The Research on E-mail Users' Behavior of Participating in Subjects Based on Social Network Analysis

The e-mail network is a type of social network. This study analyzes user behavior in e-mail subject participation in organizations by using social network analysis. First, the Enron dataset and the position-related information of an employee are introduced, and methods for deletion of false data are presented. Next, the three-layer model(User, Subject, Keyword) is proposed for analysis of user behavior. Then, the proposed keyword selection algorithm based on a greedy approach, and the influence and propagation of an e-mail subject are defined. Finally, the e-mail user behavior is analyzed for the Enron organization. This study has considerable significance in subject recommendation and character recognition.

Container Networking Performance Analysis for Large-Scale User Behavior Simulation

Accurately simulating large-scale user behavior is important to improve the similarity between the cyber range and the real network environment. The Linux Container provides a method to simulate the behavior of large-scale users under the constraints of limited physical resources. In a container-based virtualization environment, container networking is an important component. To evaluate the impact of different networking methods between the containers on the simulation performance, the typical container networking methods such as none, bridge, macvlan were analyzed, and the performance of different networking methods was evaluated according to the throughput and latency metrics. The experiments show that under the same physical resource constraints, the macvlan networking method has the best network performance, while the bridge method has the worst performance. This result provides a reference for selecting the appropriate networking method in the user behavior simulation process.

Evaluation of Microblog Users’ Influence Based on PageRank and Users Behavior Analysis

This paper explores the uses’ influences on microblog. At first, according to the social network theory, we present an analysis of information transmitting network structure based on the relationship of following and followed phenomenon of microblog users. Informed by the microblog user behavior analysis, the paper also addresses a model for calculating weights of users’ influence. It proposes a U-R model, using which we can evaluate users’ influence based on PageRank algorithms and analyzes user behaviors. In the U-R model, the effect of user behaviors is explored and PageRank is applied to evaluate the importance and the influence of every user in a microblog network by repeatedly iterating their own U-R value. The users’ influences in a microblog network can be ranked by the U-R value. Finally, the validity of U-R model is proved with a real-life numerical example.

On-Line Privacy Behavior: Using User Interfaces for Salient Factors

The problem of privacy in social networks is well documented within literature;users have pri- vacy concerns however, they consistently disclose their sensitive information and leave it open to unintended third parties. While numerous causes of poor behaviour have been suggested by re- search the role of the User Interface (UI) and the system itself is underexplored. The field of Per- suasive Technology would suggest that Social Network Systems persuade users to deviate from their normal or habitual behaviour. This paper makes the case that the UI can be used as the basis for user empowerment by informing them of their privacy at the point of interaction and remind- ing them of their privacy needs. The Theory of Planned Behaviour is introduced as a potential theoretical foundation for exploring the psychology behind privacy behaviour as it describes the salient factors that influence intention and action. Based on these factors of personal attitude, subjective norms and perceived control, a series of UIs are presented and implemented in con- trolled experiments examining their effect on personal information disclosure. This is combined with observations and interviews with the participants. Results from this initial, pilot experiment suggest groups with privacy salient information embedded exhibit less disclosure than the control group. This work reviews this approach as a method for exploring privacy behaviour and propos- es further work required.

Factors influencing knowledge contribution:An empirical investigation of social networking website users

Purpose: In the Web 2.0 era,leveraging the collective power of user knowledge contributions has become an important part of the study of collective intelligence. This research aims to investigate the factors which influence knowledge contribution behavior of social networking sites(SNS) users.Design/methodology/approach: The data were obtained from an online survey of 251 social networking sites users. Structural equation modeling analysis was used to validate the proposed model.Findings: Our survey shows that the individuals' motivation for knowledge contribution,their capability of contributing knowledge,interpersonal trust and their own habits positively influence their knowledge contribution behavior,but reward does not significantly influence knowledge contribution in the online virtual community.Research limitations: Respondents of our online survey are mainly undergraduate and graduate students. A limited sample group cannot represent all of the population. A larger survey involving more SNS users may be useful.Practical implications: The results have provided some theoretical basis for promoting knowledge contribution and user viscosity.Originality/value: Few studies have investigated the impact of social influence and user habits on knowledge contribution behavior of SNS users. This study can make a theoretical contribution by examining how the social influence processes and habits affect one's knowledge contribution behavior using online communities.

Subtle role of latency for information diffusion in online social networks

Information diffusion in online social networks is induced by the event of forwarding information for users, and latency exists widely in user spreading behaviors. Little work has been done to reveal the effect of latency on the diffusion process. In this paper, we propose a propagation model in which nodes may suspend their spreading actions for a waiting period of stochastic length. These latent nodes may recover their activity again. Meanwhile, the mechanism of forwarding information is also introduced into the diffusion model. Mean-field analysis and numerical simulations indicate that our model has three nontrivial results. First, the spreading threshold does not correlate with latency in neither homogeneous nor heterogeneous networks, but depends on the spreading and refractory parameter. Furthermore, latency affects the diffusion process and changes the infection scale. A large or small latency parameter leads to a larger final diffusion extent, but the intrinsic dynamics is different. Large latency implies forwarding information rapidly, while small latency prevents nodes from dropping out of interactions. In addition, the betweenness is a better descriptor to identify influential nodes in the model with latency, compared with the coreness and degree. These results are helpful in understanding some collective phenomena of the diffusion process and taking measures to restrain a rumor in social networks.

社交网络平台用户特征识别及其行为演变--以微博为例

根据微博用户特征进行聚类,识别不同类别的用户特征;依据行为数据,分析用户不同使用行为的差异。对于优化社交网络平台、推进用户持续使用等具有重要意义。使用八爪鱼采集器抓取微博平台1200个用户的样本数据以及13840个微博数据,运用数理统计方法进行统计分析和归纳,探讨用户个人特征和行为规律。研究结果显示,微博用户无论在用户特征变量还是行为变量方面,均存在高度相关性。通过聚类结果进行不同类别用户特征的比较。在微博使用中发现持续使用、中较使用、停止使用等用户不同使用行为以及动态特征。

一种融合表征的农产品推荐算法

针对农产品电商平台,产品季节性强、地域性强、用户行为多变,导致推荐效果不理想的问题,提出了一种融合表征的农产品推荐算法。首先,用长短期记忆网络和注意力网络相结合组成深度兴趣网络,以此来捕获物品的潜在特征;其次,构建用户-商品二部图;再次,利用图神经网络提取图数据的连接信息对每个节点的影响,并更新节点的嵌入式表示,以获取用户的潜在特征;最后,将两种潜在特征通过多层感知机得到待推荐农产品的购买概率,进一步提取和利用了用户行为序列中的用户深度兴趣,并将其融合深度兴趣网络进行推荐。实验结果表明:融合表征的农产品推荐算法相较于原有模型AUC指标提高9%以上,准确率和召回率提高约6%以上;相较于不考虑节点嵌入式表示的情况,AUC和准确率、召回率也均有提高。

基于异质信息网络表征学习的微博虚假信息甄别研究

[研究目的]社会网络的飞速发展与突发公共卫生事件的时有发生,使得大量的、具有迷惑性的虚假信息混杂社交媒体中,甄别此类信息已成为用户信息素养的重要组成部分。[研究方法]以微博上的异质信息网络为研究对象,综合考虑微博文本的语义特征和用户行为的非语义特征,引入多头注意力机制融合生成集成表示实现虚假信息甄别,并从信息内容、参与用户、用户与信息交互三个维度进行特征挖掘与量化分析。[研究结论]研究表明,基于异质信息网络表征学习的虚假信息甄别方法具有较好的实用性,有助于解构虚假信息的特征,为突发公共卫生健康事件的虚假信息治理及辩症施策提供有益参考。

面向加密流量的社交软件用户行为识别

随着智能终端和社交网络越来越融入人们的日常生活,针对社交软件的用户行为识别在网络管理、网络环境监管和市场调研等方面发挥越来越重要的作用.社交软件普遍使用端到端加密协议进行加密数据传输,现有方法通常提取加密数据的统计特征进行行为识别.但这些方法识别的性能不稳定且需要的数据量多,这些缺点影响了方法的实用性.提出了一种面向加密流量的社交软件用户行为识别方法.首先,从加密流量中识别出稳定的控制流数据,并提取控制服务数据分组负载长度序列.然后设计了2种神经网络模型,用于自动从控制流负载长度序列中提取特征,细粒度地识别用户行为.最后,以WhatsApp为例进行了实验,2种神经网络模型对WhatsApp用户行为的识别精准率、召回率和F1-score均超过96%.与类似工作的实验比较证明了该方法识别性能的稳定性,此外,该方法能够通过很少的控制流数据分组达到较高的识别精准率,对实时行为识别的研究具有重要的现实意义.

基于CPSS视角的需求响应能力评估综述

对信息—物理—社会系统与需求响应评估的结合进行综述。首先对信息物理系统的历史和框架进行分析。接着总结了现有的需求响应研究现状,包括需求响应潜力评估的意义、分类和评估方法。对于需求响应能力评估的数据来源进行总结,包括问卷调查和运行数据采集。在信息—物理—社会系统与需求响应结合方面,分别分析需求响应的物理域、信息域和社会域基础,介绍相应的建模方法和研究内容。最后,对未来的研究展望进行讨论,包括市场考核机制、快速仿真与建模技术以及综合能源下的需求响应管理。

基于决策树的社交网络隐式用户行为数据挖掘方法

为了解决社交网络隐式用户行为数据挖掘过程中关联相似性计算较为困难的问题,提出了基于决策树的社交网络隐式用户行为数据挖掘方法。将社交网络视为包含不同维度的向量空间,计算特定维度上用户的兴趣空间和兴趣点。确定样本属性集后,根据已知行为数据建立测试分支,计算该分支下子集的属性权重,不断迭代直至挖掘到同等属性的数据点为止。测试结果表明:该方法可对不同种类隐式用户行为精准挖掘,目标行为数据查找效果较好,实用性较强。

网络结构视角下用户异质性对知识协作效果的影响研究

[目的/意义]探讨如何优化在线知识协作平台网络结构和平衡团队内部的知识与经验,为促进跨领域信息流通,避免信息孤岛的形成,通过集体知识协作,推动知识的创造、传播和应用。[方法/过程]本研究利用编程问答网站Stack Overflow的数据,筛选出含“Python”标签的问题帖和对应的回答帖,结合负二项回归模型探讨了网络结构特征和用户异质性对知识协作质量和效率的影响。[结果/结论]网络结构特征方面,节点中心性显著提升了协作的质量和效率,较高的聚集系数和较大的连边跨度限制了信息流通,不利于知识协作效率。用户异质性方面,知识背景和注册时长的高异质性通常阻碍协作,经验异质性中注册时长异质性对协作效果均产生负面影响,回答被采纳率异质性仅对协作质量产生了负面影响,而活动强度异质性则产生了积极影响。

在线众包健康社区用户社会化交互网络结构与特征

在线众包健康社区用户通过社会化交互活动实现知识获取、共享与交流,形成了复杂社会网络结构。深入解析社会化交互网络结构,有助于理解用户在网络空间里的聚集方式,以及特定网络结构下用户行为形成的路径和意义。结合社会网络相关理论,确定在线心理服务平台用户社会化交互网络结构的分析框架,爬取国内知名在线心理服务平台壹点灵的用户交互数据,分别对用户社会化交互的整体网络结构、内部子结构网络、个体网络结构以及各网络结构下用户的行为特征进行深入剖析,为在线心理服务平台全面了解用户关系网络与行为特征,从而为发掘用户价值提供有效指导。

基于Python的社交网站用户行为数据采集方法

传统数据采集方法存在适用范围较窄、重复性工作量较大等问题,导致社交网站用户行为数据采集效率较差,提出基于Python的社交网站用户行为数据采集方法。采用情境标记法确定社交网站用户行为数据采集时机,基于Python语言搭建一个以MJU采样算法为URL地址管理中心的Scrapy爬虫框架,执行Scrapy爬虫框架完成社交网站用户行为数据的采集流程。实验结果表明,本文方法在采集社交网站的用户行为数据时,采集速率为830个/h,验证了该方法具有快速性。

基于深度学习的用户和实体行为分析技术

针对大数据环境中内部攻击行为难以有效防范的问题,在深入研究用户和实体行为分析(UEBA)技术的基础上,提出基于深度学习的用户和实体行为分析方案,并结合相关数据集进行实验分析。首先利用UEBA技术,构建单位员工和系统设备的正常活动基线、用户行为模式画像;其次使用基于深度学习的多网络模型架构,实现对内部员工窃取敏感数据、账号盗用攻击和针对Web业务系统API的异常访问请求的精准检测和异常评分。实验结果表明:单个网络模型中多层感知器的准确度最高,循环神经网络次之,径向基函数网络相对较差;相比单个网络模型,结合3个神经网络模型的多网络模型准确度有了一定的提升,误判率更低,具有一定的实际运用意义。

行为增强的多层次协同Top-N推荐

传统推荐系统只利用单一用户行为,然而用户行为间是具有关联性,忽视用户行为会丢失辅助行为对目标行为的影响。本文提出了一种行为增强的多层次协同Top-N推荐,在推荐二分图和元路径图上利用注意力机制传播信息,学习多层次高阶和异质协同信号(包括用户-项目间的和项目间的)以提高推荐性能,这样可以更好地利用推荐图结构,并充分考虑到推荐图结构上各种行为间的相互影响。在经典数据集上做了全方位实验验证模型有效性,在电商推荐数据上取得了很好效果。

面向未来千兆宽带应用的用户行为分析与网络优化

针对千兆宽带应用,分析其中的用户行为和网络优化方法。采集数据并分析用户行为特征,以准确掌握用户需求变动,评估千兆宽带网络性能指标并分析其局限性,识别网络性能的关键指标,运用以用户行为为依据的网络优化策略,如智能路由、负载均衡、差分服务流量调度及软件定义网络(Software Defined Network,SDN)切片技术,提升网络服务质量,保障安全与隐私,为实现未来千兆宽带应用的智能、高效、安全网络环境贡献力量。

基于用户行为特征的网络教学云平台资源个性化推荐方法

针对网络教学云平台的数据量大导致推荐的相关度低的问题,文章提出基于用户行为特征的网络教学云平台资源个性化推荐方法。该方法通过采集历史数据,筛选关键特征,结合K-中心聚类算法挖掘用户行为特征,建立与资源特征关联规则,结合偏好因子,为用户生成资源个性化推荐列表。实验测试结果表明,该方法推荐相关度高,满足平台实际应用需求。

基于深度学习的电力设备智能运行方式研究

在电力系统中对用户行为数据的分析过程存在主观性强的问题,且在复杂电网环境下难以做出准确的判断。对此,文中提出一种基于改进K-means聚类算法和深度学习相结合的分析算法。在改进K-means聚类算法的基础上构造出用户行为数据分析模型,实现了对行为数据的自适应学习。通过HRF-TCN预测模型筛选出用户行为数据的关键特征,降低数据维度后利用时间卷积网络进行预测,并将预测结果传递给电力系统中的设备,使其智能化地调整运行方式。基于公开用电数据集对算法进行的可行性验证结果表明,所提算法能够对用户的行为数据进行准确分析,数据分类效果指标DBI、SC分别为0.826 4和0.440 1,预测指标Ac和F1分别为96.8%、0.967 8,与同类算法相比,其分类效果更好、精度更高。