Complex Network Approach Research Articles

Histogram of the node strength and histogram of the edge weight: two new features for RGB-D person re-identification

Person re-identification is a classical task for any multi-camera surveillance system. Most of the existing researches on re-identification are based on features extracted from RGB images. However, there are many deficiencies in RGB image processing, some of which are requiring a lot of illumination and high computation. In this paper, novel features are proposed for RGB-D person re-identification. First, the complex network approach in texture recognition is modified and its threshold function is changed for using in depth images extracted by RGB-D sensors. Then, two novel measurements named the histogram of the edge weight (HEW) and the histogram of the node strength (HNS) are introduced on complex networks. Our features fit both single-shot and multi-shot person re-identification. In the single-shot case, the HNS is extracted from only one frame while for the multi-shot case it is extracted from both one frame and multi-frames. These proposed measurements are called histogram of the spatial node strength (HSNS) and histogram of the temporal node strength (HTNS) respectively. Subsequently, these measurements are combined with skeleton features using score-level fusion. The method is evaluated using two benchmark databases and the results show that ours outperforms some state-of-the-art methods.

Power System Structural Vulnerability Assessment Based on an Improved Maximum Flow Approach

With the increasing complexity of the power grid and the concerns on major blackouts, there is an urgent need for an efficient and effective tool to assess the power system structural vulnerability. To tackle this challenge, this paper proposes a maximum flow-based complex network approach to identify the critical lines in a system. The proposed method consists of two major steps. First, the power network is modeled as a graph with edges (transmission lines, transformers, etc.) and nodes (buses, substations, etc.). The critical scenarios are identified by using the principal component analysis and convex hull. Then the second step is to use an improved maximum flow-based complex network approach for topology analysis. Weighted vertices in the network are considered, enabling taking the selected operating conditions into consideration when identifying the vulnerable lines. The proposed method is validated using the western Danish power system. The vulnerable lines in the network are ranked. Simulation results demonstrate the effectiveness of the method by intentional attacks and comparison with the planning strategy from the system operator.

Interprovincial transfer of embodied primary energy in China: A complex network approach

The energy supply–demand security and climate change has continued to be problematic, making it significant and necessary to investigate embodied energy flow, particularly in a large and fast-growing developing country like China. One of the effective approaches is the energy/emissions embodied in bilateral trade (EEBT) aiming to locate the destination of energy bi-directionally to evaluate how energy flow between producer and consumer sectors. However, in addition to the flow of energy and resources, the topological structure and impact of underlying components from a system science perspective are equally important for policy-making. This study therefore constructs an energy embodied in trade network (EETN) model to track multi-layer primary energy flow by integrating the EEBT approach and complex network analysis. The embodied coal, oil, natural gas, and non-fossil fuels associated with China’s 30 provinces/municipalities are quantified at the provincial level. By the joint analysis of the network-oriented metrics, the EETN model elicits the possibility of understanding the heterogeneity distribution of different types of energy flow and the potential impact of province-specific policy interventions. We explain how resource endowment, economic growth, income inequality, cross-provincial industrial transfer, and infrastructures affect China’s provincial energy embodiments as well as the clustering features. Other findings and policy recommendations are also presented.

Word sense disambiguation: A complex network approach

The word sense disambiguation (WSD) task aims at identifying the meaning of words in a given context for specific words conveying multiple meanings. This task plays a prominent role in a myriad of real world applications, such as machine translation, word processing and information retrieval. Recently, concepts and methods of complex networks have been employed to tackle this task by representing words as nodes, which are connected if they are semantically similar. Despite the increasingly number of studies carried out with such models, most of them use networks just to represent the data, while the pattern recognition performed on the attribute space is performed using traditional learning techniques. In other words, the structural relationships between words have not been explicitly used in the pattern recognition process. In addition, only a few investigations have probed the suitability of representations based on bipartite networks and graphs (bigraphs) for the problem, as many approaches consider all possible links between words. In this context, we assess the relevance of a bipartite network model representing both feature words (i.e. the words characterizing the context) and target (ambiguous) words to solve ambiguities in written texts. Here, we focus on semantical relationships between these two type of words, disregarding relationships between feature words. The adopted method not only serves to represent texts as graphs, but also constructs a structure on which the discrimination of senses is accomplished. Our results revealed that the adopted learning algorithm in such bipartite networks provides excellent results mostly when local features are employed to characterize the context. Surprisingly, our method even outperformed the support vector machine algorithm in particular cases, with the advantage of being robust even if a small training dataset is available. Taken together, the results obtained here show that the representation/classification used for the WSD problem might be useful to improve the semantical characterization of written texts without the use of deep linguistic information.

Time irreversibility and intrinsics revealing of series with complex network approach

In this work, we analyze time series on the basis of the visibility graph algorithm that maps the original series into a graph. By taking into account the all-round information carried by the signals, the time irreversibility and fractal behavior of series are evaluated from a complex network perspective, and considered signals are further classified from different aspects. The reliability of the proposed analysis is supported by numerical simulations on synthesized uncorrelated random noise, short-term correlated chaotic systems and long-term correlated fractal processes, and by the empirical analysis on daily closing prices of eleven worldwide stock indices. Obtained results suggest that finite size has a significant effect on the evaluation, and that there might be no direct relation between the time irreversibility and long-range correlation of series. Similarity and dissimilarity between stock indices are also indicated from respective regional and global perspectives, showing the existence of multiple features of underlying systems.

Complex Network Study of the Immune Epitope Database for Parasitic Organisms.

Complex network approach allows the representation and analysis of complex systems of interacting agents in an ordered and effective manner, thus increasing the probability of discovering significant properties of them. In the present study, we defined and built for the first time a complex network based on data obtained from Immune Epitope Database for parasitic organisms. We then considered the general topology, the node degree distribution, and the local structure (triadic census) of this network. In addition, we calculated 9 node centrality measures for observed network and reported a comparative study of the real network with three theoretical models to detect similarities or deviations from these ideal networks. The results obtained corroborate the utility of the complex network approach for handling information and data mining within the database under study. They confirm that this type of approach can be considered a valuable tool for preliminary screening of the best experimental conditions to determine whether the amino acid sequences being studied are true epitopes or not.

Comparison of complex networks and tree-based methods of phylogenetic analysis and proposal of a bootstrap method.

Complex networks have been successfully applied to the characterization and modeling of complex systems in several distinct areas of Biological Sciences. Nevertheless, their utilization in phylogenetic analysis still needs to be widely tested, using different molecular data sets and taxonomic groups, and, also, by comparing complex networks approach to current methods in phylogenetic analysis. In this work, we compare all the four main methods of phylogenetic analysis (distance, maximum parsimony, maximum likelihood, and Bayesian) with a complex networks method that has been used to provide a phylogenetic classification based on a large number of protein sequences as those related to the chitin metabolic pathway and ATP-synthase subunits. In order to perform a close comparison to these methods, we selected Basidiomycota fungi as the taxonomic group and used a high-quality, manually curated and characterized database of chitin synthase sequences. This enzymatic protein plays a key role in the synthesis of one of the exclusive features of the fungal cell wall: the presence of chitin. The communities (modules) detected by the complex network method corresponded exactly to the groups retrieved by the phylogenetic inference methods. Additionally, we propose a bootstrap method for the complex network approach. The statistical results we have obtained with this method were also close to those obtained using traditional bootstrap methods.

Complex networks approach for depth of anesthesia assessment

Despite numerous attempts to develop a reliable depth of anesthesia (DoA) index to avoid patients’ intraoperative awareness during surgery, designing an accurate DoA index is a grand challenge in anesthesia research. In this paper, an attempt is made to design a new DoA index. We applied a statistical model and spectral graph wavelet transform (SGWT) to monitor the DoA. The de-noised electroencephalography (EEG) signals are partitioned into segments using a window technique. The window size is determined empirically, then each EEG segment is divided into sub-blocks to make the signal quasi stationary. 10 statistical characteristics are extracted from each sub-block. As a result, a vector of statistical characteristics is pulled out from each segment. Each vector of the features is then mapped as a weighted graph and spectral graph wavelet transform is performed. The total energy of wavelet coefficients at different scales is tested. The energy of wavelet coefficients at scale 3 is selected to form a SGWTDoA function. The SGWTDoA is evaluated using an anesthesia EEG recordings and the bispectral (BIS) from 22 subjects. The Bland-Altman, regression, Q-Q plot and Pearson correlation are used to verify the agreement between the SGWTDoA and the BIS. The experimental results demonstrate that the SGWTDoA has the ability to estimate the DoA accurately. The SGWTDoA is also compared and tested with the BIS in the case of poor signal quality. Our findings show that, the SGWTDoA can reflect the transition from unconsciousness to consciousness efficiently even for a poor signal while the BIS fails to display the DoA values on the monitor.

A Complex Network Approach for the Estimation of the Energy Demand of Electric Mobility

We study how renewable energy impacts regional infrastructures considering the full deployment of electric mobility at that scale. We use the Sardinia Island in Italy as a paradigmatic case study of a semi-closed system both by energy and mobility point of view. Human mobility patterns are estimated by means of census data listing the mobility dynamics of about 700,000 vehicles, the energy demand is estimated by modeling the charging behavior of electric vehicle owners. Here we show that current renewable energy production of Sardinia is able to sustain the commuter mobility even in the theoretical case of a full switch from internal combustion vehicles to electric ones. Centrality measures from network theory on the reconstructed network of commuter trips allows to identify the most important areas (hubs) involved in regional mobility. The analysis of the expected energy flows reveals long-range effects on infrastructures outside metropolitan areas and points out that the most relevant unbalances are caused by spatial segregation between production and consumption areas. Finally, results suggest the adoption of planning actions supporting the installation of renewable energy plants in areas mostly involved by the commuting mobility, avoiding spatial segregation between consumption and generation areas.

A complex network approach reveals a pivotal substructure of genes linked to schizophrenia.

Research on brain disorders with a strong genetic component and complex heritability, such as schizophrenia, has led to the development of brain transcriptomics. This field seeks to gain a deeper understanding of gene expression, a key factor in exploring further research issues. Our study focused on how genes are associated amongst each other. In this perspective, we have developed a novel data-driven strategy for characterizing genetic modules, i.e., clusters of strongly interacting genes. The aim was to uncover a pivotal community of genes linked to a target gene for schizophrenia. Our approach combined network topological properties with information theory to highlight the presence of a pivotal community, for a specific gene, and to simultaneously assess the information content of partitions with the Shannon’s entropy based on betweenness. We analyzed the publicly available BrainCloud dataset containing post-mortem gene expression data and focused on the Dopamine D2 receptor, encoded by the DRD2 gene. We used four different community detection algorithms to evaluate the consistence of our approach. A pivotal DRD2 community emerged for all the procedures applied, with a considerable reduction in size, compared to the initial network. The stability of the results was confirmed by a Dice index ≥80% within a range of tested parameters. The detected community was also the most informative, as it represented an optimization of the Shannon entropy. Lastly, we verified the strength of connection of the DRD2 community, which was stronger than any other randomly selected community and even more so than the Weighted Gene Co-expression Network Analysis module, commonly considered the standard approach for such studies. This finding substantiates the conclusion that the detected community represents a more connected and informative cluster of genes for the DRD2 community, and therefore better elucidates the behavior of this module of strongly related DRD2 genes. Because this gene plays a relevant role in Schizophrenia, this finding of a more specific DRD2 community will improve the understanding of the genetic factors related with this disorder.

Towards a comprehensive understanding of emerging dynamics and function of pancreatic islets: A complex network approach: Comment on “Network science of biological systems at different scales: A review” by Gosak et al.

Towards a comprehensive understanding of emerging dynamics and function of pancreatic islets: A complex network approach: Comment on “Network science of biological systems at different scales: A review” by Gosak et al.

A Comparative Analysis of the Predictive Abilities of Economic Complexity Metrics Using International Trade Network

The complex networks approach has proven to be an effective tool to understand and predict the evolution of a wide range of complex systems. In this work, we consider the network representing the exchange of goods between countries: the international trade network. According to the type of goods they export, the complex networks approach allows inferring which countries will have a bigger growth compared to others. The aim of this work is to study three different methods characterizing the complex networks and study their behaviour on two main topics. Can the method predict the economic evolution of a country? What happens to those methods when we merge the economies?

Credit Card Fraud Detection through Parenclitic Network Analysis

The detection of frauds in credit card transactions is a major topic in financial research, of profound economic implications. While this has hitherto been tackled through data analysis techniques, the resemblances between this and other problems, like the design of recommendation systems and of diagnostic/prognostic medical tools, suggest that a complex network approach may yield important benefits. In this paper we present a first hybrid data mining/complex network classification algorithm, able to detect illegal instances in a real card transaction data set. It is based on a recently proposed network reconstruction algorithm that allows creating representations of the deviation of one instance from a reference group. We show how the inclusion of features extracted from the network data representation improves the score obtained by a standard, neural network-based classification algorithm and additionally how this combined approach can outperform a commercial fraud detection system in specific operation niches. Beyond these specific results, this contribution represents a new example on how complex networks and data mining can be integrated as complementary tools, with the former providing a view to data beyond the capabilities of the latter.

Network Structure Reveals Patterns of Legal Complexity in Human Society: The Case of the Constitutional Legal Network

Complexity in nature has been broadly found not only in physical and biological systems but also in social and economic systems. Although many studies have examined complex systems and helped us understand real-world complexity, the investigation to the legal complexity has not been thoroughly investigated. Here we introduce a novel approach to studying complex legal systems using complex network approaches. On the basis of the bipartite relations among Constitution articles and Court decisions, we built a complex legal network and found the system shows the heterogeneous structure as generally observed in many complex social systems. By treating legal networks as unique political regimes, we examine whether structural properties of the systems have been influenced as the society changes, or not. On one hand, there is a core structure in all legal networks regardless of any social circumstances. On the other hand, with relative comparison among different regimes’ networks, we could identify characteristic structural properties that reveal their identity. Our analysis would contribute to provide a better understanding of legal complexity and practical guidelines for use in various legal and social applications.

Mapping the hierarchical structure of the global shipping network by weighted ego network analysis

An improved complex network approach named weighted ego network analysis (WENA) was employed to explore the hierarchical structure of the global shipping network. First, it was found that all centrality indices exhibited scale-free properties with obvious power-law distributions. That is, less than 30% of shipping ports had quite high centrality values, which are powerful hubs of the global shipping system. Those very accessible shipping ports showed significant preferential attachment properties such as 'rich-club' and 'Matthew' effect. Second, results of WENA revealed its core-peripheral hierarchical structure, which could be divided into five levels of sub-networks: first tier (top 5%), second tier (top 5%~15%), third tier (top 15%~30%), fourth tier (top 30%~50%), and fifth tier (the remainder). Finally, five kinds of correlation analyses have been carried out, including degree vs. degree, degree vs. strength, degree vs. clustering coefficient, degree vs. distance and centrality vs. GDP. Results indicated that the connectivity of a shipping port was both significantly influenced by the connectivity of its directly connected ports, the shipping distance and hinterland economic scale.

Mapping the hierarchical structure of the global shipping network by weighted ego network analysis

An improved complex network approach named weighted ego network analysis (WENA) was employed to explore the hierarchical structure of the global shipping network. First, it was found that all centrality indices exhibited scale-free properties with obvious power-law distributions. That is, less than 30% of shipping ports had quite high centrality values, which are powerful hubs of the global shipping system. Those very accessible shipping ports showed significant preferential attachment properties such as 'rich-club' and 'Matthew' effect. Second, results of WENA revealed its core-peripheral hierarchical structure, which could be divided into five levels of sub-networks: first tier (top 5%), second tier (top 5%~15%), third tier (top 15%~30%), fourth tier (top 30%~50%), and fifth tier (the remainder). Finally, five kinds of correlation analyses have been carried out, including degree vs. degree, degree vs. strength, degree vs. clustering coefficient, degree vs. distance and centrality vs. GDP. Results indicated that the connectivity of a shipping port was both significantly influenced by the connectivity of its directly connected ports, the shipping distance and hinterland economic scale.

Patterns of precipitation and soil moisture extremes in Texas, US: A complex network analysis

Abstract Understanding of the spatial and temporal dynamics of extreme precipitation not only improves prediction skills, but also helps to prioritize hazard mitigation efforts. This study seeks to enhance the understanding of spatiotemporal covariation patterns embedded in precipitation (P) and soil moisture (SM) by using an event-based, complex-network-theoretic approach. Events concurrences are quantified using a nonparametric event synchronization measure, and spatial patterns of hydroclimate variables are analyzed by using several network measures and a community detection algorithm. SM–P coupling is examined using a directional event coincidence analysis measure that takes the order of event occurrences into account. The complex network approach is demonstrated for Texas, US, a region possessing a rich set of hydroclimate features and is frequented by catastrophic flooding. Gridded daily observed P data and simulated SM data are used to create complex networks of P and SM extremes. The uncovered high degree centrality regions and community structures are qualitatively in agreement with the overall existing knowledge of hydroclimate extremes in the study region. Our analyses provide new visual insights on the propagation, connectivity, and synchronicity of P extremes, as well as the SM–P coupling, in this flood-prone region, and can be readily used as a basis for event-driven predictive analytics for other regions.

Research on the development efficiency of regional high-end talent in China: A complex network approach

In this paper, based on the panel data of 31 provinces and cities in China from 1991 to 2016, the regional development efficiency matrix of high-end talent is obtained by DEA method, and the matrix is converted into a continuous change of complex networks through the construction of sliding window. Using a series of continuous changes in the complex network topology statistics, the characteristics of regional high-end talent development efficiency system are analyzed. And the results show that the average development efficiency of high-end talent in the western region is at a low level. After 2005, the national regional high-end talent development efficiency network has both short-range relevance and long-range relevance in the evolution process. The central region plays an important intermediary role in the national regional high-end talent development system. And the western region has high clustering characteristics. With the implementation of the high-end talent policies with regional characteristics by different provinces and cities, the relevance of high-end talent development efficiency in various provinces and cities presents a weakening trend, and the geographical characteristics of high-end talent are more and more obvious.

Analysis of the main active ingredients and bioactivities of essential oil from Osmanthus fragrans Var. thunbergii using a complex network approach

BackgroundOsmanthus fragrans has been used as folk medicine for thousands of years. The extracts of Osmanthus fragrans flowers were reported to have various bioactivities including free radical scavenging, anti-inflammation, neuroprotection and antitumor effects. However, there is still lack of knowledge about its essential oil.MethodsIn this work, we analyzed the chemical composition of the essential oil from Osmanthus fragrans var. thunbergii by GC-MS. A complex network approach was applied to investigate the interrelationships between the ingredients, target proteins, and related pathways for the essential oil. Statistical characteristics of the networks were further studied to explore the main active ingredients and potential bioactivities of O. fragrans var. thunbergii essential oil.ResultsA total of 44 ingredients were selected from the chemical composition of O. fragrans var. thunbergii essential oil, and that 191 potential target proteins together with 70 pathways were collected for these compounds. An ingredient-target-pathway network was constructed based on these data and showed scale-free property as well as power-law degree distribution. Eugenol and geraniol were screened as main active ingredients with much higher degree values. Potential neuroprotective and anti-tumor effect of the essential oil were also found. A core subnetwork was extracted from the ingredient-target-pathway network, and indicated that eugenol and geraniol contributed most to the neuroprotection of this essential oil. Furthermore, a pathway-based protein association network was built and exhibited small-world property. MAPK1 and MAPK3 were considered as key proteins with highest scores of centrality indices, which might play an important role in the anti-tumor effect of the essential oil.ConclusionsThis work predicted the main active ingredients and bioactivities of O. fragrans var. thunbergii essential oil, which would benefit the development and utilization of Osmanthus fragrans flowers. The application of complex network theory was proved to be effective in bioactivities studies of essential oil. Moreover, it provides a novel strategy for exploring the molecular mechanisms of traditional medicines.

Analysing journey-to-work data using complex networks

It is well known that journey-to-work (JTW) data can be represented using complex network graphs. What is less evident is the way in which this approach can be used to quantitatively analyse the structure, connectivity and dynamics of commuting behaviour. This paper employs a complex network approach to spatially disaggregated JTW data in order to examine commuting behaviour for six different modes of transport (car, car passengers, train, bus, cycling and walking) within three of the most populous metropolitan areas in Australia. A set of network measures (degree, strength, clustering coefficient, maximum cliques, average shortest path length and betweenness) are computed from both the unweighted and weighted graphs corresponding to JTW data for the Sydney, Melbourne and the South East Queensland regions from the time periods: 2001, 2006 and 2011. Results reveal a number of interesting dynamics, one being that Melbourne exhibits shorter (and presumed to be faster) alternate commuting paths than either Sydney or South East Queensland given its lower betweenness and shortest path values allied with higher clustering coefficients. The interpretation of these metrics demonstrates that complex networks have the capacity to reveal new insights from JTW data, by enabling a more comprehensive, systematic, empirical and fine-grained analysis of changes in commuting behaviour over time.