Network Unfolding Map by Vertex-Edge Dynamics Modeling

Verri, Filipe Alves Neto; Urio, Paulo Roberto; Zhao, Liang

doi:10.1109/tnnls.2016.2626341

Cited by 13 publications

(10 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In this section, we briefly introduce the particle competition model, originally proposed in 33 and improved in [34][35][36] . When investigated the behavior of this framework, it was showed that a certain level of randomness can largely enhance the learning process, similar to the phenomenon of stochastic resonance, in which the performance of a nonlinear deterministic system can be largely improved by a certain level of noise 4 .…”

Section: Community Detection By Particle Competitionmentioning

confidence: 99%

“…Therefore, depending on the system complexity, the use of only deterministic rules might be insufficient to learn the process behind the system 4 . A recent development of this framework showed that it can identify nonlinear features in boundaries between classes with overlapping structural data 36 . Although other community detection algorithms could be selected in this work, the particle competition method was adopted due to its performance in overlapping data and low computational complexity order, which is lower than other network-based semi-supervised algorithms 34,36 .…”

Section: Community Detection By Particle Competitionmentioning

confidence: 99%

“…A recent development of this framework showed that it can identify nonlinear features in boundaries between classes with overlapping structural data 36 . Although other community detection algorithms could be selected in this work, the particle competition method was adopted due to its performance in overlapping data and low computational complexity order, which is lower than other network-based semi-supervised algorithms 34,36 . Other than that, this method outputs the participation rate of each node on each community (or how strong a node relates to a certain pattern), enabling us to take that into consideration in further developments of this work.…”

Section: Community Detection By Particle Competitionmentioning

confidence: 99%

See 2 more Smart Citations

Temporal Network Pattern Identification by Community Modelling

Gao

Zheng

Vega-Oliveros

et al. 2020

Sci Rep

Self Cite

View full text Add to dashboard Cite

Temporal network mining tasks are usually hard problems. This is because we need to face not only a large amount of data but also its non-stationary nature. In this paper, we propose a method for temporal network pattern representation and pattern change detection following the reductionist approach. The main idea is to model each stable (durable) state of a given temporal network as a community in a sampled static network and the temporal state change is represented by the transition from one community to another. For this purpose, a reduced static single-layer network, called a target network, is constructed by sampling and rearranging the original temporal network. Our approach provides a general way not only for temporal networks but also for data stream mining in topological space. Simulation results on artificial and real temporal networks show that the proposed method can group different temporal states into different communities with a very reduced amount of sampled nodes. In real-world applications, data often arrives in streams and must be analyzed in real-time. Patterns and relations in such data are usually not stable but evolve over time 1,2. In dynamical and non-stationary environments, the data distribution can change, yielding the phenomenon of concept drift 2,3. One example is the stock market, where the massive data exchange is strongly related to macro and micro political events, economic events and other factors, such as natural and man-made disasters. Another example is the pattern of customersâ€ ™ buying preferences, which may depend on the season, availability, inflation rate, etc. Some common types of changes may include a gradual change over time, a recurring or cyclical change, and a sudden or abrupt change. Different concept drift detection and handling schemes may be required for each situation. In this context, uncovering data stream patterns is a fundamental task not only to detect concept drift but also to predict future behaviors. Recently, some works have employed networks as a way for representing data from many real systems 4-7 , leading to complex network research. Complex network refers to large scale graphs with nontrivial connection patterns 8-11. One of the most important features of a complex network is the presence of communities. Detecting communities in these systems has become a fundamental task to help us understand how local patterns (represented by sub-networks) interact and produce global behavior. From a network topology point of view, a community is a sub-graph in which the inner links are dense while the outer connections are relatively sparse 4,12,13. In terms of the information-theoretic field, a community is seen as the module that can diminish or retard the propagation flow of information in the system, for a considerable period of time 14-16. In the machine learning domain, community detection provides new unsupervised learning methods, where each community corresponds to a data cluster in the clustering problem 4. Large systems usually are high-dimensional...

show abstract

Section: Community Detection By Particle Competitionmentioning

confidence: 99%

Section: Community Detection By Particle Competitionmentioning

confidence: 99%

Section: Community Detection By Particle Competitionmentioning

confidence: 99%

See 1 more Smart Citation

Temporal Network Pattern Identification by Community Modelling

Gao

Zheng

Vega-Oliveros

et al. 2020

Sci Rep

Self Cite

View full text Add to dashboard Cite

show abstract

“…A presença e o estudo de estrutura de comunidades são comumente encontrados em diversas redes reais, tais como redes sociais, redes de interação de proteínas na biologia e na bioinformática (JONSSON et al, 2006;ZHANG, 2009), estruturas de fluxo óleo-água (GAO et al, 2015), redes de mobilidade humana (THIEMANN et al, 2010), estrutura espacial de movimento urbano em grandes cidades (ZHONG et al, 2014), redes de elites corporativas nos campos da política e da economia (HEEMSKERK; TAKES, 2016), e muito mais. Comunidades, também conhecidas como estruturas modulares, são comumente definidas como sendo grupos de vértices densamente conectados (CLAUSET; MOORE, 2004;GIRVAN, 2004;COSTA et al, 2007;BLONDEL et al, 2008;FORTUNATO, 2010;LANCICHINETTI et al, 2010;ZHAO, 2012c;MACAU;RUBIDO, 2016;ZHAO, 2016).…”

Section: Detecção De Comunidades Via Remoção Iterativa De Arestasunclassified

“…Encontrar a partição ótima é considerado um problema NP-difícil na maioria dos casos (FORTUNATO, 2010), consequentemente abrindo espaço para ampla pesquisa em meios de se encontrar soluções sub-ótimas em tempo admissível. Diferentes abordagens para detecção de comunidades já foram desenvolvidas, incluindo propriedades espectrais em matrizes de grafos (DONETTI;MUñOZ, 2004;CAPOCCI et al, 2005;YANG;LIU, 2008;MITROVIć;TADIć, 2009), sincronização de osciladores acoplados (ARENAS; DíAZ-GUILERA; PéREZ-VICENTE, 2006), modelo de Potts (REICHARDT; BORNHOLDT, 2004), caminhada e competição de partículas em redes (QUILES et al, 2008;BREVE;QUILES, 2009;ZHAO, 2012c;ZHAO, 2016;VERRI;URIO;ZHAO, 2016), e dinâmica de partículas com forças de atração e repulsão (QUILES; MACAU; RUBIDO, 2016).…”

Section: Detecção De Comunidades Via Remoção Iterativa De Arestasunclassified

Desenvolvimento de técnicas de aprendizado de máquina via sistemas dinâmicos coletivos

Gueleri¹

View full text Add to dashboard Cite

Agradeço à minha família, por me ajudar e me apoiar em todas as decisões que tomei ao longo da vida, inclusive na decisão de realizar esta tese de doutorado. Agradeço ao Prof. Dr. Zhao Liang, orientador no desenvolvimento desta tese, pela paciência e dedicação que teve ao longo desses anos. Mais que um orientador, ele se tornou um grande amigo. Agradeço aos demais professores e pesquisadores, do ICMC-USP e de outras instituições, do doutorado e de outras épocas, que de alguma forma contribuíram para o desenvolvimento desta tese e para a minha formação. Agradeço a todos os amigos que estiveram comigo nesses anos, que não só me ajudaram profissionalmente, mas permitiram também ótimos momentos de descontração. Enfim, expresso aqui minha gratidão a todas as pessoas que, direta ou indiretamente, contribuíram para mais esta etapa que agora estou completando. Felizmente tive saúde, motivação, energia, oportunidades e circunstâncias favoráveis para conseguir concluir todo este trabalho, embora alguns momentos difíceis e noites mal-dormidas também façam parte. Muito importante também foi o apoio financeiro concedido pela Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), por meio de bolsa (Processo 2013/08666-8), que permitiu minha dedicação exclusiva ao desenvolvimento desta tese de doutorado.

show abstract