Catálogo de publicaciones - libros

This paper describes the application of percolation theory to a Japanese business relation network composed of approximately 3,000,000 links. In this network, we examined the process in which links are randomly removed. At the percolation transition point, we calculate the survival rate for each node as an indicator of its global network connectivity. The basic properties of each node are determined in connection with the values characterising these complex networks, such as the link number and job category. We confirm that this index has strong correlation with degree and shell number, also has significant correlation with sales and number of employee. Finally, we define the network robustness for each prefecture in Japan by using this new indicator.

Graph partitioning, or community detection, is an important tool for investigating the structures embedded in real data. The spectral method is a major algorithm for graph partitioning and is also analytically tractable. In order to analyze the performance of the spectral method, we consider a regular graph of two loosely connected clusters, each of which consists of a random graph, i.e., a random graph with a planted partition. Since we focus on the bisection of regular random graphs, whether the unnormalized Laplacian, the normalized Laplacian, or the modularity matrix is used does not make a difference. Using the replica method, which is often used in the field of spin-glass theory, we estimate the so-called detectability threshold; that is, the threshold above which the partition obtained by the method is completely uncorrelated with the planted partition.

Infectious diseases spread through human networks. Susceptible-Infected-Removed (SIR) model is one of the epidemic models to describe infection dynamics on a complex network connecting individuals. In the metapopulation SIR model, each node represents a population (group) which has many individuals. In this paper, we propose a modified metapopulation SIR model in which a latent period is taken into account. We call it SIIR model. We divide the infection period into two stages: an infected stage, which is the same as the previous model, and a seriously ill stage, in which individuals are infected and cannot move to the other populations. The two infectious stages in our modified metapopulation SIR model produce a discontinuous final size distribution. Individuals in the infected stage spread the disease like individuals in the seriously ill stage and never recover directly, which makes an effective recovery rate smaller than the given recovery rate.

The agglomeration effect of population, which explains why many people live near places where many other people also live, is one important interaction that influences human population. We examine the agglomeration effect by measuring the distribution of the logarithmic differences between populations living in two places separated by some distance. The shapes of the distributions of the logarithmic differences closely resemble each other without depending on the regions or the country in cases of small scale of separation distance. This result suggests a unified explanation to understand the population distributions of various regions.

An influenza epidemic is a complicated phenomenon influenced by numerous social and biological factors, including geography, climate, population, transport network, and biological statuses of humans and viruses, among others. To investigate the strength of these influences, we evaluated data from influenza epidemics that occurred in Japan between April 1999 and December 2014 using wavelet analysis. We calculated wavelet transform and phase difference, which was defined as the phase difference in a prefecture other than Tokyo. The time-averaged phase differences revealed the following: (1) the epidemics were earlier and more strongly synchronized in 7 prefectures in the Kanto region, which includes Tokyo, and in 7 prefectures in the Kinki region, which includes Osaka; (2) except for these urban regions, the epidemics propagated from western to eastern prefectures, and finally to northern prefectures; (3) epidemic jumps occurred in several prefectures (e.g., Miyagi Prefecture); and (4) epidemics occasionally occurred at different times in two adjacent prefectures (e.g., Yamanashi and Tokyo). We then attempted to qualitatively deduce the causes for these observations. This study is expected to be important for integrating knowledge to derive trends in epidemics, both nationally and internationally.

Lifetime distributions of social entities, such as enterprises, products, and media contents, are one of the fundamental statistics characterizing the social dynamics. To investigate the lifetime distribution of mutually interacting systems, simple models having a rule for additions and deletions of entities are investigated. We found a quite universal lifetime distribution for various kinds of inter-entity interactions, and it is well fitted by a stretched-exponential function with an exponent close to 1/2. We propose a “modified Red-Queen” hypothesis to explain this distribution. We also review empirical studies on the lifetime distribution of social entities, and discuss the applicability of the model.

In this study, we investigated around one million pieces of Japanese firm-size data, which are included in the database ORBIS, and confirmed that the age distribution of firms approximately obeys an exponential function. We estimated the decay rate of firms by comparing their activities in 2008 and 2013 and found that it does not depend on firm age and can be regarded to be constant. Here, decay rate of firms denotes the state transition probability of firm activities. These two observations are qualitatively consistent when the number of newly founded firms is nearly constant. This phenomenon is analogous to nuclear decay. We quantitatively confirmed this consistency by comparing the parameters of exponential age distribution with the decay rate of firm activities. At the same time, using this result, we estimated the number of firms founded annually and the decay rate of firm activities in Japan before World War II.

We analyze Japanese interfirm trade network data for 20 years from the viewpoint of the metabolism of scale-free network evolution. We find that the preferential attachment effect of established firms is stronger than that of merged firms. This shows that merging firms should choose counterparties using delicate business strategies that may not be related to the degree. We also find that the distribution of lifespan of links is approximated well by an exponential function with the characteristic time of 6 years. The results imply the link creation and deletion is well characterized by a Poisson process.

This paper presents a methodology to study the flow of funds in large value payment systems (LVPSs). The presented algorithm separates the flow of payments in two categories: (1) external funds, i.e. funds transferred from other financial market infrastructures (FMIs) or provided by the central bank and (2) the reuse of incoming payments. Our method further studies the flow of intraday liquidity under the framework of its provision within the Mexican FMIs. The aim is to evaluate the impact of the intraday liquidity provision, and understand how liquidity is transmitted to participants in the Mexican LVPS SPEI.

In this paper, some achievements of research on pedestrian dynamics in Jamology are reviewed. The author focuses on three situations, i.e., one-dimensional unidirectional flow, egress process and queuing process. Experimental, theoretical and simulation results, which give us some prescriptions of easing jam in the situations above, are presented.