Characterizing and predicting community members from evolutionary and heterogeneous networks

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.

The Taihu Lake plain is a highly urbanized region in China with many water-related environmental problems. Although point-source pollution has been effectively controlled by government legislation, urban surface runoff pollution is still a major issue. Different types of urban communities were selected for rainfall runoff experiments. According to the monitoring data of rainfall events, multiple methods were used to analyze the characteristics of surface runoff pollution and estimate the pollution load for different types of communities. The results indicated that surface runoff from urban communities reduced the river water quality. Certain degrees of the 'first flush' effect occurred in different types of urban communities. The surface runoff pollution in the commercial residential community was weaker than that in commercial and private residential communities; however, the first flush occurred more frequently in the commercial residential community. Holding back 30% of the surface runoff could effectively improve the runoff water quality in commercial and private residential communities as well as the commercial residential community with restaurants. In the commercial residential community, 25% of surface runoff should be held to improve runoff water quality effectively. The loads of pollutants, especially nitrogen and phosphorus, in urban communities in the Taihu Lake basin were higher than those in other regions in China. This research can assist with the reduction of surface runoff pollution in highly urbanized communities.

In February of 2016, China formally promulgated and pointed out that the new communities should promote the block system. This new policy sparked heated debate. To demonstrate the rationality of this decision, the article studies on the problem that the influence on surrounding traffic causing by the opening of different types of community. The result shows that there is a big difference of the influence on surrounding traffic between opening different types of communities. So the decision of opening communities should fully consider the particularity of communities and adjust measures according to local conditions.

Many complex systems derived from nature and society consist of multiple types of entities and heterogeneous interactions, which can be effectively modeled as heterogeneous information network (HIN). Structural analysis of heterogeneous networks is of great significance by leveraging the rich semantic information of objects and links in the heterogeneous networks. And, clustering heterogeneous networks aims to group vertices into classes, which sheds light on revealing the structure–function relations of the underlying systems. The current algorithms independently perform the feature extraction and clustering, which are criticized for not fully characterizing the structure of clusters. In this study, we propose a learning model by joint <underline>G</underline>raph <underline>E</underline>mbedding and <underline>N</underline>onnegative <underline>M</underline>atrix <underline>F</underline>actorization (aka GEjNMF ), where feature extraction and clustering are simultaneously learned by exploiting the graph embedding and latent structure of networks. We formulate the objective function of GEjNMF and transform the heterogeneous network clustering problem into a constrained optimization problem, which is effectively solved by l 0 -norm optimization. The advantage of GEjNMF is that features are selected under the guidance of clustering, which improves the performance and saves the running time of algorithms at the same time. The experimental results on three benchmark heterogeneous networks demonstrate that GEjNMF achieves the best performance with the least running time compared with the best state-of-the-art methods. Furthermore, the proposed algorithm is robust across heterogeneous networks from various fields. The proposed model and method provide an effective alternative for heterogeneous network clustering.

In the heterogeneous networks, a complex signal-interference environment is formed, characterized by a intra-system interference high level, which significantly reduces the heterogeneous networks throughput. One of the ways to increase such networks throughput by effectively reducing the intra-system interference influence is to use the integrated optimization method of these networks. However, its application is hampered by the complexity of implementing the collective dynamic routing method that underlies it. Therefore, modification of the collective dynamic routing method in relation to large heterogeneous networks is relevant. As a result of the application analysis of the collective dynamic routing method for optimizing the data flows distribution in heterogeneous networks, it is shown that the main limitation of this method applicability is the computational complexity of finding the optimal data transmission route. For this value the corresponding expressions are obtained. A modification of the collective dynamic routing method has been developed, in which the influence of intra-system interference is taken into account at the subnet segment level, and data distribution between subnets is carried out at the zone level. Using the obtained expressions, it is shown that the computational complexity of implementing the collective dynamic routing method in this case depends only on the number of users in the zone, which ensures the scalability of such a solution. Therefore, the use of the developed modification of the collective dynamic routing method will make it possible to use it to increase the throughput of large heterogeneous networks.

The goal of this study was to assess whether different types of communities (ie, urban, suburban, small town, and rural) vary in the alcohol enforcement activities they conduct. We conducted a cross-sectional study using data from a national survey of local law enforcement agencies. The survey assessed enforcement of a range of alcohol policies at 1,082 law enforcement agencies. U.S. Census data were used to categorize agencies based on community type. Agencies in urban areas conducted more enforcement activities than agencies in other community types. Urban agencies were more likely than rural agencies to conduct underage compliance checks (prevalence ratio [PR]: 0.42; CI: 0.34-0.53), saturation patrols (PR: 0.80; CI: 0.67-0.95), sobriety checkpoints (PR: 0.68; CI: 0.53-0.86), and enforcement aimed at illegal sales to intoxicated patrons (PR: 0.59; CI: 0.42-0.81). Urban agencies were also more likely than small town agencies to do compliance checks (PR: 0.66; CI: 0.56-0.79) and sobriety checkpoints (PR: 0.75; CI: 0.61-0.91), and they were more likely than suburban agencies to do compliance checks (PR: 0.67; CI: 0.57-0.78) and enforcement actions around the sale of alcohol to intoxicated patrons (PR: 0.64; CI: 0.45-0.90), provision of alcohol to minors (PR: 0.77; CI: 0.65-0.92), and consumption by minors (PR: 0.90; CI: 0.82-0.99). Enforcement of alcohol laws differs by community type. Future research is needed to identify mechanisms to increase enforcement by agencies in different types of communities.

In the face of the sudden outbreak of COVID-19, the community, as the interface between the state and the population, has played a large role in stemming the impact of COVID-19, and community resilience has become a key part of community governance. However, the negative response of some residents is an important factor that hinders the community's emergency response capabilities. Therefore, starting from empirical research, this chapter uses factor analysis, multiple linear regression analysis, and cluster analysis to explore the differences in community resilience and residents’ risk response methods in different types of communities, as well as the impact mechanism of community resilience on residents’ risk response methods. The study found that first, community resilience has a significant positive impact on positive risk response, among which governance performance has a more significant impact; second, community resilience has a negative correlation with negative coping styles, in which governance performance has a more significant impact; third, there are obvious differences in the level of resilience between different types of communities, with urban communities being the best, mixed communities second, and transformation communities the last. After the shock of COVID-19, the formation of an individualised society and an all-powerful governmental governance system has led to an increasingly prominent role for communities in social management. Community resilience continues to influence the way residents respond and becomes a key factor in controlling COVID-19.

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.

Traditional learning-to-rank problem mainly focuses on one single type of objects. However, with the rapid growth of the Web 2.0, ranking over multiple interrelated and heterogeneous objects becomes a common situation, e.g., the heterogeneous academic network. In this scenario, one may have much training data for some type of objects (e.g. conferences) while only very few for the interested types of objects (e.g. authors). Thus, the two important questions are: (1) Given a networked data set, how could one borrow supervision from other types of objects in order to build an accurate ranking model for the interested objects with insufficient supervision? (2) If there are links between different objects, how can we exploit their relationships for improved ranking performance? In this work, we first propose a regularized framework called HCDRank to simultaneously minimize two loss functions related to these two domains. Then, we extend the approach by exploiting the link information between heterogeneous objects. We conduct a theoretical analysis to the proposed approach and derive its generalization bound to demonstrate how the two related domains could help each other in learning ranking functions. Experimental results on three different genres of data sets demonstrate the effectiveness of the proposed approaches.

A relation-aware heterogeneous graph convolutional network for relationship prediction

A similarity-based community detection method with multiple prototype representation

Aiming at the transmission of heterogeneous data in heterogeneous networks, a topology optimization algorithm of heterogeneous wireless sensor networks based on local tree reconstruction is proposed, which can achieve better data transmission in the heterogeneous networks. First, the algorithm divides the nodes of the network into different layers by their hops and chooses different numbers of nodes as relay nodes in different layers. Second, the nodes are set with different initial energies in different layers. Because the packets of different nodes have different sizes, we adopt the corresponding data aggregation coefficients according to the actual data requirements of the network in data transmission. Finally, the lifetime of the network is prolonged by real-time updating of the topology of the tree during the data transmission. The simulations indicate that after the aforementioned three steps, the proposed algorithm prolongs the lifetime of the heterogeneous networks and improves the nodes utilization effectively.

Entrepreneurs can benefit from the communities they build. Therefore, many entrepreneurs create online communities that allow self-selected stakeholders, such as customers, crowd investors, or enthusiasts, to interact with the venture and other like-minded individuals. However, research on how entrepreneurs can successfully engage community members and grow such online communities is only slowly emerging. In particular, it is unclear if, how much, and which content entrepreneurs should contribute to foster engagement in different types of communities and which role these community types play in the community’s overall growth. Based on a longitudinal case study in the video game industry, we first theorize and show that—depending on the community type—both too much and too little entrepreneur-provided content fails to leverage community engagement potential and that different communities require more or less diverging content. We then theorize and show that community growth is largely driven by engagement in open communities, such as those hosted on social media. We outline the implications this has for entrepreneurs, our understanding of online communities, and entrepreneurial communities more generally.

Recent studies have shown that sectors defined by the denominational status and by the era in which schools were founded account for some of the variation in the effectiveness of schooling in Scotland. Other components of the total variation among schools can be attributed to variation in the social composition of schools, and to variation in the process and teaching characteristics of schools. The present paper complements these studies by considering other sources of intra‐system variation. In particular, it investigates variation in pupil‐attainment among different types of communities, and among educational authorities (EAs). It focuses on analyses of pupil attainment in fourth year Ordinary grade English and arithmetic, and on an overall attainment measured across subjects in the Scottish Certificate of Education (SCE) examinations. The effectiveness of schools is then assessed as the variation in pupil‐level attainment that is associated with schools, both before and after adjustment for pupil social background characteristics. The recently developed statistical technique of multilevel modelling is employed to model variation in the effectiveness of schools. This methodology has been developed in order to overcome some of the statistical problems that have bedevilled studies of the performance of schools. Its two main strengths are that it explicitly takes account of the multilevel nature of the information embodied in pupils’ examination results, and that it allows each school to vary in terms of its performance characteristics to a considerably greater extent than previous aggregate‐level methods. In analysing the variation among schools it is important to take account of the selective nature of school intakes, otherwise the apparent differences among schools may reflect no more than the variation in the quality of their intakes. The data available to the present study contain information on pupils’ social background characteristics, but not on their prior cognitive attainments. The selectivity of school intakes derives principally from two sources: variation in the residential pattern of socioeconomic groups, and voluntary choice of school. The data analysed here relate to pupils who entered their secondary schooling in the mid‐1970s before the Parent's Charter was enacted; thus voluntary choice of secondary schooling was constrained mainly to the independent sector. In addition to formal selection in relation to entry to the independent sector, the schools in the cities and large burghs were subject to informal selection through the creaming of their more able and socially advantaged pupils by the independent sector. The study attempts to control for some of the effects of these formal and informal selection processes by stratifying the analysis by the degree of selectivity of schools. Variation among schools in five types of community are investigated: cities, large burghs and towns, small burghs and towns, new towns, and rural areas. The cities account for about one‐third of all pupils in the system, while the two burgh sectors combined account for slightly more than half of all pupils in the system. Although the rural areas account for 10 per cent of pupils, they account for 17 per cent of the schools in the system, because of the smaller size of rural schools. The lowest performing community sector is the city sector, but it also contains the widest range of variation among its schools. The analysis of variation among EAs reveals that the median effectiveness of schools varies to a much smaller extent than the variation among schools, but that the assessment of EA performance is sensitive to the population of schools included in the analysis.

Phenotype prediction aims to infer the traits of living organisms based on genomics data, which has important applications in biology such as cancer subtype diagnosis and crop breedings. Traditional phenotype prediction approaches only learn the low-dimensional representation of samples, but ignore the interaction between biomolecules and cannot use the topology structure of heterogeneous molecular networks. Furthermore, most of them lack interpretability and do not effectively identify key biomolecules associated with phenotypes. In this paper, we propose a heterogeneous network embedding based solution (PhenoHNE) to predict phenotype by fusing topology information of molecules. PhenoHNE firstly utilizes variational graph autoencode (VGAE) to obtain the embedding representation of molecules in the heterogeneous network. Secondly, PhenoHNE adopts multilayer perceptron (MLP) with attention mechanism to learn sample representation. Finally, it fuses molecular embedding representation and sample representation to predict the phenotype of samples. In this way, the genetics information of heterogeneous molecular network can further guide the feature learning of samples and improve the prediction performance. Experimental results on human and maize datasets confirm that PhenoHNE outperforms competitive methods by a large margin under different evaluation protocols, and it also can effectively identify the key molecules associated with phenotypes of interests.