Scheduled System Maintenance:
On May 6th, single article purchases and IEEE account management will be unavailable from 8:00 AM - 5:00 PM ET (12:00 - 21:00 UTC). We apologize for the inconvenience.
By Topic

Computational Aspects of Social Networks (CASoN), 2012 Fourth International Conference on

Date 21-23 Nov. 2012

Filter Results

Displaying Results 1 - 25 of 67
  • [Front matter]

    Publication Year: 2012 , Page(s): i - xix
    Save to Project icon | Request Permissions | PDF file iconPDF (193 KB)  
    Freely Available from IEEE
  • A discussion on Internet Governance

    Publication Year: 2012 , Page(s): 1 - 7
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (279 KB) |  | HTML iconHTML  

    The Internet has created a universal medium wherein peoples of the world engage in dialogue and participate in a myriad of activities, despite the medium lacking universally enforceable rules of conduct. This absence, or perceived weakness, of Governance in and of the Internet, has aided the creation of a sphere of existence wherein issues such as censorship, violation of the end-to-end principle, Intellectual Rights protection, cannot be debated adequately due to the lack of a suitable framework, nor can decisions by a stakeholder be enforced universally. Drawing on the literature in the field of Information Systems dealing with Governance of the Internet and Philosophy of Ontology, we question the implicit assumption that the Internet can be governed. We also investigate whether the Internet within the field of Information Sciences is ontologically defined. In stating our arguments, we utilize philosophical hyperbolic doubt as a guiding methodology. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Cross-domain vulnerabilities over social networks

    Publication Year: 2012 , Page(s): 8 - 13
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (860 KB) |  | HTML iconHTML  

    Recent years have seen a tremendous growth of social networks such as Facebook and Twitter. At the same time, the share of video traffic in the Internet has also significantly increased, and the two functions are getting closer to one another. YouTube, the most famous video sharing site, allows people to comment on videos with other people while Facebook and Twitter are important vectors into sharing videos. Both video channels and social networks are increasingly vulnerable attack targets. For example, social networks are also considerable spam and phishing vectors, and Adobe Flash as the premier video streaming application is associated with numerous software vulnerabilities. This is a good way for attackers to compromise sites with embedded Flash objects. In this paper, we present the technical background of the cross-domain mechanisms and the security implications. Several recent studies have demonstrated the weakness of the cross-domain policy, leading to session hijacking or the leakage of sensitive information. Current solutions to detect these vulnerabilities use a client-side approach. The purpose of our work is to present a new approach based on network flows analysis to detect malicious behavior. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Informativity-based graph: Exploring mutual kNN and labeled vertices for semi-supervised learning

    Publication Year: 2012 , Page(s): 14 - 19
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (628 KB) |  | HTML iconHTML  

    Data repositories are getting larger and in most of the cases, only a small subset of their data items is labeled. In such scenario semi-supervised learning (SSL) techniques have become very relevant. Among these algorithms, those based on graphs have gained prominence in the area. An important step in graph-based SSL methods is the conversion of tabular data into a weighted graph. However, most of the SSL literature focuses on developing label inference algorithms without studying graph construction methods and its effect on the base algorithm performance. This paper provides a novel technique for building graph by using mutual kNN and labeled vertices. The use of prior information, i.e., to consider the small fraction of labeled vertices, has been underexplored in SSL literature and mutual kNN has been only explored in clustering. The empirical evaluation of the proposed graph showed promising results in terms of accuracy, when it is applied to the label propagation task. Additionally, the resultant networks have lower average degree than kNN networks. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Primary role identification in e-mail networks using pattern subgraphs and sequence diagrams

    Publication Year: 2012 , Page(s): 20 - 25
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (352 KB) |  | HTML iconHTML  

    Social networks often forms very complex structures that additionally change over time. Description of actors' roles in such structures requires to take into account this dynamics reflecting behavioral characteristics of the actors. A role can be defined as a sequence of different types of activities. Various types of activities are modeled by pattern subgraphs, whereas sequences of these activities are modeled by sequence diagrams. For such defined roles, an original role identification procedure is applied assigning to each e-mail user a predefined role of spammer or regular user. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Event detection in evolving networks

    Publication Year: 2012 , Page(s): 26 - 32
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (467 KB) |  | HTML iconHTML  

    This paper describes a methodology for finding and describing significant events in time evolving complex networks. We first group the nodes of the network in clusters, according to their similarity in terms of a set of local properties such as degree and clustering coefficient. We then monitor the behavior of these groups over time, looking for significant changes on the size of the groups. These events are notable since they show that the position of a number of nodes in the network has changed. We describe this evolution by extracting the correspondent transition patterns. We examined our methodology on three different real network datasets. Our experiments show that the discovered rules are significant and can describe the occurring events. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Estimating network parameters using random walks

    Publication Year: 2012 , Page(s): 33 - 40
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (575 KB) |  | HTML iconHTML  

    Sampling from large graphs is an area of great interest, especially since the emergence of huge structures such as Online Social Networks (OSNs) and the World Wide Web (WWW). These networks, when viewed as graphs, often contain hundreds of millions of vertices and billions of edges. The large scale properties of a network can be summarized in terms of parameters of the underlying graph, such as the total number of vertices, edges and triangles. The large size of these networks makes it computationally expensive to obtain such structural properties of the underlying graph by exhaustive search. If we can estimate these properties by taking small but representative samples from the network, then size is no longer such a problem. In this paper we present a general framework to estimate network properties using random walks. These methods work under the assumption we are able to obtain local characteristics of a vertex during each step of the random walk, for example the number and labels of the neighboring vertices of a specific vertex These assumptions are relatively reasonable in practice, but may add some additional query cost to each step of the random walk. We also present some practical methods to estimate the total number of edges/links m, number of vertices/nodes n and number of connected triads of vertices (triangles) t in graphs with degree distributions which follow a power-law and higher number of triangles higher than expected in random graphs. We use these graphs since they tend to better correspond to the structure of large online networks, and in fact some of the data used are taken from such a network. Additionally we present experimental estimates for n, m, t we obtained using our methods on real or manufactured networks. In order to make the methods practical, the total number of steps made by the walk was limited to at most the size n of the network. In fact the results appear to converge for a lower number of steps, indicating that our proposed met- ods are feasible in practice. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Graphical analysis of social group dynamics

    Publication Year: 2012 , Page(s): 41 - 46
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (452 KB) |  | HTML iconHTML  

    Identifying communities in social networks becomes an increasingly important research problem. Several methods for identifying such groups have been developed, however, qualitative analysis (taking into account the scale of the problem) still poses serious problems. This paper describes a tool for facilitating such an analysis, allowing to visualize the dynamics and supporting localization of different events (such as creation or merging of groups). In the final part of the paper, the experimental results performed using the benchmark data (Enron emails) provide an insight into usefulness of the proposed tool. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Logical push framework for real-time SNS processing

    Publication Year: 2012 , Page(s): 47 - 51
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (467 KB) |  | HTML iconHTML  

    The development of mobile devices and Social Network Service (SNS) has explosively increased unstructured data. The purpose of this study is to suggest the abstract push framework in the Social Big Data environment especially for real-time SNS. Through the analysis of previous push services, the study found limitations such as fixed payload size, duplicated development of each mobile OS, and messages lost in the air interface. The proposed framework added new functions that serve as the temporary storage in the push façade and the local storage in the client device. It is to solve the message lost problems and enables to use Offline SNS through pre-caching the contents operation of contents. The stakeholders of the push service - Service Providers (SP), Contents Providers (CP) - can refer to the framework and improve their services. The limitation of this study is that only the abstract version of the framework is suggested. Detailed simulation results and its performance compared to other frameworks will be conducted after development of the prototype in the future. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Automatic sentiment analysis of Twitter messages

    Publication Year: 2012 , Page(s): 52 - 57
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (322 KB) |  | HTML iconHTML  

    Twitter® is a microblogging service usually used as an instant communication platform. The capacity to provide information in real time has stimulated many companies to use this service to understand their consumers. In this direction, TV stations have adopted Twitter for shortening the distance between them and their viewers, and use such information as a feedback mechanism for their shows. The sentiment analysis task can be used as one such feedback mechanism. This task corresponds to classifying a text according to the sentiment that the writer intended to transmit. A classifier usually requires a pre-classifled data sample to determine the class of new data. Typically, the sample is pre-classified manually, making the process time consuming and reducing its real time applicability for big data. This paper proposes an automatic sentiment classifier for Twitter messages, and uses TV shows from Brazilian stations for benchmarking. The automatic sentiment analysis reduces human intervention and, thus, the complexity and cost of the whole process. To assess the performance of the proposed system tweets related to a Brazilian TV show were captured in a 24h interval and fed into the system. The proposed technique achieved an average accuracy of 90%. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Machine learning algorithms applied in automatic classification of social network users

    Publication Year: 2012 , Page(s): 58 - 62
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (391 KB) |  | HTML iconHTML  

    This work shows the results of an analysis of machine learning algorithms applied in automatic classification for the users of the social network called Scientia.Net. The tests were done using a database with 2000 users. The analysis identifies which algorithm performs better in automatic classification of users within a social network. The algorithms tested were Multilayer Perceptron, Support Vector Machine, Kohonen Network and K-means Algorithm. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • What does everybody know? Identifying event-specific sources from social media

    Publication Year: 2012 , Page(s): 63 - 68
    Cited by:  Papers (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (531 KB) |  | HTML iconHTML  

    Social media is increasingly becoming a popular platform for the public to voice their opinion and present them to a huge audience in the web. The year 2011 saw one of the greatest use of social media in the rise and spread of various events, and has been rightly defined as the year of “Social Media Democracy”, with “The Protester” being named as the TIME magazine's person of the year 2011. Due to the power law distribution of the Internet, it is highly likely that the social media sites are buried in the Long Tail. It is therefore, of utmost importance to identify quality social media sources from the Long Tail, for understanding and exploring the real-life events in depth. In this work, we propose a framework to distinguish the disparate sources from social media that provide extremely significant information about various events. Specifically, we propose information theoretic measures to identify “specific” sources for an event (often buried in the Long Tail) and “closer” entities (individuals, groups, organizations, places, etc.) for an event. We also introduce a novel evaluation strategy for validating the proposed measures. Data for the research is collected from various blogging platforms. Experiments demonstrate promising results with interesting findings. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Quantifying social network dynamics

    Publication Year: 2012 , Page(s): 69 - 74
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (523 KB) |  | HTML iconHTML  

    The dynamic character of most social networks requires to model evolution of networks in order to enable complex analysis of theirs dynamics. The following paper focuses on the definition of differences between network snapshots by means of Graph Differential Tuple. These differences enable to calculate the diverse distance measures as well as to investigate the speed of changes. Four separate measures are suggested in the paper with experimental study on real social network data. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Graph-based cross-validated committees ensembles

    Publication Year: 2012 , Page(s): 75 - 80
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (437 KB) |  | HTML iconHTML  

    Ensemble techniques combine several individual classifiers to obtain a composite classifier that outperforms each of them alone. Despite of these techniques have been successfully applied to many domains, their applications on networked data still need investigation. There are not many known strategies for sampling with replacement from interconnected relational data. To contribute in this direction, we propose a cross-validated committee ensemble procedure applied to graph-based classifiers. The proposed ensemble either maintains or significantly improves the accuracy of the tested relational graph-based classifiers. We also investigate the role played by diversity among the several individual classifiers, i.e., how much they agree in their predictions, to explain the technique success or failure. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Is ARPU1; the right choice for wireless data-based communication services?

    Publication Year: 2012 , Page(s): 81 - 86
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (770 KB) |  | HTML iconHTML  

    Communication services based on wireless data, such as Social Network Service (SNS) and Mobile Instant Messenger (MIM), developed rapidly with the advent of communication networks and smartphones. As SNS and MIM gradually replace voice and Short Message Service (SMS) communication, the revenue structures of mobile carriers also change into data. Although mobile carriers are still measuring their revenues using voice-based indices such as Average Revenue Per User (ARPU). The use of accurate indices and measurement is important, since revenue measurement determines the future developmental direction of businesses and industries. Therefore, this study analyzes changes in revenue sources in this age of data, the utility of ARPU, and the standards for data-based revenue calculations. The analysis shows that the utility of ARPU is very low, since it cannot reflect the growth of data services and the existence of multiple subscribers. Also, it was identified that services and subscribers should be redefined in consideration of data revenues. The results of the study propose measurement factors considering data-based revenue, which could be referred as standards for measuring diverse revenues. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Representing scientific associations through the lens of Actor-Network Theory

    Publication Year: 2012 , Page(s): 87 - 92
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (468 KB) |  | HTML iconHTML  

    Knowledge on the social web presupposes to gather information about its current and potential users and document their relationships, interests and needs. A recent branch of sociology, the Actor-Network Theory or ANT, states that relations among human and nonhuman actors are equally important to comprehend social phenomena. Since scientists are potential users of huge computational support, their communities provide relevant cases for domain characterization and software design. This paper investigates the possibilities of using ANT to characterize a real instance of those social networks. The active role of nonhuman actors allows us to trace the relations based on material clues left behind by the actors, and also to bring forth features to be explored by social software. The results of this study present a graphical representation that allows quantitative and qualitative analysis of the social network, which may inform a better design of systems for those communities. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Detection of head experts in social network

    Publication Year: 2012 , Page(s): 93 - 98
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (248 KB) |  | HTML iconHTML  

    This paper introduces a method for head expert identification in a social network based on local community detection and formal concept analysis. There are several methods for expert identification, but most of these methods try to find an expert for a particular area. In this paper, we propose a novel approach to identify a head expert. This person is in the background and most of the time he is working across different areas of research, and he can establish teams of experts for particular areas. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Performance of Modified Iterative Decoding Algorithm for Multilevel codes

    Publication Year: 2012 , Page(s): 99 - 104
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (611 KB) |  | HTML iconHTML  

    In this paper, performance of Modified Iterative Decoding Algorithm (MIDA) is investigated for decoding of Multi-level codes. MIDA is a hard decision decoder that was initially proposed for decoding of Product codes, by same authors. In this paper same concept is being extended for Multi-level codes. The basic iterative decoder has a huge complexity which is unaffordable for long constituent codes especially when used in real time systems. In MIDA, the idea of Syndrome Decoding is used for search space reduction. Moreover, it is found that complexity reduction factor increases with each iteration. The worst case complexity of MIDA is closed to that of basic algorithm but in average case there is huge reduction in complexity, with negligible performance degradation. Performance of the proposed algorithm is investigated over an OFDM system. Additive White Gaussian Noise (AWGN) channel and Rayleigh fading channel are used for simulations. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Identifying focal patterns in social networks

    Publication Year: 2012 , Page(s): 105 - 108
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (432 KB) |  | HTML iconHTML  

    Identifying authoritative individuals is a well-known approach in extracting actionable knowledge, known as “Knowledge Representation”, in a social network. Previous researches suggest measures to identify influential individuals, however, such individuals might not represent the appropriate context (relationships, interactions, etc.). For example, it is nearly an impossible task for a single individual to organize a mass protest of the scale of Occupy Wall Street. Similarly, other events such as the Arab Spring, coordinating crisis responses for natural disasters (e.g., the Haiti earthquake), or even organizing flash mobs would require a key set of individuals rather than a single or the most authoritative one. These events demonstrate the need and importance of examining influential structures rather than single individuals in social networks. A new methodology is proposed to identify such influential structures and recognizing their importance. The proposed methodology is evaluated empirically with real-world data from NIST's Tweets2011 corpus. We also introduce a novel and objective evaluation strategy to ascertain the efficacy of the focal patterns. Challenges with future research directions are outlined. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • RunMyCode: An innovative platform for social production and evaluation of scientific research

    Publication Year: 2012 , Page(s): 109 - 114
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (359 KB) |  | HTML iconHTML  

    In this paper we describe RunMyCode (RMC), a cloud based Platform as a Service tool (PaaS) which involves researchers collaborating across different time and geographical zones to produce research results along with their related papers. It helps researchers to incrementally co-create, deploy, evaluate, optimize and reproduce their research. It allows also a continuous integration and enhancement of different types of research artefacts including visual models, software and scripts. RMC extends traditional collaborative platforms by means of a scientific research-oriented social layer which allows intensive user contribution, interactions, and community building along with a content management system that lets researchers share, evaluate and develop valuable knowledge. RMC features include load balancing, elastic context aware and data-driven resource allocation, and parallel and “parameterizable” task execution. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Solving maximum clique problem in stochastic graphs using learning automata

    Publication Year: 2012 , Page(s): 115 - 119
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (299 KB) |  | HTML iconHTML  

    The maximum clique of a given graph G is the subgraph C of G such that two vertices in C are adjacent in G with maximum cardinality. Finding the maximum clique in an arbitrary graph is an NP-Hard problem, motivated by the social networks analysis. In the real world applications, the nature of interaction between nodes is stochastic and the probability distribution function of the vertex weight is unknown. In this paper a learning automata-based algorithm is proposed for solving maximum clique problem in the stochastic graph. The simulation results on stochastic graph demonstrate that the proposed algorithm outperforms standard sampling method in terms of the number of samplings taken by algorithm. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • DIFSoN: Discovering influential friends from social networks

    Publication Year: 2012 , Page(s): 120 - 125
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (304 KB) |  | HTML iconHTML  

    Social networks, which are made of social entities (e.g., individual users) linked by some specific types of interdependencies such as friendship, have become popular to facilitate collaboration and knowledge sharing among users. Such interactions or interdependencies can be dependent on or influenced by user characteristics such as connectivity, centrality, weight, importance, and activity in the networks. As such, some users in the social networks can be considered as highly influential to others. In this paper, we propose a computational model that integrates data mining with social computing to help users to discover influential friends from the social networks. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An algorithm to achieve k-anonymity and l-diversity anonymisation in social networks

    Publication Year: 2012 , Page(s): 126 - 131
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (271 KB) |  | HTML iconHTML  

    The development of several popular social networks in recent days and publication of social network data has led to the danger of disclosure of sensitive information of individuals. This necessitated the preservation of privacy before the publication of such data. There are several algorithms developed to preserve privacy in micro data. But these algorithms cannot be applied directly as in social networks the nodes have structural properties along with their labels. k-anonymity and l-diversity are efficient tools to anonymise micro data. So efforts have been made to find out similar algorithms to handle social network anonymisation. In this paper we propose an algorithm which can be used to achieve k-anonymity and l-diversity in social network anonymisation. This algorithm is based upon some existing algorithms developed in this direction. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Structural link prediction using community information on Twitter

    Publication Year: 2012 , Page(s): 132 - 137
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (262 KB) |  | HTML iconHTML  

    Currently, social networks and social media have attracted increasing research interest. In this context, link prediction is one of the most important tasks since it can predict the existence or missing of a future relation between user members in a social network. In this paper, we describe experiments to analyze the viability of applying the within and inter cluster (WIC) measure for predicting the existence of a future link on a large-scale online social network. Compared with undirected social networks, directed social networks have received less attention and still are not well understood, mainly due to the occurrence of asymmetric links. The WIC measure combines the local structural similarity information and community information to improve link prediction accuracy. We compare the WIC measure with classical measures based on local structural similarities, using real data from Twitter, a directed and asymmetric large-scale online social network. Our experiments show that the WIC measure can be used efficiently on directed and asymmetric large-scale networks. Moreover, it outperforms all compared measures employed for link prediction. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Tweets mining for French Presidential Election

    Publication Year: 2012 , Page(s): 138 - 143
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (702 KB) |  | HTML iconHTML  

    This paper describes a system which surveys the French Presidential Election trends from Twitter's discussions. This system carries out the automatic collection, evaluation and rating of tweets for evaluating the trends. The objective of this paper is to discuss the variety of issues and challenges surrounding the perspectives regarding the use of Social Network Analyses and Text Mining methods for applications in politics. The article will first present a review of the literature in Social Media and Text Mining analysis for political purposes and then describe the work done for our system for collecting and manipulating Twitter's data. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.