User Alignment Across Social Networks Based On ego-Network Embedding

Users' addiction to online social networks is discovered to be highly correlated with their social connections in the networks. Dense social connections can effectively help online social networks retain their active users and improve the social network services. Therefore, it is of great importance to make a good prediction of the social links among users. Meanwhile, to enjoy more social network services, users nowadays are usually involved in multiple online social networks simultaneously. Formally, the social networks which share a number of common users are defined as the "aligned networks". With the information transferred from multiple aligned social networks, we can gain a more comprehensive knowledge about the social preferences of users in the pre-specified target network, which will benefit the social link prediction task greatly. However, when transferring the knowledge from other aligned source networks to the target network, there usually exists a shift in information distribution between different networks, namely domain difference. In this paper, we study the social link prediction problem of the target network, which is aligned with multiple social networks concurrently. To accommodate the domain difference issue, we project the features extracted for links from different aligned networks into a shared lower-dimensional feature space. Moreover, users in social networks usually tend to form communities and would only connect to a small number of users. Thus, the target network structure has both the low-rank and sparse properties. We propose a novel optimization framework, SLAMPRED, to combine both these two properties aforementioned of the target network and the information of multiple aligned networks with nice domain adaptations. Since the objective function is a linear combination of convex and concave functions involving nondifferentiable regularizers, we propose a novel optimization method to iteratively solve it. Extensive experiments have been done on real-world aligned social networks, and the experimental results demonstrate the effectiveness of the proposed model.

Social networks have been part of people's daily life and plenty of users have registered accounts in multiple social networks. Interconnections among multiple social networks add a multiplier effect to social applications when fully used. With the sharp expansion of network size, traditional standalone algorithms can no longer support computing on large scale networks while alternatively, distributed and parallel computing become a solution to utilize the data-intensive information hidden in multiple social networks. As such, synergistic partitioning, which takes the relationships among different networks into consideration and focuses on partitioning the same nodes of different networks into same partitions. With that, the partitions containing the same nodes can be assigned to the same server to improve the data locality and reduce communication overhead among servers, which are very important for distributed applications. To date, there have been limited studies on multiple large scale network partitioning due to three major challenges: 1) the need to consider relationships across multiple networks given the existence of intricate interactions, 2) the difficulty for standalone programs to utilize traditional partitioning methods, 3) the fact that to generate balanced partitions is NP-complete. In this paper, we propose a novel framework to partition multiple social networks synergistically. In particular, we apply a distributed multilevel k-way partitioning method to divide the first network into k partitions. Based on the given anchor nodes which exist in all the social networks and the partition results of the first network, using MapReduce, we then develop a modified distributed multilevel partitioning method to divide other networks. Extensive experiments on two real data sets demonstrate that our method can significantly outperform baseline independent-partitioning method in accuracy and scalability.

In recent years, social networks have shown diversity in function and applications. People begin to use multiple online social networks simultaneously for different demands. The ability to uncover a user’s latent topic and social network preference is critical for community detection, recommendation, and personalized service across social networks. Unfortunately, most current works focus on the single network, necessitating new technology and models to address this issue. This paper proposes a user preference discovery model on multiple social networks. Firstly, the global and local topic concepts are defined, then a latent semantic topic discovery method is used to obtain global and local topic word distributions, along with user topic and social network preferences. After that, the topic distribution characteristics of different social networks are examined, as well as the reasons why users choose one network over another to create a post. Next, a Gibbs sampling algorithm is adopted to obtain the model parameters. In the experiment, we collect data from Twitter, Instagram, and Tumblr websites to build a dataset of multiple social networks. Finally, we compare our research to previous works, and both qualitative and quantitative evaluation results have demonstrated the effectiveness.

Many real world networks are very large and constantly change over time. These dynamic networks exist in various domains such as social networks, traffic networks and biological interactions. To handle large dynamic networks in downstream applications such as link prediction and anomaly detection, it is essential for such networks to be transferred into a low dimensional space. Recently, network embedding, a technique that converts a large graph into a low-dimensional representation, has become increasingly popular due to its strength in preserving the structure of a network. Efficient dynamic network embedding, however, has not yet been fully explored. In this paper, we present a dynamic network embedding method that integrates the history of nodes over time into the current state of nodes. The key contribution of our work is 1) generating dynamic network embedding by combining both dynamic and static node information 2) tracking history of neighbors of nodes using LSTM 3) significantly decreasing the time and memory by training an autoencoder LSTM model using temporal walks rather than adjacency matrices of graphs which are the common practice. We evaluate our method in multiple applications such as anomaly detection, link prediction and node classification in datasets from various domains.

Context: The importance of the involved stakeholders and their networks in vocational education and training (VET) focussing on international transfer and cooperation is highlighted in various empirical studies. A systematic empirical survey of these by means of social network analysis, however, has hardly been applied to date. This article is concerned with the development of social capital in the course of network formation and its sustainability. The object of investigation is the funded European VET innovation project AI Pioneers within the Erasmus+ program of the European Union. The main objective of the project is to establish and expand an international network in the context of VET in order to support the exchange of expertise on the use of artificial intelligence (AI) in education. Approach: To answer the research questions, the first step was to combine theoretical approaches from a social network perspective from psychology in relation to the analysis of interpersonal trust, sociology regarding the social capital approach and business administration by addressing the roles of actors in innovation processes. Among others, the social network perspective in this study is based on the work of Granovetter as well as Marsden and Campbell. For the data collection, a fully structured interview questionnaire and a semi-structured interview guideline were developed based on the theoretical framework of the study. In the second step, a multi-perspective egocentric network analysis was carried out: Data on a total of N = 10 egocentric networks were collected from the funded partners in the AI Pioneers project to gain an overall picture of the combined social capital and network structures. For the visualisation of the network data, the type of structured and standardised network maps was used. Findings: Regarding the establishment of social capital in the analysed innovation project AI Pioneers, it can be emphasised that a total of 74 relationships have been recorded in the 10 egocentric networks combined. In line with the project objectives, the education sector is addressed by the majority of the analysed relationships (n = 54), with (technical) vocational schools making up a substantial part of these. Focussing on the sustainability of the surveyed network structures: Most of the analysed relationships already existed before the project start and were consolidated during it (n = 57), while new ones were also established (n = 17). In addition, the continuous development of mutual trust and the need for equal cooperation is emphasised: A relatively high level of mutual trust can be recorded overall in the analysed egocentric networks (n = 55), while a low mutual trust is present in 19 relationships which is described due to e.g. asymmetrical power relations or a lack of commitment. The results show that the relationships analysed primarily contribute their resources in the form of expertise and their networking knowledge to the egocentric networks. Furthermore, a high level of interest and willingness to support the AI Pioneers project can be captured, particularly due to the novelty of the topic and the application of AI in VET. Conclusions: The study makes a significant contribution to VET research and its methodological set by using social network analysis with a combination of qualitative approaches for analysing egocentric networks from multiple perspectives. The importance of allocating resources to the creation of social capital regarding cooperation, network building and the sustainable maintenance of established structures can be emphasised. In this respect the benefits of a network-based approach can be highlighted in the context of the Erasmus+ program and the partnerships for innovation on forward-looking topics. In addition, the development of the two structured survey instruments in this study can be emphasised, which can be further developed on the basis of future research. Further quantitative network analyses would be valuable for VET research, especially against the background of innovation drivers and network formation, such as market and trend-related drivers due to demands and developments in the field of AI in education.

Anchor link prediction exacerbates the risk of privacy leakage via the de-anonymization of social network data. Embedding-based methods for anchor link prediction are limited by the excessive similarity of the associated nodes in a latent feature space and the variation between latent feature spaces caused by the semantics of different networks. In this article, we propose a novel method which reduces the impact of semantic discrepancies between different networks in the latent feature space. The proposed method consists of two phases. First, graph embedding focuses on the network structural roles of nodes and increases the distinction between the associated nodes in the embedding space. Second, a federated adversarial learning framework which performs graph embedding on each social network and an adversarial learning model on the server according to the observable anchor links is used to associate independent graph embedding approaches on different social networks. The combination of distinction enhancement and the association of graph embedding approaches alleviates variance between the latent feature spaces caused by the semantics of different social networks. Extensive experiments on real social networks demonstrate that the proposed method significantly outperforms the state-of-the-art methods in terms of both precision and robustness.

Nowadays, the user linkage or link prediction task is usually based on user profile or some sensitive data (i.e., name, gender, ID, life or health insurance accounts, etc.). With the enhancement of laws and regulations, the difficulty of personal sensitive data acquisition is increasing. Moreover, the abnormal users called online water army often camouflage themselves to achieve specific goals. They often register false user information such as name, gender, age, etc. To protect privacy and satisfy the needs of camouflage, users and ISPs often hide those sensitive data (i.e., user profile). In this paper, we want to link same user in multiple social networks, which is formally defined as ULASN (User Linkage across Anonymized Social Networks) problem. ULASN is very challenging to address due to (1) the lack of enough ground-truth to build models and obtain accurate prediction results, (2) the studied networks are anonymized, where no user profile or sensitive data is available, and (3) the need of scalable algorithms for user linkage task in large-scale social nateworks, and (4) users in social network are interrelated. To resolve these challenges, a noval user linkage framework based on social structures called ULA is proposed in this paper. ULA tackles these problems by considering massive, low-quality and interrelated user information. It uses few ground-truth to partition users into blocks, which reduces the size of candidates. By extending Fellegi-Sunter methods, our proposed algorithm can handle social network similarity complying to continuous distributions. A probabilistic generative model is proposed and solved by EM algorithm. Simultaneously, missing value problem can also solved when we use EM algorithm to learning parameters. Extensive experiments conducted on two real-world social networks demonstrate that ULA can perform very well in solving ULASN problem.

Many famous online social networks, e.g., Facebook and Twitter, have achieved great success in the last several years. Users in these online social networks can establish various connections via both social links and shared attribute information. Discovering groups of users who are strongly connected internally is defined as the community detection problem. Community detection problem is very important for online social networks and has extensive applications in various social services. Meanwhile, besides these popular social networks, a large number of new social networks offering specific services also spring up in recent years. Community detection can be even more important for new networks as high quality community detection results enable new networks to provide better services, which can help attract more users effectively. In this paper, we will study the community detection problem for new networks, which is formally defined as the “New Network Community Detection” problem. New network community detection problem is very challenging to solve for the reason that information in new networks can be too sparse to calculate effective similarity scores among users, which is crucial in community detection. However, we notice that, nowadays, users usually join multiple social networks simultaneously and those who are involved in a new network may have been using other well-developed social networks for a long time. With full considerations of network difference issues, we propose to propagate useful information from other well-established networks to the new network with efficient information propagation models to overcome the shortage of information problem. An effective and efficient method, Cat (Cold stArT community detector), is proposed in this paper to detect communities for new networks using information from multiple heterogeneous social networks simultaneously. Extensive experiments conducted on real-world heterogeneous online social networks demonstrate that Cat can address the new network community detection problem effectively.

The era of today’s world is based on multiple online social network such as twitter, Facebook, LinkedIn and many more. User’s use this online social network for commination in terms of text, audio, video, images, gif’s and so on which leads to enormous amount of unstructured data generation. Hence, it becomes inevitable to analyze these unstructured data into meaningful knowledge which can be applied to various applications such as link prediction, criminology, public health, recommendation system and many more. Most applications of social networks require user profile data to analyze the data. In this paper, we propose a graph based methodology to connect user profiles based on their attributes similarity and build a social network graph of connected users. The methodology is tested on LinkedIn data set and results are promising. The methodology addresses various issues associated with unstructured data analysis.

Social network alignment, identifying social accounts of the same individual across different social networks, shows fundamental importance across a wide spectrum of applications. Individuals more often than not join in multiple social networks and it is in fact intractable or even impossible to acquiring supervision for guiding the alignment. However, to the best of our knowledge, none of existing methods can align multiple social networks without supervision. In this paper, we propose to study the problem of unsupervised multiple social network alignment. To address this problem, we propose a novel unsupervised model of Matrix factorization with diagonal Cone under orthogonal Constraint, referred to as MC <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> . Its core idea is to embed and align multiple social networks in the common subspace via an unsupervised approach. Specifically, in MC <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> model, we first design a matrix optimization to infer the common subspace from different social networks. To address the nonconvex optimization, we then design an efficient alternating algorithm by leveraging its inherent functional property. Through extensive experiments on real-world datasets, we demonstrate that the proposed MC <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> model significantly outperforms the state-of-the-art methods.

The increasing interest in dynamically changing networks has led to growing interest in a more general link prediction problem called temporal link prediction in the data mining and machine learning communities. However, only links in identical time frames are considered in temporal link prediction. We propose a new link prediction problem called cross-temporal link prediction in which the links among nodes in different time frames are inferred. A typical example of cross-temporal link prediction is cross-temporal entity resolution to determine the identity of real entities represented by data objects observed in different time periods. In dynamic environments, the features of data change over time, making it difficult to identify cross-temporal links by directly comparing observed data. Other examples of cross-temporal links are asynchronous communications in social networks such as Face book and Twitter, where a message is posted in reply to a previous message. We adopt a dimension reduction approach to cross-temporal link prediction, that is, data objects in different time frames are mapped into a common low-dimensional latent feature space, and the links are identified on the basis of the distance between the data objects. The proposed method uses different low-dimensional feature projections in different time frames, enabling it to adapt to changes in the latent features over time. Using multi-task learning, it jointly learns a set of feature projection matrices from the training data, given the assumption of temporal smoothness of the projections. The optimal solutions are obtained by solving a single generalized eigenvalue problem. Experiments using a real-world set of bibliographic data for cross-temporal entity resolution showed that introducing time-dependent feature projections improves the accuracy of link prediction.

In today’s online social networks, it becomes essential to help newcomers as well as existing community members to find new social contacts. In scientific literature, this recommendation task is known as link prediction. Link prediction has important practical applications in social network platforms. It allows social network platform providers to recommend friends to their users. Another application is to infer missing links in partially observed networks. The shortcoming of many of the existing link prediction methods is that they mostly focus on undirected graphs only. This work closes this gap and introduces link prediction methods and metrics for directed graphs. Here, we compare well-known similarity metrics and their suitability for link prediction in directed social networks. We advance existing techniques and propose mining of subgraph patterns that are used to predict links in networks such as GitHub, GooglePlus, and Twitter. Our results show that the proposed metrics and techniques yield more accurate predictions when compared with metrics not accounting for the directed nature of the underlying networks.

Recommender systems based on multiple social networks correlation

The increasing popularity and diversity of social media sites has encouraged more and more people to participate on multiple online social networks to enjoy their services. Each user may create a user identity, which can includes profile, content, or network information, to represent his or her unique public figure in every social network. Thus, a fundamental question arises -- can we link user identities across online social networks? User identity linkage across online social networks is an emerging task in social media and has attracted increasing attention in recent years. Advancements in user identity linkage could potentially impact various domains such as recommendation and link prediction. Due to the unique characteristics of social network data, this problem faces tremendous challenges. To tackle these challenges, recent approaches generally consist of (1) extracting features and (2) constructing predictive models from a variety of perspectives. In this paper, we review key achievements of user identity linkage across online social networks including stateof- the-art algorithms, evaluation metrics, and representative datasets. We also discuss related research areas, open problems, and future research directions for user identity linkage across online social networks.

Recently, the number of people who are members of multiple online social networks simultaneously has increased. However, if these people share everything with others, they risk their privacy. Users may be unaware of the privacy risks involved with sharing their sensitive information on a network. Currently, there are many research efforts focused on social identity linkage (SIL) on multiple online social networks for commercial services, which exacerbates privacy issues. Many existing studies consider methods of encrypting or deleting sensitive information without considering if this is unreasonable for social networks. Meanwhile, these studies ignore privacy awareness, which is rudimentary and critical. To enhance privacy awareness, we discuss a user privacy exposure measure for users who are members of multiple online social networks. With this measure, users can be aware of the state of their privacy and their position on a privacy measurement scale. Additionally, we propose a straightforward method through our framework to reduce information loss and foster user privacy awareness by using spurious content for required fields.