USST_Arts_117160727微博用户的行为动力学研究

VIP免费
3.0 赵德峰 2024-11-11 4 4 800.8KB 50 页 15积分
侵权投诉
的行为是杂多样的。于人的定和定的分是对
特性挖掘建模当前网络和人动力研究重要方
向。传统的人动力究中,大数学者程来
相继为发间隔是均的。,人高度的复
杂性机制很难某些为特性普适。许研究
表明的行偏离过程高度时间静默短时间
高频爆这也研究的关学者们开聚焦
特性研究试图找到某种建立相对其进行理的解
本论文以新浪微博究对,以人为动力理论基础合运用复
杂网络学和理学等法对微博网络拓扑结和内机制行了
,并从周发性间隔数分个方微博
微博论行行了入的,进了用
些共论和研究
微博户为节点以用之间的关连接边,构一个微博
网络过对微博网络扑结的分非线连接机制建立微博
演化,并对模行了实证仿真表明微博网络
点度布服从指数截断且具特性小世界
个体微博消息间间隔从幂且具
发性。大数用户发微博时间中中1214
1922 点这时间段内。受兴趣的影微博时间间隔也服从幂
微博论行为表指数截断
其中指数1.5 左右指数
1同的论数和用粉丝之间也
表明的行社会广泛
于上微博情控可以
节点免疫来对论的进行微博营销则可以借
领袖的影力来微博信息播范围
本文微博网络构、信息制和用的行为特性进行了入的
,有了解信息和用普世特性,并情控制、
信息推荐营销提供价值参考
关键微博 社交网络 传播 复杂网络 动力学
ABSTRACT
The human behavior is complex and diverse. The quantitative and qualitative
analysis of human behaviors, especially mining and modeling for the characteristics, is
an important direction of the research on complex networks and human behavior
dynamics currently.In the traditional studies of human dynamics, most scholars use to
describe human behavior in Poisson process which the time interval distribution of
successive acts is uniform. However, the human behavior is highly complex, single
drive mechanism is difficult to explain some of the behavioral characteristics, and has
no universality. Many studies show that the human behavior has deviated from Poisson
process with a high degree of non-uniformity: a long silence and a short high-frequency
outbreak. These results have caused the researchers' attention, more and more scholars
began to focus on studying user behavior characteristics, trying to find some law and
establish the model to have a reasonable explanation.
This paper study the object of Sina Weibo, based on the theoretical of human
behavior dynamics, adding with complex networks, statistical and management theory
to have a research on the microblogging network topology and internal drive
mechanisms. Then, have a depth statistical analysis of the users posting, forwarding
and comment behavior, respectively from the periodic, episodic, interval time
distribution, frequency distribution and correlation aspects, and get some behavior
universal characteristics inherent. The main conclusions and results are as follows:
The Weibo-blog could be regards a user behavior network based on the user notes
and relationship edges. By the analysis of network topology, we established a
microblogging network evolution model based on the nonlinear preferential attachment
mechanism. The empirical and simulation study results show that network node degree
distribution follows exponential truncated power-law distribution, and has scale-free
and small-world characteristics.
The blog posting interval time of individual and group users follow a power law
distribution, and have a significant periodic and episodic effect. Most users blog posting
time are centralized in 12~14 and 19~22 time periods. Driven by the interest, microblog
forwarding interval time also follows a power law distribution. The media, websites and
celebritys microblog forwarding and comment behavior follow as exponential
truncated power-law distribution, which the media and website exponent are about 1.5,
but the celebrity exponent is less than 1, showing two different statistical characteristics.
In addition, the relation between forwarding, comment and followings shows a positive
correlation, which indicates the user behavior is widespread effected by social concerns.
Based on the above analysis, there have some suggestions to the microblog
application. In public opinion control, we can use targeted immunization of the key note
to control the public opinion predation. As microblog marketing, microblogging
informations influence can expand by the help of opinion leaders and celebrity user.
In conclusion, this paper has a deep research on the microblog network structure,
information dissemination mechanism and user behavior characteristics, it has a great
help to understand the information propagation pattern and user behaviors universal
characteristics. Whats more, it can also provide some valuable references in the public
opinion control, information recommendation and marketing services.
Key Word: Micro-blog, Social network, Public opinion propagation,
Complex network, Human dynamics
摘要
ABSTRACT
…………………………………………1
1.1 选题背景研究 …………………………………………….....1
1.2 本文的工作 …………………………………………………….2
第二章 为动力学的理论基础和文 …………………………5
2.1 的分 ………………………………...5
2.1.1 泊松 ............................................................................. 5
2.1.2 指数截断 ............................................................................. 7
2.2 为动力学的本理论 ………………..8
2.2.1 排队 ......................................................................... 8
2.2.2 齐次泊松过程理论 ............................................................................. 9
2.2.3 兴趣动模 ...................................................................................... 10
2.2.4 记忆 ...................................................................................... 12
2.3 为动力学的研究 …………13
第三章 微博网络特性研究 ………………………….16
3.1 微博网络拓扑结 ……………………………………16
3.1.1 微博网络 .................................................................................. 16
3.1.2 微博网络拓扑 ............................................................................... 17
3.2 微博网络模 ………………………….19
3.2.1 .............................................................................................. 20
3.2.2 ............................................................................... 20
3.2.3 型仿真 ...................................................................................... 22
3.3 小结 …………………………………………………..23
第四章 微博为特性统 ………………...24
4.1 数据采 ………………………………………………...24
4.2 微博信息为统 ………………...25
4.2.1 个体微博布时间概率 .................................................... 25
4.2.2 微博 ........................................................... 26
4.3 微博信息 ………………27
4.3.1 微博时间间隔 ........................................................ 27
4.3.2 微博转论数分 .................................................... 28
4.4 为相 ………………………….29
4.5 小结 …………………………………………………..31
第五章 微博营销策略 ………………...32
5.1 微博论的 ………………………….32
5.1.1 论的 .................................................................................. 32
5.1.2 免疫策略 ...................................................................................... 33
5.1.3 免疫效 ...................................................................................... 35
5.2 微博营销策略 ……………………………………36
5.2.1 微博 .................................................................................. 36
5.2.2 实证 .............................................................................................. 37
5.2.3 微博营销策略 ...................................................................................... 38
5.3 小结 …………………………………………………..39
第六章 及展望 ………………………………………………...40
6.1 全文 …………………………………………………..40
6.2 研究展望 …………………………………………………..41
参考 …………………….42
公开发表的论文和承担目及取得 ……………………..46
…………………………………………………………………………47
摘要:

摘要人类的行为是复杂多样的。对于人类行为的定量和定性的分析,特别是对用户行为特性的挖掘和建模,是当前复杂网络和人类行为动力学研究的一个重要方向。在传统的人类动力学研究中,大多数学者利用泊松过程来描述人类行为,认为人们的相继行为发生的时间间隔分布是均匀的。但是,人类行为具有高度的复杂性,单一的驱动机制很难解释某些行为特性,且不具备普适性。许多研究结果表明人类的行为偏离了泊松过程,具有高度的非均匀性:长时间的静默和短时间内高频爆发。这也引起了研究者们的关注,越来越多的学者们开始聚焦用户行为特性的研究,试图找到某种规律并建立相应的模型对其进行合理的解释。本论文以新浪微博为研究对象,以人类行为动力学为...

展开>> 收起<<
USST_Arts_117160727微博用户的行为动力学研究.pdf

共50页,预览5页

还剩页未读, 继续阅读

作者:赵德峰 分类:高等教育资料 价格:15积分 属性:50 页 大小:800.8KB 格式:PDF 时间:2024-11-11

开通VIP享超值会员特权

  • 多端同步记录
  • 高速下载文档
  • 免费文档工具
  • 分享文档赚钱
  • 每日登录抽奖
  • 优质衍生服务
/ 50
客服
关注