arXiv subjects

Liudmila Ostroumova Prokhorenkova

Publications and source records attributed to Liudmila Ostroumova Prokhorenkova.

11 recordsLinked to original sources

arXiv2019

Preferential placement for community structure formation

Various models have been recently proposed to reflect and predict different properties of complex networks. However, the community structure, which is one of the most important properties, is not well studied and modeled. In this paper, we suggest a principle called "preferential placement", which allows to model a realistic clustering structure. We provide an extensive empirical analysis of the obtained structure as well as some theoretical results.

physics.soc-ph↗

arXiv2018

Clustering Properties of Spatial Preferential Attachment Model

In this paper, we study the clustering properties of the Spatial Preferential Attachment (SPA) model introduced by Aiello et al. in 2009. This model naturally combines geometry and preferential attachment using the notion of spheres of influence. It was previously shown in several research papers that graphs generated by the SPA model are similar to real-world networks in many aspects. For example, the vertex degree distribution was shown to follow a power law. In the current paper, we study the behaviour of C(d), which is the average local clustering coefficient for the vertices of degree d. This characteristic was not previously analyzed in the SPA model. However, it was empirically shown that in real-world networks C(d) usually decreases as d^{-a} for some a>0 and it was often observed that a=1. We prove that in the SPA model C(d) decreases as 1/d. Furthermore, we are also able to prove that not only the average but the individual local clustering coefficient of a vertex v of degree d behaves as 1/d if d is large enough. The obtained results are illustrated by numerous experiments with simulated graphs.

cs.SI↗

arXiv2017

Local Clustering Coefficient in Generalized Preferential Attachment Models

In this paper, we analyze the local clustering coefficient of preferential attachment models. A general approach to preferential attachment was introduced in earlier, where a wide class of models (PA-class) was defined in terms of constraints that are sufficient for the study of the degree distribution and the clustering coefficient. It was previously shown that the degree distribution in all models of the PA-class follows a power law. Also, the global clustering coefficient was analyzed and a lower bound for the average local clustering coefficient was obtained. We expand the results by analyzing the local clustering coefficient for the PA-class of models. Namely, we analyze the behavior of C(d) which is the average local clustering for the vertices of degree d.

math.PR↗

arXiv2017

Modularity of complex networks models

Modularity is designed to measure the strength of division of a network into clusters (known also as communities). Networks with high modularity have dense connections between the vertices within clusters but sparse connections between vertices of different clusters. As a result, modularity is often used in optimization methods for detecting community structure in networks, and so it is an important graph parameter from a practical point of view. Unfortunately, many existing non-spatial models of complex networks do not generate graphs with high modularity; on the other hand, spatial models naturally create clusters. We investigate this phenomenon by considering a few examples from both sub-classes. We prove precise theoretical results for the classical model of random d-regular graphs as well as the preferential attachment model, and contrast these results with the ones for the spatial preferential attachment (SPA) model that is a model for complex networks in which vertices are embedded in a metric space, and each vertex has a sphere of influence whose size increases if the vertex gains an in-link, and otherwise decreases with time. The results obtained in this paper can be used for developing statistical tests for models selection and to measure statistical significance of clusters observed in complex networks.

math.PR↗

arXiv2017

Assortativity in Generalized Preferential Attachment Models

In this paper, we analyze assortativity of preferential attachment models. We deal with a wide class of preferential attachment models (PA-class). It was previously shown that the degree distribution in all models of the PA-class follows a power law. Also, the global and the average local clustering coefficients were analyzed. We expand these results by analyzing the assortativity property of the PA-class of models. Namely, we analyze the behavior of $d_{nn}(d)$ which is the average degree of a neighbor of a vertex of degree $d$.

math.PR↗

arXiv2017

PageRank in Undirected Random Graphs

PageRank has numerous applications in information retrieval, reputation systems, machine learning, and graph partitioning. In this paper, we study PageRank in undirected random graphs with an expansion property. The Chung-Lu random graph is an example of such a graph. We show that in the limit, as the size of the graph goes to infinity, PageR- ank can be approximated by a mixture of the restart distribution and the vertex degree distribution. We also extend the result to Stochastic Block Model (SBM) graphs, where we show that there is a correction term that depends on the community partitioning.

math.PR↗

arXiv2016

Generating maximally disassortative graphs with given degree distribution

In this paper we consider the optimization problem of generating graphs with a prescribed degree distribution, such that the correlation between the degrees of connected nodes, as measured by Spearman's rho, is minimal. We provide an algorithm for solving this problem and obtain a complete characterization of the joint degree distribution in these maximally disassortative graphs, in terms of the size-biased degree distribution. As a result we get a lower bound for Spearman's rho on graphs with an arbitrary given degree distribution. We use this lower bound to show that for any fixed tail exponent, there exist scale-free degree sequences with this exponent such that the minimum value of Spearman's rho for all graphs with such degree sequences is arbitrary close to zero. This implies that specifying only the tail behavior of the degree distribution, as is often done in the analysis of complex networks, gives no guarantees for the minimum value of Spearman's rho.

math.PR↗

arXiv2015

Recency-based preferential attachment models

Preferential attachment models were shown to be very effective in predicting such important properties of real-world networks as the power-law degree distribution, small diameter, etc. Many different models are based on the idea of preferential attachment: LCD, Buckley-Osthus, Holme-Kim, fitness, random Apollonian network, and many others. Although preferential attachment models reflect some important properties of real-world networks, they do not allow to model the so-called recency property. Recency property reflects the fact that in many real networks vertices tend to connect to other vertices of similar age. This fact motivated us to introduce a new class of models - recency-based models. This class is a generalization of fitness models, which were suggested by Bianconi and Barabasi. Bianconi and Barabasi extended preferential attachment models with pages' inherent quality or fitness of vertices. When a new vertex is added to the graph, it is joined to some already existing vertices that are chosen with probabilities proportional to the product of their fitness and incoming degree. We generalize fitness models by adding a recency factor to the attractiveness function. This means that pages are gaining incoming links according to their attractiveness, which is determined by the incoming degree of the page (current popularity), its inherent quality (some page-specific constant) and age (new pages are gaining new links more rapidly). We analyze different properties of recency-based models. In particular, we show that some distributions of inherent quality lead to the power-law degree distribution.

math.PR↗

arXiv2015

Global clustering coefficient in scale-free weighted and unweighted networks

In this paper, we present a detailed analysis of the global clustering coefficient in scale-free graphs. Many observed real-world networks of diverse nature have a power-law degree distribution. Moreover, the observed degree distribution usually has an infinite variance. Therefore, we are especially interested in such degree distributions. In addition, we analyze the clustering coefficient for both weighted and unweighted graphs. There are two well-known definitions of the clustering coefficient of a graph: the global and the average local clustering coefficients. There are several models proposed in the literature for which the average local clustering coefficient tends to a positive constant as a graph grows. On the other hand, there are no models of scale-free networks with an infinite variance of the degree distribution and with an asymptotically constant global clustering coefficient. Models with constant global clustering and finite variance were also proposed. Therefore, in this paper we focus only on the most interesting case: we analyze the global clustering coefficient for graphs with an infinite variance of the degree distribution. For unweighted graphs, we prove that the global clustering coefficient tends to zero with high probability and we also estimate the largest possible clustering coefficient for such graphs. On the contrary, for weighted graphs, the constant global clustering coefficient can be obtained even for the case of an infinite variance of the degree distribution.

cs.SI↗

arXiv2015

Global clustering coefficient in scale-free networks

In this paper, we analyze the behavior of the global clustering coefficient in scale free graphs. We are especially interested in the case of degree distribution with an infinite variance, since such degree distribution is usually observed in real-world networks of diverse nature. There are two common definitions of the clustering coefficient of a graph: global clustering and average local clustering. It is widely believed that in real networks both clustering coefficients tend to some positive constant as the networks grow. There are several models for which the average local clustering coefficient tends to a positive constant. On the other hand, there are no models of scale-free networks with an infinite variance of degree distribution and with a constant global clustering. In this paper we prove that if the degree distribution obeys the power law with an infinite variance, then the global clustering coefficient tends to zero with high probability as the size of a graph grows.

math.PR↗

arXiv2014

Quick Detection of High-degree Entities in Large Directed Networks

In this paper, we address the problem of quick detection of high-degree entities in large online social networks. Practical importance of this problem is attested by a large number of companies that continuously collect and update statistics about popular entities, usually using the degree of an entity as an approximation of its popularity. We suggest a simple, efficient, and easy to implement two-stage randomized algorithm that provides highly accurate solutions for this problem. For instance, our algorithm needs only one thousand API requests in order to find the top-100 most followed users in Twitter, a network with approximately a billion of registered users, with more than 90% precision. Our algorithm significantly outperforms existing methods and serves many different purposes, such as finding the most popular users or the most popular interest groups in social networks. An important contribution of this work is the analysis of the proposed algorithm using Extreme Value Theory -- a branch of probability that studies extreme events and properties of largest order statistics in random samples. Using this theory, we derive an accurate prediction for the algorithm's performance and show that the number of API requests for finding the top-k most popular entities is sublinear in the number of entities. Moreover, we formally show that the high variability among the entities, expressed through heavy-tailed distributions, is the reason for the algorithm's efficiency. We quantify this phenomenon in a rigorous mathematical way.

cs.SI↗