of Sciences, 2008,47
  • [7] Malmgren R D;Stouffer D B;Campanharo A S L O. On universality in human correspondence activity. {H}SCIENCE, 2009,5948
  • [8] Han X P;Zhou T;Wang B H. Modeling human dynamics with adaptive interest. {H}NEW JOURNAL OF PHYSICS, 2008,07
  • [9] Wu Y;Zhou C;Xiao J. Evidence for a bimodal distribution in human communication. Proceedings of the National Academy of Sciences, 2010,44
  • [10] Oliveira J G;Vazquez A. Impact of interactions on human dynamics. Physica A:Statistical Mechanics and its Applications, 2009,02
  • [11] Aly M;Hatch A;Josifovski V. Web-scale user modeling for targeting. New York:ACM, 2012
  • [12] Aly M;Pandey S;Josifovski V. Towards a robust modeling of temporal interest change patterns for behavioral targeting. New York:ACM, 2013
  • [13] Zheng N;Jin X;Li L. Cross-region collaborative filtering for new point-of-interest recommendation. New York:ACM, 2013
  • [14] Zhao Z;Yang Z;Zhang Z. Emergence of scaling in human-interest dynamics. Scientific Reports, 2013,3472
  • [15] Larson J W;Briggs P R;Tobis M. Block-entropy analysis of climate data. Procedia Computer Science, 2011,01
  • [16] Christoph B;Bernd P. Permutation entropy:A natural complexity measure for time series. {H}Physical Review Letters, 2002,17
  • [17] James P C;David P F. Regularities unseen,randomness observed:Levels of entropy convergence. {H}CHAOS, 2003,01
  • [18] Claypool M;Le P;Wased M. Implicit interest indicators. New York:ACM, 2001
  • [19] Shannon C E. A mathematical theory of communication. {H}Bell System Technical Journal, 1948,03
  • [20] Hao Bolin. Symbolic dynamics and characterization of complexity. Physica D:Nonlinear Phenomena, 1991,01
  • [21] Ebeling W;P?schel T. Entropy and long-range correlations in literary English. {H}Europhysics Letters, 1994,04
  • [22] Anishchenko V S;Ebeling W;Neiman A B. Power law distributions of spectral density and higher order entropies. {H}Chaos Solitons and Fractals, 1994,01
  • [23] Lazer D;Pentland A S;Adamic L. Computational social science. {H}SCIENCE, 2009,5915
  • [24] Lazer D;Kennedy R;King G. The parable of Google flu:T raps in big data analysis. {H}SCIENCE, 2014,6176
  • [25] Miller G A. The magical number seven,plus or minus two:Some limits on our capacity for processing information. {H}Psychological Review, 1956,01
  • [26] Wyner A D;Ziv J. Some asymptotic properties of the entropy of a stationary ergodic data source with applications to data compression. {H}IEEE Transactions on Information Theory, 1989,06
  • [27] Schurmann T;Grassberger P. Entropy estimation of symbol sequences. {H}CHAOS, 1996,03
  • [28] Ebeling W;Nicolis G. Entropy of symbolic sequences:The role of correlations. {H}Europhysics Letters, 1991,03
  • [29] Lesne A;Blanc J-L;Pezard L. Entropy estimation of very short symbolic sequences. {H}Physical Review E, 2009,04
  • [30] Chierichetti F;Kumar R;Raghavan P. Are web users really Markovian. New York:ACM, 2012
  • [31] Martyushev L M;Seleznev V D. Maximum entropy production principle in physics,chemistry and biology. {H}PHYSICS REPORTS-REVIEW SECTION OF PHYSICS LETTERS, 2006,01
  • [32] Wissner-Gross A D;Freer C E. Causal entropic forces. {H}Physical Review Letters, 2013,16
  • [33] Schr?dinger E. What is Life? The Physical Aspect of the Living Cell and Mind. Cambridge,UK:Cambridge University Press, 1967
  • [34] Mises L. Human Action. Auburn,Alabama:Ludwig von Mises Institute, 1998
  • [35] Barabasi A L. The origin of bursts and heavy tails in human dynamics. {H}NATURE, 2005,7039
查看更多︾
相似文献 查看更多>>
3.83.188.254