A link stream is a collection of triplets (t,u,v) indicating that an interaction occurred between u and v at time t. Link streams model many real-world situations like email exchanges between individuals, connections between devices, and others. Much work is currently devoted to the generalization of classical graph and network concepts to link streams. In this paper, we generalize the existing notions of intra-community density and inter-community density. We focus on emails exchanges in the Debian mailing-list, and show that threads of emails, like communities in graphs, are dense subsets loosely connected from a link stream perspective.
In this paper we show how graph structure can be used to drastically reduce the computational bottleneck of the Breadth First Search algorithm (the foundation of many graph traversal techniques). In particular, we address parallel implementations where the bottleneck is the number of messages between processors emitted at the peak iteration. First, we derive an expression for the expected degree distribution of vertices in the frontier of the algorithm which is shown to be highly skewed. Subsequently, we derive an expression for the expected message along an edge in a particular iteration. This skew suggests a weighted, iteration based, partition would be advantageous. Employing the METIS algorithm we then show empirically that such partitions can reduce the message overhead by up to 50% in some particular instances and in the order of 20% on average. These results have implications for graph processing in multiprocessor and distributed computing environments.
Comprehending the virality of a meme can help us in addressing the problems pertaining to disciplines like epidemiology and digital marketing. Therefore, it is not surprising that memetics remains a highly analyzed research topic ever since the mid 1990s. Some scientists choose to investigate the intrinsic contagiousness of a meme while others study the problem from a network theory perspective. In this paper, we revisit the idea of a core-periphery structure and apply it to understand the trajectory of a viral meme in a social network. We have proposed shell-based hill climbing algorithms to determine the path from a periphery shell(where the meme originates) to the core of the network. Further simulations and analysis on the networks behavioral characteristics helped us unearth specialized shells which we term Pseudo-Cores. These shells emulate the behavior of the core in terms of size of the cascade triggered. In our experiments, we have considered two sets for the target nodes, one being core and the other being any of the pseudo-core. We compare our algorithms against already existing path finding algorithms and validate the better performance experimentally.
PropelledMin, Semi by thePark, Juyong recent advances in digitization of books and computational methods for automated text analysis, we are witnessing a promising opportunity for a serious scientific study of narratives. The importance of such an endeavor stems from the fact that a good story, albeit often fictional and artificial, is composed of highly believable characters who interact and experience a sequence of events together in a realistic world setting, and thus a better understanding of narratives may yield new insights for comprehending various real social phenomena as well as literary fiction. Here we present the basic scientific framework for modeling narrative as complex networks, which allows us to study how the narrative structure is reflected in the network of characters and how they allow us to understand the dynamics of narrative progression. This paper contains the fundamental network model of narratives and its properties that serves as the starting point for a more comprehensive future work.
EvaluatingRossetti, Giulio a community detectionPappalardo, Luca algorithmRinzivillo, Salvatore is a complex task due to the lack of a shared and universally accepted definition of community. In literature, one of the most common way to assess the performances of a community detection algorithm is to compare its output with given ground truth communities by using computationally expensive metrics (i.e., Normalized Mutual Information). In this paper we propose a novel approach aimed at evaluating the adherence of a community partition to the ground truth: our methodology provides more information than the state-of-the-art ones and is fast to compute on large-scale networks. We evaluate its correctness by applying it to six popular community detection algorithms on four large-scale network datasets. Experimental results show how our approach allows to easily evaluate the obtained communities on the ground truth and to characterize the quality of community detection algorithms.
The aim of this studyTerekhov, Ivan is to define an appropriateEvans, Antony approach to forecast the appearance and disappearanceGollnick, Volker of air passenger demand between cities worldwide. For the air passenger demand link forecasting, a weighted similarity-based algorithm is used, with an analysis of nine indices. The weighted resource allocation index demonstrates the best metrics. The accuracy of this method has been determined through a comparison of modeled and known data from three separate years. The known data was used to establish boundaries when applying the similarity-based algorithm. As a result, it is found that a weighted resource allocation index, with defined boundaries, should be utilized for link prediction in the air passenger demand network. Furthermore, it is shown that grouping cities within the air passenger demand network, based on socio-economic indicators, increases the accuracy of the forecast.
CommunityChen, Mingming detectionBahulkar, Ashwin isKuzmin, Konstantin anSzymanski, Boleslaw K. important step of network analysis that relies on the correctness of edges. However, incompleteness and inaccuracy of network data collection methods often cause the communities based on the collected datasets to be different from the ground truth. In this paper, we aim to recover or improve the network community structure using scores provided by different link prediction techniques to replace a fraction of low ranking existing links with top ranked predicted links. Experimental results show that applying our approach to different networks can significantly refine community structure. We also show that predictions of edge additions and persistence are confirmed by the future states of evolving social networks. Another important finding is that not every metric performs equally well on all networks. We observe that performance of link prediction ranking is correlated with certain network properties, such as the network size or average node degree.
Cyclic behaviour and circadian patterns emerging from the editing activity of Wikipedia are hereby considered. Such patterns affect many human activities, mobility routes, energy storage and synchronization, among others. Because the editing of Wikipedia is the result of a voluntary process made by many independent human beings, the question about the signature of such circadian patterns on such data is not straightforward. We however show in this work that Wikipedia editing presents well defined periodic patterns with respect to daily, weekly and monthly activity. In addition, we also show the periodic nature of the number of inter-event in time. The results of our work shed some light on the activity scheduling present in our society, contributing to the circadian patterns understanding.
Good news should answer the following questions: ‘Who?’, ‘Where?’, ‘When?’, ‘What?’, and possibly ‘Why?’. We present an approach whichSluban, Borut extracts interestingGrčar, Miha events from thousands of daily news. We construct a time-varying, three-layer network where the nodes are entities of interestMozetič, Igor in the news. The temporal aspect of the network answers the ‘When?’ question. The layers are: (1) the co-occurrence of entities which answers the ‘Who?’ or ‘Where?’, (2) the summary layer which answers the ‘What?’, and (3) the sentiment layer which labels the links as ‘good’ or ‘bad’ news. We demonstrate the news network evolution over a period of four years in an interactive web portal.
The organizationUzun, Thais Gobet of individualsRibeiro, Carlos Henrique Costa in groups or communities is an observed property of complex social networks and this structural organization emerges naturally due to the relationships built between people on a daily basis. We believe that the opinion exchange among individuals is a key factor to this community construction, given that sharing opinions bounds people together, and disagreeing constantly would probably weaken a relationship. In this work, we analyse three models of opinion exchange that uncover the community structure of a network, based on the Naming Game (NG), a classic model of linguistic interactions of agreement. The NG-based models applied in this work insert time-changing social features to the NG dynamics in order to form communities of nodes sharing different language conventions. For this matter, we explore the models NG-AW—that incorporates trust—, NG-LEF—that incorporates uncertainty—and NG-SM—finally incorporating opinion preference. We test the algorithms in LFR networks and show that the separate addition of each social feature in the Naming Game results in improvements in community detection. Our simulations show that opinions coexist at the end of the game in non-convergent executions, each name tagging a different community, identifying, by a socially guided language dynamics, the topological communities present on the network. Moreover, the resulting trust in edges and uncertainty in nodes classify them according to role and position in the network, respectively. We observed this behavior in large networks with disjoint communities generated using LFR benchmark, and we compared our results with existing results from the literature, focusing on the quality of the community detection per se. Our model with secondary memory has shown accuracy comparable with algorithms designed specifically for topological community detection, while modeling social features that reveal communities as an emergent property, as observed in real-world social systems.
A better understandingBarbosa, Hugo S. of the fundamentalDe Lima Neto, Fernando B. mechanisms underlying complex human dynamics is of major interest inEvsukoff, Alexandre contemporaryMenezes, Ronaldo social research. Over the last few years, researchers have made huge strides towards this understanding, thanks especially to the increasing availability of datasets containing digital traces of many human activities. In this work, we investigate Web browsing trajectories using a human mobility approach based on approximately four years of browsing history data. Our findings strongly suggest that return visitation patterns in browsing behaviors and in human mobility exhibit very similar scaling properties. Moreover, we classify Web users as returners and explorers based on their on-line activities, and show that at a population level, the distribution of both profiles agrees with empirical observations in human mobility. Finally, we create a network representation of the most popular websites from the aggregated browsing trajectories and uncover many functional clusters related with different users’ activities.
When linearityGueye, Ibrahima can be rigorouslyNdong, Joseph assumed for stochastic processes, the linear Kalman filter can be used as a powerful tool for anomaly detection in communication networks. However, this assumption done with a strong evidence is not generally proved in a rigorous way. So it is important to develop other methodology, for the scope of anomaly detection, which are not obliged to be based on that assumption. This paper is focused on the use of particle filtering to build a normal behavioral model for an anomaly detector. The particle filter is calibrated for entropy reduction for the scope of noise reduction in the measurements. With the help of a mixture of normal distributions, we can reuse the filtered observations to identify anomalous events in a few number of classes. Generally anomalies might be rare and thus they might happen on a few clusters. So, using a new decision process based on a hidden markov model, we can track and identify the potential abnormal clusters. We study the performances of this system by analyzing the false alarm rate vs detection rate trade-off by means of Receiver Operating Characteristic curve, and compare the results with the Kalman filter. We validate the approach to track volume anomalies over real network traffic.
Traditional music recommendation systems rely on collaborative filtering to recommend songs or artists. This is computationally efficient and performs well method but is not effective when there is limited or no user input. For these cases, it may be useful to consider content-based recommendation. This paper considers a content-based recommendation system based on lyrical data. We compare a complex network of lyrical recommendations to an equivalent collaborative filtering network. We used user generated tag data from Last.fm to produce 23 subgraphs of each network based on tag categories representing musical genre, mood, and gender of vocalist. We analyzed these subgraphs to determine how recommendations within each network tend to stay within tag categories. Finally, we compared the lyrical recommendations to the collaborative filtering recommendations to determine how well lyrical recommendations perform. We see that the lyrical network is significantly more clustered within tag categories than the collaborative filtering network, particularly within small musical niches, and recommendations based on lyrics alone perform 12.6 times better than random recommendations.
NetworkAmorim, Bernardo growthFigueiredo, Daniel andIacobelli, Giulio evolutionNeglia, Giovanni is a fundamental theme that has puzzled scientists for the past decades. A number of models have been proposed to capture important properties of real networks. In an attempt to better describe reality, more recent growth models embody local rules of attachment, however they still require a primitive to randomly select an existing network node and then some kind of global knowledge about the network (at least the set of nodes and how to reach them). We propose a purely local network growth model that makes no use of global sampling across the nodes. The model is based on a continuously moving random walk that after s steps connects a new node to its current location, but never restarts. Through extensive simulations and theoretical arguments, we analyze the behavior of the model finding a fundamental dependency on the parity of s, where networks with either exponential or a conditional power law degree distribution can emerge. As s increases parity dependency diminishes and the model recovers the degree distribution of Barabási-Albert preferential attachment model. The proposed purely local model indicates that networks can grow to exhibit interesting properties even in the absence of any global rule, such as global node sampling.
The study of user diversitySilveira, Ricardo in online socialIacobelli, Giulio networks is an importantFigueiredo, Daniel and ongoing research effort to better understand human behavior. This work takes a step in this direction by providing an empirical study of around 8,000 athletes divided into 13 categories and followed by 197 million users in Twitter. We propose a metric for follower diversity at the category level that factors the vast popularity difference between categories (e.g., soccer versus golf). Using this metric, we propose a measure for athlete heterogeneity based on the diversity of his/her followers. Our findings reveal that follower diversity is spread across two scales with the vast majority of users having very small diversity. We also find that athlete heterogeneity is inversely proportional to its number of followers. This indicates that very popular athletes are followed by users that (on average) do not follow other sports.
RecentAlrasheed, Hend advancesDragan, Feodor F. in the field of genetic data analysis reveal promising findings in the field of human history; especially when combined with proper data analysis tools. Within the field of modern genetics, there is evidence that the human populations have genetically interacted as a result of several events. The genetic admixture contains multiple pieces of DNA that have been passed down subsequently through generations making it combine DNA from different source groups. In this paper, we construct and analyze the network of human genetic admixture. We study the topology of this network, we investigate its δ -hyperbolicity (negative curvature), and, using it, identify the core vertices by proposing the δ -hyperbolicity-neighborhood measure that we assign to each vertex.
Inferring the structureDavis, Benjamin of an unknownRye, Erik C. network is a difficult problemGera, Ralucca of interest to researchers, academicsLim, Bing Yong , and industrialistsLazzaro, Gary . We develop a novel algorithm to infer nodes and edges in an unknown network. Our algorithm utilizes monitors that detect incident edges and adjacent nodes with their labels and degrees. The algorithm infers the network through a preferential random walk with a probabilistic restart at a previously discovered but unmonitored node, or a random teleportation to an unexplored node. Our algorithm outperforms random walk inference and random placement of monitors inference in edge discovery in all test cases. Our algorithm outperforms both methodologies in node inference in synthetic test networks; on real networks it outperforms them in the beginning of the inference. Finally, a website was created where these algorithms can be tested live on preloaded networks or custom networks as desired by the user. The visualization also displays the network as it is being inferred, and provides other statistics about the real and inferred networks.
Many researchers identify influentials in a network by their betweenness centrality. Whereas betweenness centrality can be calculated in small, static, connected networks, its calculation in complex, large, evolving networks frequently causes some problems. Hence, we propose a proxy variable for a node’s betweenness centrality that can be calculated in large, evolving networks. We illustrate our approach using the example of Key Opinion Leader (KOL) identification in an evolving co-authorship network of researchers who have published articles about PCSK9 (a protein that regulates cholesterol levels).
