Aruane M Pineda, Francisco A Rodrigues, Caroline L Alves, Michael Möckel, Thaise G L de O Toutain, Joel Augusto Moura Porto
Abstract Heart disease, also known as cardiovascular disease, encompasses a variety of heart conditions that can result in sudden death for many people. Examples include high blood pressure, ischaemia, irregular heartbeats and pericardial effusion. Electrocardiogram (ECG) signal analysis is frequently used to diagnose heart diseases, providing crucial information on how the heart functions. To analyse ECG signals, quantile graphs (QGs) is a method that maps a time series into a network based on the time-series fluctuation proprieties. Here, we demonstrate that the QG methodology can differentiate younger and older patients. Furthermore, we construct networks from the QG method and use machine-learning algorithms to perform the automatic diagnosis, obtaining high accuracy. Indeed, we verify that this method can automatically detect changes in the ECG of elderly and young subjects, with the highest classification performance for the adjacency matrix with a mean area under the receiver operating characteristic curve close to one. The findings reported here confirm the QG method’s utility in deciphering intricate, nonlinear signals like those produced by patient ECGs. Furthermore, we find a more significant, more connected and lower distribution of information networks associated with the networks from ECG data of the elderly compared with younger subjects. Finally, this methodology can be applied to other ECG data related to other diseases, such as ischaemia.
{"title":"Analysis of quantile graphs in EGC data from elderly and young individuals using machine learning and deep learning","authors":"Aruane M Pineda, Francisco A Rodrigues, Caroline L Alves, Michael Möckel, Thaise G L de O Toutain, Joel Augusto Moura Porto","doi":"10.1093/comnet/cnad030","DOIUrl":"https://doi.org/10.1093/comnet/cnad030","url":null,"abstract":"Abstract Heart disease, also known as cardiovascular disease, encompasses a variety of heart conditions that can result in sudden death for many people. Examples include high blood pressure, ischaemia, irregular heartbeats and pericardial effusion. Electrocardiogram (ECG) signal analysis is frequently used to diagnose heart diseases, providing crucial information on how the heart functions. To analyse ECG signals, quantile graphs (QGs) is a method that maps a time series into a network based on the time-series fluctuation proprieties. Here, we demonstrate that the QG methodology can differentiate younger and older patients. Furthermore, we construct networks from the QG method and use machine-learning algorithms to perform the automatic diagnosis, obtaining high accuracy. Indeed, we verify that this method can automatically detect changes in the ECG of elderly and young subjects, with the highest classification performance for the adjacency matrix with a mean area under the receiver operating characteristic curve close to one. The findings reported here confirm the QG method’s utility in deciphering intricate, nonlinear signals like those produced by patient ECGs. Furthermore, we find a more significant, more connected and lower distribution of information networks associated with the networks from ECG data of the elderly compared with younger subjects. Finally, this methodology can be applied to other ECG data related to other diseases, such as ischaemia.","PeriodicalId":15442,"journal":{"name":"Journal of complex networks","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135253942","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract We consider several families of network centrality measures induced by graph kernels, which include some well-known measures and many new ones. The Self-consistency and Bridge axioms, which appeared earlier in the literature, are closely related to certain kernels and one of the families. We obtain a necessary and sufficient condition for Self-consistency, a sufficient condition for the Bridge axiom, indicate specific measures that satisfy these axioms and show that under some additional conditions they are incompatible. PageRank centrality applied to undirected networks violates most conditions under study and has a property that according to some authors is ‘hard to imagine’ for a centrality measure. We explain this phenomenon. Adopting the Self-consistency or Bridge axiom leads to a drastic reduction in survey time in the culling method designed to select the most appropriate centrality measures.
{"title":"Selection of centrality measures using Self-consistency and Bridge axioms","authors":"Pavel Chebotarev","doi":"10.1093/comnet/cnad035","DOIUrl":"https://doi.org/10.1093/comnet/cnad035","url":null,"abstract":"Abstract We consider several families of network centrality measures induced by graph kernels, which include some well-known measures and many new ones. The Self-consistency and Bridge axioms, which appeared earlier in the literature, are closely related to certain kernels and one of the families. We obtain a necessary and sufficient condition for Self-consistency, a sufficient condition for the Bridge axiom, indicate specific measures that satisfy these axioms and show that under some additional conditions they are incompatible. PageRank centrality applied to undirected networks violates most conditions under study and has a property that according to some authors is ‘hard to imagine’ for a centrality measure. We explain this phenomenon. Adopting the Self-consistency or Bridge axiom leads to a drastic reduction in survey time in the culling method designed to select the most appropriate centrality measures.","PeriodicalId":15442,"journal":{"name":"Journal of complex networks","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135364437","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Shenwen Chen, Yisen Wang, Ziquan Liu, Wenbo Du, Lei Zheng, Runran Liu
Abstract Scientific collaboration is an essential aspect of the educational field, offering significant reference value in resource sharing and policy making. With the increasing diversity and inter-disciplinary nature of educational research, understanding scientific collaboration within and between various subfields is crucial for its development. This article employs topic modelling to extract educational research topics from publication metadata obtained from 265 scientific journals spanning the period from 2000 to 2021. We construct a multilayer co-authorship network whose layers represent the scientific collaboration in different subfields. The topological properties of the layers are compared, highlighting the differences and common features of scientific collaboration between hot and cold topics, with the main difference being the existence of a significant largest connected component. Further, the cross-layer cooperation behaviour is investigated by studying the structural measures of the multilayer network and reveals authors’ inclination to collaborate with familiar individuals in familiar subfields. Moreover, the relationships between the authors’ features on the network topology and their H-index are investigated. The results emphasize the significance of establishing a clear research direction to enhance the academic reputation of authors, as well as the importance of cross-layer collaboration for expanding their research groups. Finally, based on the above results, we propose a multilayer network generation model of scientific collaboration and verify its validity.
{"title":"Analysing educational scientific collaboration through multilayer networks: patterns, impact and network generation model","authors":"Shenwen Chen, Yisen Wang, Ziquan Liu, Wenbo Du, Lei Zheng, Runran Liu","doi":"10.1093/comnet/cnad033","DOIUrl":"https://doi.org/10.1093/comnet/cnad033","url":null,"abstract":"Abstract Scientific collaboration is an essential aspect of the educational field, offering significant reference value in resource sharing and policy making. With the increasing diversity and inter-disciplinary nature of educational research, understanding scientific collaboration within and between various subfields is crucial for its development. This article employs topic modelling to extract educational research topics from publication metadata obtained from 265 scientific journals spanning the period from 2000 to 2021. We construct a multilayer co-authorship network whose layers represent the scientific collaboration in different subfields. The topological properties of the layers are compared, highlighting the differences and common features of scientific collaboration between hot and cold topics, with the main difference being the existence of a significant largest connected component. Further, the cross-layer cooperation behaviour is investigated by studying the structural measures of the multilayer network and reveals authors’ inclination to collaborate with familiar individuals in familiar subfields. Moreover, the relationships between the authors’ features on the network topology and their H-index are investigated. The results emphasize the significance of establishing a clear research direction to enhance the academic reputation of authors, as well as the importance of cross-layer collaboration for expanding their research groups. Finally, based on the above results, we propose a multilayer network generation model of scientific collaboration and verify its validity.","PeriodicalId":15442,"journal":{"name":"Journal of complex networks","volume":"132 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135363082","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract Temporal networks are known as the most important tools for representing and storing dynamic systems. This type of network accurately demonstrates all the dynamic changes that occur in a dynamic system. In different applications of dynamic systems, different representation of network models has been used to represent temporal networks. In the last decade, controllability in dynamic systems has become one of the most important challenges in this field. Controllability means the transfer of the network from an initial state to a desired final state in a certain period of time. The most common representation of network model used in control processes is the layered model. But this model has a high overhead, and on the other hand, it slows down the network control processes. In this article, we have proposed a new model for storing and representing temporal networks, which uses a tree structure to save all dynamics of network. Considering that in the proposed model only essential network control information is stored, this model has a very low data overhead compared to the layered model, and this makes the control processes run at a higher speed.
{"title":"Novel network representation model for improving controllability processes on temporal networks","authors":"Yan Liu, Jianhang Zeng, Yue Xu","doi":"10.1093/comnet/cnad036","DOIUrl":"https://doi.org/10.1093/comnet/cnad036","url":null,"abstract":"Abstract Temporal networks are known as the most important tools for representing and storing dynamic systems. This type of network accurately demonstrates all the dynamic changes that occur in a dynamic system. In different applications of dynamic systems, different representation of network models has been used to represent temporal networks. In the last decade, controllability in dynamic systems has become one of the most important challenges in this field. Controllability means the transfer of the network from an initial state to a desired final state in a certain period of time. The most common representation of network model used in control processes is the layered model. But this model has a high overhead, and on the other hand, it slows down the network control processes. In this article, we have proposed a new model for storing and representing temporal networks, which uses a tree structure to save all dynamics of network. Considering that in the proposed model only essential network control information is stored, this model has a very low data overhead compared to the layered model, and this makes the control processes run at a higher speed.","PeriodicalId":15442,"journal":{"name":"Journal of complex networks","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135364422","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Correction to: Analysis of quantile graphs in EGC data from elderly and young individuals using machine learning and deep learning","authors":"","doi":"10.1093/comnet/cnad041","DOIUrl":"https://doi.org/10.1093/comnet/cnad041","url":null,"abstract":"","PeriodicalId":15442,"journal":{"name":"Journal of complex networks","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135368983","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-07-01DOI: 10.48550/arXiv.2307.12461
Tong Mao, Ding-Xuan Zhou
Neural networks activated by the rectified linear unit (ReLU) play a central role in the recent development of deep learning. The topic of approximating functions from H"older spaces by these networks is crucial for understanding the efficiency of the induced learning algorithms. Although the topic has been well investigated in the setting of deep neural networks with many layers of hidden neurons, it is still open for shallow networks having only one hidden layer. In this paper, we provide rates of uniform approximation by these networks. We show that ReLU shallow neural networks with $m$ hidden neurons can uniformly approximate functions from the H"older space $W_infty^r([-1, 1]^d)$ with rates $O((log m)^{frac{1}{2} +d}m^{-frac{r}{d}frac{d+2}{d+4}})$ when $r
{"title":"Rates of Approximation by ReLU Shallow Neural Networks","authors":"Tong Mao, Ding-Xuan Zhou","doi":"10.48550/arXiv.2307.12461","DOIUrl":"https://doi.org/10.48550/arXiv.2307.12461","url":null,"abstract":"Neural networks activated by the rectified linear unit (ReLU) play a central role in the recent development of deep learning. The topic of approximating functions from H\"older spaces by these networks is crucial for understanding the efficiency of the induced learning algorithms. Although the topic has been well investigated in the setting of deep neural networks with many layers of hidden neurons, it is still open for shallow networks having only one hidden layer. In this paper, we provide rates of uniform approximation by these networks. We show that ReLU shallow neural networks with $m$ hidden neurons can uniformly approximate functions from the H\"older space $W_infty^r([-1, 1]^d)$ with rates $O((log m)^{frac{1}{2} +d}m^{-frac{r}{d}frac{d+2}{d+4}})$ when $r<d/2 +2$. Such rates are very close to the optimal one $O(m^{-frac{r}{d}})$ in the sense that $frac{d+2}{d+4}$ is close to $1$, when the dimension $d$ is large.","PeriodicalId":15442,"journal":{"name":"Journal of complex networks","volume":"8 3 1","pages":"101784"},"PeriodicalIF":2.1,"publicationDate":"2023-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80563859","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Zhao-Yan Li, Jing Zhang, Guozhong Zheng, Li Chen, Jiqiang Zhang, Weiran Cai
While network science has become an indispensable tool for studying complex systems, the conventional use of pairwise links often shows limitations in describing high-order interactions properly. Hypergraphs, where each edge can connect more than two nodes, have thus become a new paradigm in network science. Yet, we are still in lack of models linking network growth and hyperedge expansion, both of which are commonly observable in the real world. Here, we propose a generative hypergraph model by employing the preferential attachment mechanism in both nodes and hyperedge formation. The model can produce bi-heterogeneity, exhibiting scale-free distributions in both hyperdegree and hyperedge size. We provide a mean-field treatment that gives the expression of the two scaling exponents, which agree with the numerical simulations. Our model may help to understand the networked systems showing both types of heterogeneity and facilitate the study of complex dynamics thereon.
{"title":"A generative hypergraph model for double heterogeneity","authors":"Zhao-Yan Li, Jing Zhang, Guozhong Zheng, Li Chen, Jiqiang Zhang, Weiran Cai","doi":"10.1093/comnet/cnad048","DOIUrl":"https://doi.org/10.1093/comnet/cnad048","url":null,"abstract":"While network science has become an indispensable tool for studying complex systems, the conventional use of pairwise links often shows limitations in describing high-order interactions properly. Hypergraphs, where each edge can connect more than two nodes, have thus become a new paradigm in network science. Yet, we are still in lack of models linking network growth and hyperedge expansion, both of which are commonly observable in the real world. Here, we propose a generative hypergraph model by employing the preferential attachment mechanism in both nodes and hyperedge formation. The model can produce bi-heterogeneity, exhibiting scale-free distributions in both hyperdegree and hyperedge size. We provide a mean-field treatment that gives the expression of the two scaling exponents, which agree with the numerical simulations. Our model may help to understand the networked systems showing both types of heterogeneity and facilitate the study of complex dynamics thereon.","PeriodicalId":15442,"journal":{"name":"Journal of complex networks","volume":"25 1","pages":""},"PeriodicalIF":2.1,"publicationDate":"2023-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139368856","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Complex networks are a powerful way to reason about systems with non-trivial patterns of interaction. The increased attention in this research area is accelerated by the increasing availability of complex network data sets, with data often being reused as secondary data sources. Typically, multiple data sources are combined to create a larger, fuller picture of these complex networks and in doing so scientists have to make sometimes subjective decisions about how these sources should be integrated. These seemingly trivial decisions can sometimes have significant impact on both the resultant integrated networks and any downstream network models executed on them. We highlight the importance of this impact in online social networks and dark networks, two use-cases where data are regularly combined from multiple sources due to challenges in measurement or overlap of networks. We present a method for systematically testing how different, realistic data integration approaches can alter both the networks themselves and network models run on them, as well as an associated Python package (NIDMod) that implements this method. A number of experiments show the effectiveness of our method in identifying the impact of different data integration setups on network diffusion models.
{"title":"An approach for analysing the impact of data integration on complex network diffusion models","authors":"J. Nevin, Paul Groth, M. Lees","doi":"10.1093/comnet/cnad025","DOIUrl":"https://doi.org/10.1093/comnet/cnad025","url":null,"abstract":"\u0000 Complex networks are a powerful way to reason about systems with non-trivial patterns of interaction. The increased attention in this research area is accelerated by the increasing availability of complex network data sets, with data often being reused as secondary data sources. Typically, multiple data sources are combined to create a larger, fuller picture of these complex networks and in doing so scientists have to make sometimes subjective decisions about how these sources should be integrated. These seemingly trivial decisions can sometimes have significant impact on both the resultant integrated networks and any downstream network models executed on them. We highlight the importance of this impact in online social networks and dark networks, two use-cases where data are regularly combined from multiple sources due to challenges in measurement or overlap of networks. We present a method for systematically testing how different, realistic data integration approaches can alter both the networks themselves and network models run on them, as well as an associated Python package (NIDMod) that implements this method. A number of experiments show the effectiveness of our method in identifying the impact of different data integration setups on network diffusion models.","PeriodicalId":15442,"journal":{"name":"Journal of complex networks","volume":"11 1","pages":""},"PeriodicalIF":2.1,"publicationDate":"2023-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74704046","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Complex networks can often exhibit a high degree of bipartivity. There are many well-known ways for testing this, and in this article, we give a systematic analysis of characterizations based on the spectra of the adjacency matrix and various graph Laplacians. We show that measures based on these characterizations can be drastically different results and leads us to distinguish between local and global loss of bipartivity. We test several methods for finding approximate bipartitions based on analysing eigenvectors and show that several alternatives seem to work well (and can work better than more complex methods) when augmented with local improvement.
{"title":"Spectral techniques for measuring bipartivity and producing partitions","authors":"Azhar Aleidan, P. Knight","doi":"10.1093/comnet/cnad026","DOIUrl":"https://doi.org/10.1093/comnet/cnad026","url":null,"abstract":"\u0000 Complex networks can often exhibit a high degree of bipartivity. There are many well-known ways for testing this, and in this article, we give a systematic analysis of characterizations based on the spectra of the adjacency matrix and various graph Laplacians. We show that measures based on these characterizations can be drastically different results and leads us to distinguish between local and global loss of bipartivity. We test several methods for finding approximate bipartitions based on analysing eigenvectors and show that several alternatives seem to work well (and can work better than more complex methods) when augmented with local improvement.","PeriodicalId":15442,"journal":{"name":"Journal of complex networks","volume":"16 1","pages":""},"PeriodicalIF":2.1,"publicationDate":"2023-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84564844","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Abstract Mean field theory models of percolation on networks provide analytic estimates of network robustness under node or edge removal. We introduce a new mean field theory model based on generating functions that includes information about the tree-likeness of each node’s local neighbourhood. We show that our new model outperforms all other generating function models in prediction accuracy when testing their estimates on a wide range of real-world network data. We compare the new model’s performance against the recently introduced message-passing models and provide evidence that the standard version is also outperformed, while the ‘loopy’ version is only outperformed on a targeted attack strategy. As we show, however, the computational complexity of our model implementation is much lower than that of message-passing algorithms. We provide evidence that all discussed models are poor in predicting networks with highly modular structure with dispersed modules, which are also characterized by high mixing times, identifying this as a general limitation of percolation prediction models.
{"title":"Improving mean-field network percolation models with neighbourhood information","authors":"Chris Jones, Karoline Wiesner","doi":"10.1093/comnet/cnad029","DOIUrl":"https://doi.org/10.1093/comnet/cnad029","url":null,"abstract":"Abstract Mean field theory models of percolation on networks provide analytic estimates of network robustness under node or edge removal. We introduce a new mean field theory model based on generating functions that includes information about the tree-likeness of each node’s local neighbourhood. We show that our new model outperforms all other generating function models in prediction accuracy when testing their estimates on a wide range of real-world network data. We compare the new model’s performance against the recently introduced message-passing models and provide evidence that the standard version is also outperformed, while the ‘loopy’ version is only outperformed on a targeted attack strategy. As we show, however, the computational complexity of our model implementation is much lower than that of message-passing algorithms. We provide evidence that all discussed models are poor in predicting networks with highly modular structure with dispersed modules, which are also characterized by high mixing times, identifying this as a general limitation of percolation prediction models.","PeriodicalId":15442,"journal":{"name":"Journal of complex networks","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136085163","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：