隐私保护关联规则挖掘的研究

数据挖掘已经融入到社会生活的各个领域,随之个人隐私保护的研究也凸显其重要性。随着隐私保护在数据挖掘研究中的深入,人们发现一些规律,也相继的发现一些隐藏的问题。总结归纳目前人们在数据挖掘领域中对隐私保护关联规则研究的现状,分析目前基于隐私保护的数据挖掘主要方法的优缺点,给出隐私保护在未来数据挖掘中的发展方向。     

基于分簇和启发式的隐私保护算法研究

针对社交网络存在安全隐患和隐私风险,提出了基于分簇算法的启发式修改。基于最短路径,提出了一个最大贪心保护算法,追求数据隐私和数据功用的平衡。从数学分析和实验结果看,实现了一定的隐私保护和功用要求。     

隐私保护增强的用户兴趣模型的构建与研究

用户兴趣模型可提高搜索引擎的检索效率,为用户提供满足需求的检索结果,并在上下文检索中快速发展。用户在得到高质量检索结果的同时,也开始担忧个人敏感或重要信息的安全,本文在传统构建用户兴趣模型的方法基础上,增加隐私保护的功能,构建隐私保护增强的用户兴趣模型,解决用户的后顾之忧,为用户搭建一个安全的网络平台。     

基于大数据分析的隐私信息保护策略研究

网络在人们的生活中越来越重要,保障个人信息的安全成了重点议题。因此,该文进行了基于大数据分析的隐私信息保护策略研究。构建隐私信息保护模型,创建基于大数据分析的Agent保护信息算法,控制隐私数据的存储与访问情况。基于大数据分析的隐私信息保护策略研究的对比试验表明：信息保护策略能有效减少隐私信息被破解,使用该方法时隐私保护性能显著提高。     

大数据安全与隐私保护技术探究

随着大数据时代的到来,"大数据"已然成为广受社会各界关注的热点问题。大数据为人们的生活、生产等方方面面带来了积极的影响,然而也带来了一定的风险。如何在现有条件下保护大数据的安全与隐私,是一个亟待解决的重要问题。文章简要介绍了大数据,并结合当前大数据面临的安全考验,对大数据安全与隐私的保护技术进行了探究。     

互联网时代计算机隐私的保护研究

本文阐述了在互联网环境下关于计算机隐私权方面的现状,以及侵犯计算机隐私的方式,并且进一步提出如何保护计算机隐私权。     

基于Android平台的智能手机新型隐私保护应用研究与开发

智能手机隐私是用户便捷、安全地使用手机的保障和基础,为了更有效地保护智能手机用户隐私,本文从功能设计开发、技术研究等多方面阐述基于Android平台的智能手机新型隐私保护应用的理论支持和具体实现,在传统的程序锁应用上结合用户的使用习惯实现更高效的隐私保护方式。     

一种路网环境下LBS隐私保护算法

近年来,基于位置服务(LBS)的应用越来越广泛.用户在享受这种位置服务带来的方便快捷之同时,也要承担可能暴露自身隐私信息(如位置等)的风险.目前,很多工作已经在诸如保护用户隐私信息方面取得了重大进展,但大多是集中于欧氏空间下的隐私保护技术,而路网环境下的隐私保护研究相对较少.针对路网环境下用户边权分布不均的问题,提出了基于哑元的边权均衡算法,即在保证匿名集中各路段邻近性的同时,以生成哑元的方式均衡各边边权分布.这样既能最大程度降低查询代价,又能使边权分布不会太分散.最后,通过实验验证了本算法的有效性,同时显示该算法还能有效防止边权分布不均引发的推断攻击.     

FVSk-匿名:一种基于k-匿名的隐私保护方法

为了确保数据发布应用环节中个人敏感隐私数据信息的安全,深入研究了k-匿名技术的机制及性能,针对其不能完全有效地防止敏感属性数据信息泄漏的问题,通过引入真子树的概念和全新的敏感属性值选择手段,在实验探索的基础上,提出了一种基于k-匿名隐私保护模型的新的数据发布隐私保护方法——FVS k-匿名隐私保护方法。这种隐私保护方法继承了k-匿名技术实现简单、处理数据便捷的优点,而且弥补了其保护个人敏感隐私数据信息不完全、不充分的缺点。优化后的FVS k-匿名方法能有效地防止个人敏感隐私数据信息的泄漏,确保个人敏感隐私数据信息的安全。     

基于社交关系强度的兴趣社交APP设计研究

目的对兴趣社交APP这一社交APP的发展新方向进行研究分析,探讨兴趣社交APP的交互设计方法。方法对目前国内兴趣社交APP的发展以及特点进行总结归纳,并运用社交关系强度的理论,结合界面交互设计的现有研究成果,探讨如何将这些因素运用到兴趣社交APP的设计上。结论总结出兴趣社交APP的发展前景以及特点,针对其鲜明的特点,基于社交关系强度理论,探讨兴趣社交APP的设计策略,并以隔壁故事社交APP为例,在具体的设计中将理论进行实践。     

A fast variable neighborhood search for p-hub median problem using social network analysis concepts

This study will contribute to the uncapacitated single allocation p-hub median problem (USApHMP) which is known as an NP-hard problem in the literature. This problem is concerned with locating of hub facilities in a network and allocating of each non-hub node to just one hub in order to minimize total transportation costs in the network. A hybrid variable neighborhood search (VNS) algorithm is proposed considering three structures of local search which are used as a combination of nested VNS and sequential VNS in the algorithm. For reduction of the dimensions in the nested part, social network analysis centrality measures for the node are used to choose elite points instead of all existing points in the local search structures. The obtained results demonstrate that this will not only retain quality of the solutions, but also reduce run time of the algorithm significantly. Three standard data sets (AP, CAB, and URAND) were used for numerical analysis. Computational results show that quality of the obtained solutions is good and able to compete with other heuristics addressed in the literature. From the viewpoint of execution time, it considerably outperforms all other algorithms. The intelligent search embedded in this algorithm makes it robust and efficient on networks with up to 400 nodes.     

Using social network analysis to combat counterfeiting

Counterfeiting undoubtedly produces great harm to companies and the society. This article presents a study on the application of social network analysis (SNA) to combating this problem. The supply chain can be viewed as a network of parties involved in delivering value to consumers. Normally, SNA is used to analyse relationships between people. In this study, it is utilised to analyse supply chains for the identification of parties that are likely to be involved in counterfeiting activities. After identifying the suspects in a supply chain, the company can deter and detect counterfeiting by tightening surveillance on these parties. The feasibility of using SNA as an anti-counterfeiting tool is investigated in a case study, the findings of which indicate that SNA can effectively help companies to prevent sources of counterfeit products from infiltrating into the supply chain. Problematic parties can be identified by characterising the following features: high degree of centrality, high closeness of centrality and high betweenness of centrality.     

Exponential random graph modeling for diagnosis of out-of-control signals in social network surveillance: Signal interpretation in social networks with ERG modeling

Monitoring of the social networks for detecting anomalous behavior could be vital for the system's survival. This anomalous behavior could raise from any changes in behavior or attributes of a particular individual or groups of individuals in the network and causes structural changes. Multivariate statistical process control charts are effective tools for this purpose while Exponential Random Graph Models are used to model highly interdependent data of the network. So after selecting a model for specific network, T² control charts are used for monitoring the network data to detect any anomalous behavior. Then the Mason, Tracy, and Young method is utilized for interpreting an out-of-control condition. Finally, some real-world examples are used to evaluate the performance of the proposed diagnosis approach. Since complicated dependency in a social network makes different choices in model selection for Exponential Random Graph Models and this causes various results in the evaluation study, if the impact of diagnosis result is not seen in model selection, the appropriate model will not be necessarily selected and this will affect the effectiveness of the whole system. So, in this paper for improving the performance of diagnosis, two indices are introduced and added to model selection criteria and then the appropriate model could be selected based on the decision-maker's preferences.     

A survey on deep learning for patent analysis

Patent document collections are an immense source of knowledge for research and innovation communities worldwide. The rapid growth of the number of patent documents poses an enormous challenge for retrieving and analyzing information from this source in an effective manner. Based on deep learning methods for natural language processing, novel approaches have been developed in the field of patent analysis. The goal of these approaches is to reduce costs by automating tasks that previously only domain experts could solve. In this article, we provide a comprehensive survey of the application of deep learning for patent analysis. We summarize the state-of-the-art techniques and describe how they are applied to various tasks in the patent domain. In a detailed discussion, we categorize 40 papers based on the dataset, the representation, and the deep learning architecture that were used, as well as the patent analysis task that was targeted. With our survey, we aim to foster future research at the intersection of patent analysis and deep learning and we conclude by listing promising paths for future work.     

Influencing dynamics on social networks without knowledge of network microstructure

Social network-based information campaigns can be used for promoting beneficial health behaviours and mitigating polarization (e.g. regarding climate change or vaccines). Network-based intervention strategies typically rely on full knowledge of network structure. It is largely not possible or desirable to obtain population-level social network data due to availability and privacy issues. It is easier to obtain information about individuals’ attributes (e.g. age, income), which are jointly informative of an individual’s opinions and their social network position. We investigate strategies for influencing the system state in a statistical mechanics based model of opinion formation. Using synthetic and data-based examples we illustrate the advantages of implementing coarse-grained influence strategies on Ising models with modular structure in the presence of external fields. Our work provides a scalable methodology for influencing Ising systems on large graphs and the first exploration of the Ising influence problem in the presence of ambient (social) fields. By exploiting the observation that strong ambient fields can simplify control of networked dynamics, our findings open the possibility of efficiently computing and implementing public information campaigns using insights from social network theory without costly or invasive levels of data collection.     

The evolution of patent mining: Applying bibliometrics analysis and keyword network analysis

Text mining methods allow researchers to investigate technical documents (tech mining) and specifically explore patents for valuable information (patent mining. To the review literature and analyze the evolution of patent analysis and patent mining methods, bibliometrics analysis and keyword-based network analysis is applied on 143 papers extracted from the 'Web of science' database. Bibliometrics analysis was applied to determine top players researching in patent mining. Applying cluster analysis on the keyword network shows three main stages of patent analysis evolution. Also, it is discussed how patent mining is evolutionized in terms of information retrieval, pattern recognition and pattern analysis.     

Inferring social network structure in ecological systems from spatio-temporal data streams

We propose a methodology for extracting social network structure from spatio-temporal datasets that describe timestamped occurrences of individuals. Our approach identifies temporal regions of dense agent activity and links are drawn between individuals based on their co-occurrences across these ‘gathering events’. The statistical significance of these connections is then tested against an appropriate null model. Such a framework allows us to exploit the wealth of analytical and computational tools of network analysis in settings where the underlying connectivity pattern between interacting agents (commonly termed the adjacency matrix) is not given a priori. We perform experiments on two large-scale datasets (greater than 10⁶ points) of great tit Parus major wild bird foraging records and illustrate the use of this approach by examining the temporal dynamics of pairing behaviour, a process that was previously very hard to observe. We show that established pair bonds are maintained continuously, whereas new pair bonds form at variable times before breeding, but are characterized by a rapid development of network proximity. The method proposed here is general, and can be applied to any system with information about the temporal co-occurrence of interacting agents.     

A knowledge accumulation approach based on bilayer social wiki network for computer-aided process innovation

Under the fierce competition, manufacturing companies pay more attention to innovation and the knowledge that enables innovation. Manufacturing process innovation is a knowledge-intensive activity, and efficient knowledge accumulation is the prerequisite and basis for computer-aided process innovation (CAPI). Hence, this research aims to build an open knowledge accumulation approach to obtain organised and refined process innovation knowledge (PIK). By considering the similarity of PIK network with biological neural network and combining the technical characteristics of social network with wiki, a novel PIK accumulation schema based on bilayer social wiki network is proposed. In social wiki network environment, PIK is accumulated in public knowledge space through participants’ social interactions and knowledge activities. The process of knowledge fusion is investigated to form the preliminary knowledge containing collective intelligence, and the mechanisms of collaborative editing and collaborative evolution are studied to refine the knowledge. The outcomes of this study lay the foundation for knowledge application of CAPI. Finally, a case study is presented to demonstrate the applicability of the proposed approach.     

Visualizing patent statistics by means of social network analysis tools

The present paper reviews the literature on social network analysis with applications to bibliometric data, and in particular, patent information. Several approaches of network analysis are conducted in the field of optoelectronics to exemplify the power of network analysis tools. Cooperation networks between inventors and applicants are illustrated, emphasizing bibliometric measures such as activity, citation frequency, etc. as well as network theoretical measures, e.g. centrality or betweenness. In this context it is found that inventors who serve as interfaces or links between different inventor groups apply for technologically broader patents, hence, benefiting from their access to different knowledge through their position. Furthermore, citation networks of patent documents as well as patent applicants were drawn. Here, patent thickets could be identified. The position of applicants within citation networks seems to be useful in explaining behaviour of the applicants in the marketplace, such as cooperation or patent infringement trials.     

Reviewing the intellectual structure and evolution of the innovation systems approach: A social network analysis

The increased attention of policymakers and researchers to the concept of innovation systems in recent decades has led to an increase in studies in this field and, consequently, its dynamics. This increase in the number of studies indicates the necessity of conducting studies to delineate the intellectual structure and process of development and evolution of this field. The present study aims to present a picture of the structure, evolution, and dynamics of innovation systems through the analysis of academic social networks based on bibliometric criteria and identify emerging issues for research. Accordingly, using citation, word co-occurrence, co-citation, and bibliographic coupling analyses, we analyzed 3250 documents which had been published between 1988 and 2018 and extracted from the Web of Science database. Through a systematic review, theoretical roots and frameworks affecting this field were identified and the existing research streams were introduced. According to the analyses, historical and contemporary views were categorized into four clusters. In addition, the most important and emerging issues in this field were identified and the most influential documents, journals, and authors were introduced.