Display omitted
•We review current approaches to detecting anomalies in social networks.•We identify key attributes of anomalies and use these to categorise detection methods.•We generalise the ...process of anomaly detection into five main steps.•We identify important areas for future research.
Anomalies in online social networks can signify irregular, and often illegal behaviour. Detection of such anomalies has been used to identify malicious individuals, including spammers, sexual predators, and online fraudsters. In this paper we survey existing computational techniques for detecting anomalies in online social networks. We characterise anomalies as being either static or dynamic, and as being labelled or unlabelled, and survey methods for detecting these different types of anomalies. We suggest that the detection of anomalies in online social networks is composed of two sub-processes; the selection and calculation of network features, and the classification of observations from this feature space. In addition, this paper provides an overview of the types of problems that anomaly detection can address and identifies key areas for future research.
Amidst the ongoing technological expansion, IoT-based applications have established their presence in various scenarios, including smart homes, healthcare, wearable devices, and more. However, ...fostering robust social relationships among objects within intelligent environments requires considering a multitude of parameters, from historical transactions data to peer opinions on interaction experiences. Our primary objective is to introduce the VISO concept, a finely tuned approach for community formation and facilitating autonomous interactions among Social IoT objects. Key contributions of VISO include: (i) a recommendation mechanism for cultivating friendships; (ii) the virtualization of devices, known as Virtual Objects, based on the Digital Twin concept; (iii) object ranking, determined by peer evaluations at the edge; and (iv) profiling the identity of each object within the environment. The VISO approach is rooted in the link analysis model, which seeks to uncover connections between individuals and qualify interactions. Additionally, the construction of social circles is guided by the Organizational Network Analysis model, a concept grounded in social networks designed to simplify the initiation of new friendships. The VISO approach stands out through its amalgamation of dynamic trust mechanisms, object virtualization, real-time interaction monitoring, and the highlighting of the most active objects. These characteristics collectively contribute to its effectiveness in managing relationships within the social IoT, while simultaneously addressing challenges posed by alternative models. Preliminary results in a testing environment substantiate the model’s capability to categorize interactions and identify the most beneficial object roles for establishing new relationships. In particular, by selecting the most active objects 30%, they account for a staggering 90% of local actions. These validation outcomes lay the foundation for planned future work and enhancements to the VISO approach. Encouraged by the promising results, the inception of a significant project is on the horizon. The application of the VISO approach to relationship management within a smart campus for ubiquitous education underscores the remarkable versatility and potential of Social IoT. This application has the power to elevate various facets of campus life, ranging from education to resource management, creating a transformative impact within the educational landscape.
•Present the general trends in publications within the field of Social IoT relationship management.•Introduce the VISO approach, emphasizing the unique strategic aspects of the proposed model.•Provide details on the algorithms that aid in the classification and suggestion of relationships.•Conceptualize object virtualization through the incorporation of digital twin technology.•Describe the system’s architecture and its mechanisms for ensuring fluid communication.
Scientific software is a fundamental player in modern science, participating in all stages of scientific knowledge production. Software occasionally supports the development of trivial tasks, while ...at other instances it determines procedures, methods, protocols, results, or conclusions related with the scientific work. The growing relevance of scientific software as a research product with value of its own has triggered the development of quantitative science studies of scientific software. The main objective of this study is to illustrate a link-based webometric approach to characterize the online mentions to scientific software across different analytical frameworks. To do this, the bibliometric software VOSviewer is used as a case study. Considering VOSviewer’s official website as a baseline, online mentions to this website were counted in three different analytical frameworks: academic literature via Google Scholar (988 mentioning publications), webpages via Majestic (1,330 mentioning websites), and tweets via Twitter (267 mentioning tweets). Google scholar mentions shows how VOSviewer is used as a research resource, whilst mentions in webpages and tweets show the interest on VOSviewer’s website from an informational and a conversational point of view. Results evidence that URL mentions can be used to gather all sorts of online impacts related to non-traditional research objects, like software, thus expanding the analytical scientometric toolset by incorporating a novel digital dimension.
Annually, money laundering activities threaten the global economy. Proceeds of these activities may be used to fund further criminal activities and to undermine the integrity of financial systems ...worldwide. For these reasons, money laundering is recognized as a critical risk in many countries. There is an emerging interest from both researchers and practitioners concerning the use of software tools to enhance detection of money laundering activities. In the current economic environment, regulators struggle to stay ahead of the latest scam, and financial institutions are challenged to ensure that they can identify and stop criminal activities, while ensuring that legitimate customers are served more effectively and efficiently. Effective technological solutions are an essential element in the fight against money laundering. Improved data and analytics are key in assisting investigators to focus on suspicious activities. Continually evolving regulations, together with recent instances of money laundering violations by some of the largest financial institutions, have highlighted the need for better technology in managing anti-money laundering activities. This study explores the use of visualization techniques that may assist in efficient identification of patterns of money laundering activities. It demonstrates how link analysis may be applied in detecting suspicious bank transactions. A prototype application (AML2ink) is used for proof-of-concept purposes.
•Study proposed a framework for detection of money laundering activities based on visualization of monetary transactions.•A prototype application (AML2ink) was developed and tested using real data from bank transactions of a large entity.•This research highlights the effectiveness of using visualization to identify suspicious money laundering activities.•Study demonstrated the use of- visualization techniques to enhance ability to “see” patterns and target suspicious ones.•The feasibility of applying low-cost, open-source software to implement such techniques was demonstrated.
This work extends the randomized shortest paths (RSP) model by investigating the net flow RSP and adding capacity constraints on edge flows. The standard RSP is a model of movement, or spread, ...through a network interpolating between a random-walk and a shortest-path behavior (Kivimäki et al., 2014; Saerens et al., 2009; Yen et al., 2008). The framework assumes a unit flow injected into a source node and collected from a target node with flows minimizing the expected transportation cost, together with a relative entropy regularization term. In this context, the present work first develops the net flow RSP model considering that edge flows in opposite directions neutralize each other (as in electric networks), and proposes an algorithm for computing the expected routing costs between all pairs of nodes. This quantity is called the net flow RSP dissimilarity measure between nodes. Experimental comparisons on node clustering tasks indicate that the net flow RSP dissimilarity is competitive with other state-of-the-art dissimilarities. In the second part of the paper, it is shown how to introduce capacity constraints on edge flows, and a procedure is developed to solve this constrained problem by exploiting Lagrangian duality. These two extensions should improve significantly the scope of applications of the RSP framework.
Community detection is an important issue in social network analysis. Most existing methods detect communities through analyzing the linkage of the network. The drawback is that each community ...identified by those methods can only reflect the strength of connections, but it cannot reflect the semantics such as the interesting topics shared by people. To address this problem, we propose a topic oriented community detection approach which combines both social objects clustering and link analysis. We first use a subspace clustering algorithm to group all the social objects into topics. Then we divide the members that are involved in those social objects into topical clusters, each corresponding to a distinct topic. In order to differentiate the strength of connections, we perform a link analysis on each topical cluster to detect the topical communities. Experiments on real data sets have shown that our approach was able to identify more meaningful communities. The quantitative evaluation indicated that our approach can achieve a better performance when the topics are at least as important as the links to the analysis.