Large-scale Social Network Data Research Articles

A Hybrid Probabilistic Graph Based Community Clustering Model for Large Social Networking Link Prediction Data

Dynamic community clustering is essential for online social networking sites due to the high dimensionality and large data size. It aims to uncover social relationships among nodes and links within the network. However, traditional models often struggle with community structure detection because of the extensive computational time and memory required. Additionally, these models need contextual weighted node information to establish social networking feature relationships. To address these challenges, an advanced probabilistic weighted community detection framework has been developed for large-scale social network data. This framework uses a filter-based probabilistic model to eliminate sparse values and identify weighted community detection nodes for dynamic clustering analysis. Experimental results demonstrate that this filter-based probabilistic community detection framework outperforms others in terms of normalized mutual information, entropy, density, and runtime efficiency (measured in milliseconds).

Learning Distilled Graph for Large-Scale Social Network Data Clustering

Spectral analysis is critical in social network analysis. As a vital step of the spectral analysis, the graph construction in many existing works utilizes content data only. Unfortunately, the content data often consists of noisy, sparse, and redundant features, which makes the resulting graph unstable and unreliable. In practice, besides the content data, social network data also contain link information, which provides additional information for graph construction. Some of previous works utilize the link data. However, the link data is often incomplete, which makes the resulting graph incomplete. To address these issues, we propose a novel Distilled Graph Clustering (DGC) method. It pursuits a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">distilled graph</i> based on both the content data and the link data. The proposed algorithm alternates between two steps: in the feature selection step, it finds the most representative feature subset w.r.t. an intermediate graph initialized with link data; in graph distillation step, the proposed method updates and refines the graph based on only the selected features. The final resulting graph, which is referred to as the distilled graph, is then utilized for spectral clustering on the large-scale social network data. Extensive experiments demonstrate the superiority of the proposed method.

Distributed privacy preserving technology in dynamic networks

With the development of information technology, large-scale social network graph data have been produced and released to provide data analysis for scientific research and business structures, while traditional network privacy protection technology does not meet the actual requirements. In this paper, we address the privacy risks of link disclosure in sequential release of a dynamic network. To prevent privacy breaches, we proposed the privacy model km - number of mutual friend, where k indicates the privacy level and m is a time period that an adversary can monitor a victim to collect the attack knowledge. We present a distributed algorithm to generate releases by adding nodes in parallel. Further, in order to improve availability of anonymous graphs, distributed greedy merge noise node algorithm (DGMNNA) is designed to reduce the number of nodes added under the premise of satisfying the anonymous model. The experimental results show that the proposed algorithm can efficiently handle large-scale social network data while ensuring the availability of anonymous data.

Real Time Social Network Data Analysis for Community Detection

WWW becomes a widely used platform for different social networks and social medias for the social communication. This platform becomes the oasis of a huge amount of data. Therefore, this data repository draws tremendous attention from corporate, government, NGOs, social workers, politician, etc. to either promote their products or to convey their message to the targeted community. But identification of community structure and social graph becomes a challenging issue for the social network researcher and graph theory researchers since the pervasive usage of instant messaging systems and fundamental shift in publishing contents in these social medias. Although a lot of attention has been given by the researcher to introduce several algorithms for identifying the community structure, most of them are not suitable for dealing with the large scale social network data in real time. This paper presents a model for community detection from social graph using the real time data analytic. In this paper, we introduce data analytic algorithms that can analysis contextual data. These algorithms can analyze large scale social interaction data and can detect a community based on the user supplied threshold value for community detection. Experiment result shows that the proposed algorithms can identify expected number meaningful communities from the social graph.

Topic Modeling in Large Scale Social Network Data

The growing popularity of social media such as Twitter and Facebook has made these websites an important source of information. The large amount of data available on these platforms presents new opportunities for mining information about the real world. Because of its widespread usage, a lot of useful information can be extracted from the text available on these social media platforms. It can be used to infer important aspects about the users of these services and about the things happening in their surroundings. This work proposes generative probabalistic models to identify latent topics and sentiments in social media data, mainly Twitter. In contrast to the majority of earlier work done in the field of topic modeling in social media data, this work incorporates various special characteristics of this data- mainly the short-length nature and special tokens like hashtags. The models proposed in work were compared qualitatively and quantitatively against several baseline models for evaluation. Experimental results suggest several improvements over the existing baseline techniques.

MapReduce-based efficient betweenness approximation pivot method for large graphs

The important person is vital for information spreading efficiency in social networks, which has always been the hottest research field in recent years. Nowadays, with the increased information spreading speed in our daily life such as cellphone or microblog, the government and enterprise have realised that it is necessary and important to manage and control the information spreading in social networks of people by recognising the important person or VIP customer. Meanwhile, with the rapid increase of information in society activities recently, the distributed information system engineering such as MapReduce-based computation system has become more and more popular in large-scale social network data analysis in emergency case management or VIP customer recognition. In this article, a novel effective and efficient node betweenness approximation pivot selection method for largescale graphs is proposed based on MapReduce-based system for social network analysis SNA. Furthermore, it has been proved useful in four discrete datasets and one continuous dataset six month cellphone call graph in VIP customer discovery for large graphs from real telecom data in China to help deal with customer relationship management more efficiently.

DynamoGraph: extending the Pregel paradigm for large-scale temporal graph processing

Large-scale temporal graphs can serve as a model in many application scenarios. Owing to the popularity of online social networks and increased research interest in gathering and analysing data about human behaviour and interaction reality mining temporal graphs gain attraction in social network analysis and more specifically in the analysis of dynamic processes in social networks. Originally developed for web-graph analysis, systems supporting a Pregel-style processing paradigm can be used to process large-scale social network data. However, current methods for social network analysis require data to be processed off-line, or lack support for temporal graphs, or support datasets of limited size only. In this work we present a cloud-based distributed processing framework extending the Pregel paradigm for large-scale temporal graphs. By using computing resources in the cloud this system is scalable and prepared to scale for the massive datasets that occur in social network analysis.

Graph sampling approach for reducing computational complexity of large-scale social network

Online social network services provide a platform for human social interactions. Nowadays, many kinds of online interactions generate large-scale social network data. Network analysis helps to mine knowledge and pattern from the relationship between actors inside the network. This approach is important to support predictions and the decision-making process in many real-world applications. The social network analysis methodology, which borrows approaches from graph theory provides several metrics that enabled us to measure specific properties of the networks. Some of the metrics calculations were built with no scalability in minds, thus it is computationally expensive. In this paper, we propose a graph sampling approach to reduce social network size, thus reducing computation operations. The performance comparison between natural graph sampling strategies using edge random sampling, node random sampling, and random walks are presented on each selected graph property. We found that the performance of graph sampling strategies depends on graph properties measured.

Predicting Positive and Negative Relationships in Large Social Networks.

In a social network, users hold and express positive and negative attitudes (e.g. support/opposition) towards other users. Those attitudes exhibit some kind of binary relationships among the users, which play an important role in social network analysis. However, some of those binary relationships are likely to be latent as the scale of social network increases. The essence of predicting latent binary relationships have recently began to draw researchers' attention. In this paper, we propose a machine learning algorithm for predicting positive and negative relationships in social networks inspired by structural balance theory and social status theory. More specifically, we show that when two users in the network have fewer common neighbors, the prediction accuracy of the relationship between them deteriorates. Accordingly, in the training phase, we propose a segment-based training framework to divide the training data into two subsets according to the number of common neighbors between users, and build a prediction model for each subset based on support vector machine (SVM). Moreover, to deal with large-scale social network data, we employ a sampling strategy that selects small amount of training data while maintaining high accuracy of prediction. We compare our algorithm with traditional algorithms and adaptive boosting of them. Experimental results of typical data sets show that our algorithm can deal with large social networks and consistently outperforms other methods.

Critical transitions in social network activity

A large variety of complex systems in ecology, climate science, biomedicine and engineering have been observed to exhibit tipping points, where the internal dynamical state of the system abruptly changes. For example, such critical transitions may result in the sudden change of ecological environments and climate conditions. Data and models suggest that detectable warning signs may precede some of these drastic events. This view is also corroborated by abstract mathematical theory for generic bifurcations in stochastic multi-scale systems. Whether the stochastic scaling laws used as warning signs are also present in social networks that anticipate a-priori {\it unknown} events in society is an exciting open problem, to which at present only highly speculative answers can be given. Here, we instead provide a first step towards tackling this formidable question by focusing on a-priori {\it known} events and analyzing a social network data set with a focus on classical variance and autocorrelation warning signs. Our results thus pertain to one absolutely fundamental question: Can the stochastic warning signs known from other areas also be detected in large-scale social network data? We answer this question affirmatively as we find that several a-priori known events are preceded by variance and autocorrelation growth. Our findings thus clearly establish the necessary starting point to further investigate the relation between abstract mathematical theory and various classes of critical transitions in social networks.

MapReduce based Betweenness Approximation Engineering in Large Scale Graph

With increased emergency cases in the world, the government has realized that it is necessary to manage and control the information spreading in social network. Recently with the rapid increase of information in society activities, distributed information system engineering such as MapReduce based cloud computing distributed system is becoming more and more important in large scale social network data computing in emergency case management. In this article, a novel MapReduce based SNA (Social Network Analysis) system was proposed on layered distributed system. According to independent identically distributed probability statistics of node Betweenness Centrality, a more efficient betweenness approximation computation method was proposed in this article in very large scale SNA graph from emergency case management with better accuracy.

Regression Analysis of Asymmetric Pairs in Large-Scale Network Data

In social network analysis, many estimation methods have been developed over the past three decades. Due to the computational complexity for analyzing large-scale social network data, however, those methods cannot be applied effectively. On the other hand, the structure of large-scale network data is often sparse so that the information loss by ignoring symmetric pairs is rather limited. Hence, we propose an asymmetric pairs regression (APR) approach to study the social network relationship. Accordingly, the computation of parameter estimations is simple and the theoretical properties can be obtained via the established logistic regression model. Simulation studies and an empirical example are presented to illustrate the usefulness of APR.

Regression Analysis of Asymmetric Pairs in Large-Scale Network Data

In social network analysis, many estimation methods have been developed over the past three decades. Due to the computational complexity for analyzing large-scale social network data, however, those methods cannot be applied effectively. On the other hand, the structure of large-scale network data is often sparse so that the information loss by ignoring symmetric pairs is rather limited. Hence, we propose an asymmetric pairs regression (APR) approach to study the social network relationship. Accordingly, the computation of parameter estimations is simple and the theoretical properties can be obtained via the established logistic regression model. Simulation studies and an empirical example are presented to illustrate the usefulness of APR.

Sequential Influence Models in Social Networks

The spread of influence among individuals in a social network can be naturally modeled in a probabilistic framework, but it is challenging to reason about differences between various models as well as to relate these models to actual social network data. Here we consider two of the most fundamental definitions of influence, one based on a small set of "snapshot'' observations of a social network and the other based on detailed temporal dynamics. The former is particularly useful because large-scale social network data sets are often available only in snapshots or crawls. The latter however provides a more detailed process model of how influence spreads. We study the relationship between these two ways of measuring influence, in particular establishing how to infer the more detailed temporal measure from the more readily observable snapshot measure. We validate our analysis using the history of social interactions on Wikipedia; the result is the first large-scale study to exhibit a direct relationship between snapshot and temporal models of social influence.