MULTILAYER STOCHASTIC BLOCK MODELS FOR COMMUNITY DETECTION IN HETEROGENEOUS NETWORKS

Dynamic heterogeneous graph representation learning with neighborhood type modeling

A relation-aware heterogeneous graph convolutional network for relationship prediction

Finding different types of communities has become a research hot spot in network science. Plenty of the real-world systems containing different types of objects and relationships can be perfectly described as the heterogeneous networks. However, most of the current research on community detection is applied for the homogeneous networks, while there is no effective function to quantify the quality of the community structure in heterogeneous networks. In this paper, we first propose the null model with the same heterogeneous node degree distribution of the original heterogeneous networks. The probability of there being an edge between two nodes is given to build the modularity function of the heterogeneous networks. Based on our modularity function, a fast algorithm of community detection is proposed for the large scale heterogeneous networks. We use the algorithm to detect the communities in the real-world twitter event networks. The experimental results show that our method perform better than other exciting algorithms and demonstrate that the modularity function of the heterogeneous networks is an effective parameter that can be used to quantify the quality of the community structure in heterogeneous networks.

Complex networks have in generally communities. These communities are very important. Network’s communities represent sets of nodes, which are very connected. In this research, we developed a new method to find the community structure in networks. Our method is based on flower pollination algorithm (FPA) witch is used in the splitting process. The splitting of networks in our method maximizes a function of quality called modularity. We provide a general framework for implementing our new method to find community structure in networks. We present the effectiveness of our method by comparison with some known methods on computer-generated and real-world networks.

Document-level relation extraction with multi-layer heterogeneous graph attention network

A heterogeneous correlation network represents relationships (edges) among source-typed and attribute-typed objects (nodes). It can be used to model an academic collaboration network, describing connections among authors and published papers. To date, there has been little research into mining communities in heterogeneous networks. The objective of our research is to discover overlapping communities that include all node and edge types in a heterogeneous correlation network. We describe an algorithm, OHC, that detects overlapping communities in heterogeneous correlation networks. Inspired by a homogeneous community scoring function, Triangle Participation Ratio (TPR), OHC finds target heterogeneous communities then expands them recursively with triangle-forming nodes. Experiments on different real world networks demonstrate that OHC identifies heterogeneous communities that are tightly connected internally according to two traditional scoring functions. Additionally, analyzing the top ranking heterogeneous communities in a case study, we evaluate the results qualitatively.

Quantifying the importance of specific types of higher-order structures in complex networks

Topology analysis of Lanzhou public transport network based on double-layer complex network theory

Homophily, the tendency of similar nodes to connect, is a fundamental phenomenon in network science and a critical factor in the performance of graph neural networks (GNNs). While existing studies primarily explore homophily in homogeneous graphs, where nodes share the same type, real-world networks are often more accurately modeled as heterogeneous graphs (HGs) with diverse node types and intricate cross-type interactions. This structural diversity complicates the analysis of homophily, as traditional homophily metrics fail to account for distinct label spaces across node types. To address this limitation, we introduce the Cross-Type Homophily Ratio (CHR), a novel metric that quantifies homophily based on the similarity of target information across different node types. Additionally, we propose Cross-Type Homophily-guided Graph Editing (CTHGE), a novel method for improving heterogeneous graph neural networks (HGNNs) performance by optimizing cross-type connectivity using Cross-Type Homophily Ratio. Extensive experiments on five HG datasets with nine HGNNs validate the effectiveness of CTHGE, which delivers a maximum relative performance improvement of over 25% for HGNNs on node classification tasks, offering a fresh perspective on cross-type homophily in HGs learning.

When studying dynamic networks, it is often of interest to understand how the community structure of the network changes. However, before studying the community structure of dynamic social networks, one must first collect appropriate network data. In this paper we present a network sampling technique to crawl the community structure of dynamic networks when there is a limitation on the number of nodes that can be queried. The process begins by obtaining a sample for the first time step. In subsequent time steps, the crawling process is guided by community structure discoveries made in the past. Experiments conducted on the proposed approach and certain baseline techniques reveal the proposed approach has at least 35% performance increase in cases when the total query budget is fixed over the entire period and at least 8% increase in cases when the query budget is fixed per time step.

Identifying and evaluating community structure in complex networks

In reality many complex networks present modules or community structures obviously. Modularity is a benefit function used in quantifying the quality of a division of a network into communities. And it usually can be used as a basis for optimization methods of detecting community structure in networks. But the most popular modularity which is proposed by M. E. J. Newman and M. Girvan has the resolution limit in community detection. Multi-resolution modularity cannot overcome the misclassifications caused by merging and splitting the communities either. In this paper, we propose a multi-resolution density modularity based on the network density. The proposed function is tested on the artificial networks. Computational results show that it can reduce the rate of misclassification considerably. And the systematicness of the community structures can be demonstrated by the multi-resolution density modularity.

Identifying meaningful cancer driver genes in a cohort of tumors is a challenging task in cancer genomics. Although existing studies have identified known cancer drivers, most of them focus on detecting coding drivers with mutations. It is acknowledged that non-coding drivers can regulate driver mutations to promote cancer growth. In this work, we propose a novel node importance-based network analysis (NIBNA) framework to detect coding and non-coding cancer drivers. We hypothesize that cancer drivers are crucial to the formation of community structures in cancer network, and removing them from the network greatly perturbs the network structure thereby critically affecting the functioning of the network. NIBNA detects cancer drivers using a three-step process: first, a condition-specific network is built by incorporating gene expression data and gene networks; second, the community structures in the network are estimated; and third, a centrality-based metric is applied to compute node importance. We apply NIBNA to the BRCA dataset, and it outperforms existing state-of-art methods in detecting coding cancer drivers. NIBNA also predicts 265 miRNA drivers, and majority of these drivers have been validated in literature. Further we apply NIBNA to detect cancer subtype-specific drivers, and several predicted drivers have been validated to be associated with cancer subtypes. Lastly, we evaluate NIBNA's performance in detecting epithelial-mesenchymal transition drivers, and we confirmed 8 coding and 13 miRNA drivers in the list of known genes. The source code can be accessed at https://github.com/mandarsc/NIBNA. Supplementary data are available at Bioinformatics online.

Networks have become a widely used tool for modeling complex systems in many different fields. This approach is extremely useful for representing interactions among genes, social relationships, Internet communications or correlations of prices within a stock market, to name just a few examples. By analyzing the structure of these networks and understanding how their different elements interact, we could improve our knowledge of the whole system. Usually, nodes that compose these networks tend to create tightly knit groups. This property, of high interest in many scientific fields, is called community structure and improving its detection and characterization is what this thesis is all about. The first objective of this work is the generation of efficient methods able to characterize the communities of a network and to understand its structure. Second, we will try to create a set of tests where such methods can be studied. Finally, we will suggest a statistical measure in order to be able to properly assess the quality of the community structure of a network. To accomplish these objectives, first, we generate a set of algorithms that can transform a network into a hierarchical tree and, from there, to determine their most relevant communities. Furthermore, we have developed a new type of benchmarks for effectively testing these and other community detection algorithms. Finally, and as the most important contribution of this work, it is shown that the community structure of a network can be accurately evaluated using a hypergeometric distribution-based index. Thus, the maximization of this measure, called Surprise, appears as the best proposed strategy for detecting the optimal partition into communities of a network. Surprise exhibits an excellent behavior in all networks analyzed, qualitatively outperforming any previous method. Thus, it appears as the best measure proposed to this end and the data suggests that it could be an optimal strategy to determine the quality of the community structure of complex networks.

Social network analysis has typically concerned analysis of one type of tie connecting nodes of the same type. It has however been recognised that people are connected through multiple types of ties and that people in addition are affiliated with multiple types of non-people nodes. Exponential random graph models (ERGM) is a family of statistical models for social networks that at this point allows for a number of different types of network data, including one-mode networks, bipartite networks, multiplex data, as well as multilevel network data. Multilevel networks have been proposed as a joint representation of associations between multiple types of entities or nodes, such as people and organization, where two types of nodes gives rise to three distinct types of ties. The typical roster data collection method may be impractical or infeasible when the node sets are hard to detect or define or because of the cognitive demands on respondents. Multilevel multilayered networks allow us to consider a multitude of different sources of data and to sample on different types of nodes and relations. We consider modelling multilevel multilayered networks using exponential random graph models and extend a recently developed Bayesian data-augmentation scheme to allow for partially missing data. We illustrate the proposed inference procedures for the case of multilevel snowball sampling and sampling with error based on the Noordin Top network.