Network Science Research Research Articles

Mining heterogeneous information networks

Most objects and data in the real world are of multiple types, interconnected, forming complex, heterogeneous but often semi-structured information networks. However, most network science researchers are focused on homogeneous networks, without distinguishing different types of objects and links in the networks. We view interconnected, multityped data, including the typical relational database data, as heterogeneous information networks, study how to leverage the rich semantic meaning of structural types of objects and links in the networks, and develop a structural analysis approach on mining semi-structured, multi-typed heterogeneous information networks. In this article, we summarize a set of methodologies that can effectively and efficiently mine useful knowledge from such information networks, and point out some promising research directions.

Throughflow centrality is a global indicator of the functional importance of species in ecosystems

To better understand and manage complex systems like ecosystems it is critical to know the relative contribution of the system components to the system function. Ecologists and social scientists have described a diversity of ways that individuals can be important; This paper makes two key contributions to this research area. First, it shows that throughflow (Tj), the total energy or matter entering or exiting a system component, is a global indicator of the relative contribution of the component to the whole system activity. It is global because it includes the direct and indirect exchanges among community members. Further, throughflow is a special case of Hubbell status or centrality as defined in social science. This recognition effectively joins the concepts, enabling ecologists to use and build on the broader centrality research in network science. Second, I characterize the distribution of throughflow in 45 empirically-based trophic ecosystem models. Consistent with theoretical expectations, this analysis shows that a small fraction of the system components are responsible for the majority of the system activity. In 73% of the ecosystem models, 20% or less of the nodes generate 80% or more of the total system throughflow. Four or fewer nodes are required to account for 50% of the total system activity and are thus defined as community dominants. 121 of the 130 dominant nodes in the 45 ecosystem models could be classified as primary producers, dead organic matter, or bacteria. Thus, throughflow centrality indicates the rank power of the ecosystems components and shows the concentration of power in the primary production and decomposition cycle. Although these results are specific to ecosystems, these techniques build on flow analysis based on economic input–output analysis. Therefore these results should be useful for ecosystem ecology, industrial ecology, the study of urban metabolism, as well as other domains using input–output analysis.

An Extensible Digital Library Service to Support Network Science

Network science research aims to understand the underlying properties of complex networks. Large-scale modeling and simulation is the core of network science research. Existing systems take a long time to run large network science experiments with high performance computing resources. Scientific data management systems currently lack the performance efficiency needed to support this type of computation and data-intensive research. Memoization provides the ability to index, archive, and reuse frequently requested and expensive to re-compute datasets. In this paper, a domain independent memoization service to increase the computational execution process performance within cyberinfrastructure-based systems is described. The service is formulated as an extensible memoization framework for the computational and simulation network science domains that is built on top of well-defined metadata objects. We present extensible concepts, discuss the proposed algorithm and framework architecture, and examine the flexible nature of the framework. The framework was utilized as a part of the CINET cyberinfrastructure-based digital library (DL). Our experimental results indicate an increase in the efficiency of the system and recommendation of the service's inclusion in scientific DLs.

The Contributions of Regional Knowledge Networks Researching Environmental Changes in Latin America and Africa: a Synthesis of what they can do and why they can be policy relevant

We provide a synthesis of what regional scientific research networks in less developed regions of the world can do and why they might be relevant for societal decisions and practice. We do so through a focus on three regional science network initiatives that aim to enhance understanding of the multiscalar dynamics of global environmental change (GEC) regionally and globally, namely the Southern Africa Regional Science Initiative (SAFARI 2000), the Large-Scale Biosphere-Atmosphere Experiment in Amazonia (LBA), and the Inter-American Institute for Global Change (IAI). With a view to aiding future efforts at regional research network formation, we assess whether and how these three networks enhanced regional science, and the extent to which they sought and managed to bridge the science-policy gap that challenges GEC science as a whole. Identifying key decisions and attributes bearing on their successes, the analysis attends specifically to how the three networks sought to build capacity, how differences and similarities between them affected their level of autonomy from governments, and how this and other factors influenced their functioning and achievements.

Mining knowledge from interconnected data

Most objects and data in the real world are interconnected, forming complex, heterogeneous but often semi-structured information networks. However, most people consider a database merely as a data repository that supports data storage and retrieval rather than one or a set of heterogeneous information networks that contain rich, inter-related, multi-typed data and information. Most network science researchers only study homogeneous networks, without distinguishing the different types of objects and links in the networks. In this tutorial, we view database and other interconnected data as heterogeneous information networks, and study how to leverage the rich semantic meaning of types of objects and links in the networks. We systematically introduce the technologies that can effectively and efficiently mine useful knowledge from such information networks.

Tutorial on biological networks

AbstractUnderstanding how the functioning of a biological system emerges from the interactions among its components is a long‐standing goal of network science. Fomented by developments in high‐throughput technologies to characterize biomolecules and their interactions, network science has emerged as one of the fastest growing areas in computational and systems biology research. Although the number of research and review articles on different aspects of network science is increasing, updated resources that provide a broad, yet concise, review of this area in the context of systems biology are few. The objective of this article is to provide an overview of the research on biological networks to a general audience, who have some knowledge of biology and statistics, but are not necessarily familiar with this research field. Based on the different aspects of network science research, the article is broadly divided into four sections: (1) network construction, (2) topological analysis, (3) network and data integration, and (4) visualization tools. We specifically focused on the most widely studied types of biological networks, which are, metabolic, gene regulatory, protein–protein interaction, genetic interaction, and signaling networks. In future, with further developments on experimental and computational methods, we expect that the analysis of biological networks will assume a leading role in basic and translational research. © 2012 Wiley Periodicals, Inc.This article is categorized under: Algorithmic Development > Biological Data Mining Application Areas > Data Mining Software Tools Application Areas > Science and Technology

Sectoral collaboration in biomedical research and development

This paper explores the role of sectors in scientific research and development networks by drawing on bibliometric analyses and innovation systems and triple helix literatures. I conducted a bibliometric study of Vancouver Canada's worldwide infection and immunity network and examined network structure through sociograms, social network metrics, as well as relational contingency table and ANOVA network analyses. Universities are the key network sector followed by hospitals and government organisations. The private sector plays a weak role. Most sectors show a preference for collaborating within, as opposed to across, sectors. This trend is most pronounced in hospitals and least pronounced among firms. Hospitals and universities collaborate well above statistical expectations. I discuss the implications of these findings for future science policy and studies of research and development networks.

Ten years of the genomic era in Brazil: Impacts on technological development assessed by scientific production and patent analysis

Over ten years ago Brazil entered the era of genome projects with the launch of its first sequencing effort. Since then, a series of projects have been undertaken in the country, greatly contributing to the formation of scientific research networks, related infrastructure and specialized labor. Today, Brazil’s competence in this area is internationally recognized and it is one of the main countries in the number of genomes sequenced. This article examines the effects of these genome initiatives on Brazilian technological innovation, through a broad bibliometric review, comprising the analysis of academic and patent publications, which revealed that the development of technology has not kept pace with scientific output.

Blasenkarzinom – Update

One of the principal objects of the scientific research network "German Bladder Cancer Network" is to consolidate research activities on bladder cancer. An overview about directions of current projects on this research topic was given at the annual meeting of the German Association of Urology in Düsseldorf from September 22 to 25 September 2010. As representatives of the"German Bladder Cancer Network" we summarize and comment on some of the most interesting projects on bladder cancer presented at this meeting. A special focus will be on current developments in the field of uropathology and on different aspects in preclinical research on bladder cancer.

A Network Perspective of Digital Competition in Online Advertising Industries: A Simulation-Based Approach

Using agent-based simulation experiments, we investigate the outcome of SAs between two smaller online search engine companies in competition with a dominant market leader in settings where an advertiser's decision making is the consequence of a combination of NI (e.g., an individual's willingness to follow others' decisions) and IP. In particular, we focus on a context in which the combined search engine company competes with a market leader holding a larger share of the market than the two runner-up “underdogs” combined. Our results indicate that, with the presence of NI and cascading effects, an alliance with “only” 35%–40% combined market share could compete with a leader whose market share, at the time of an alliance, is 60%–65%. Although important, size alone might be insufficient to build the market as suggested by the “vanilla” network effect theory. Another noteworthy finding is that a nonlinear association exists between NI and an alliance outcome; the combined runner-up companies have the best chance of success when the extent of NI is midrange, rather than on the high or low end of continuum. Contrary to the conventional view, this finding might also stimulate discussions among network science researchers. Furthermore, our results suggest that NI substantially moderates the relationship between the combined market share at the time of an alliance and the likelihood of resulting alliance success.

Behavioral dynamics and influence in networked coloring and consensus

We report on human-subject experiments on the problems of coloring (a social differentiation task) and consensus (a social agreement task) in a networked setting. Both tasks can be viewed as coordination games, and despite their cognitive similarity, we find that within a parameterized family of social networks, network structure elicits opposing behavioral effects in the two problems, with increased long-distance connectivity making consensus easier for subjects and coloring harder. We investigate the influence that subjects have on their network neighbors and the collective outcome, and find that it varies considerably, beyond what can be explained by network position alone. We also find strong correlations between influence and other features of individual subject behavior. In contrast to much of the recent research in network science, which often emphasizes network topology out of the context of any specific problem and places primacy on network position, our findings highlight the potential importance of the details of tasks and individuals in social networks.

Rete-netzwerk-red: analyzing and visualizing scholarly networks using the Network Workbench Tool

The enormous increase in digital scholarly data and computing power com- bined with recent advances in text mining, linguistics, network science, and scientometrics make it possible to scientifically study the structure and evolution of science on a large scale. This paper discusses the challenges of this 'BIG science of science'—also called 'computational scientometrics' research—in terms of data access, algorithm scalability, repeatability, as well as result communication and interpretation. It then introduces two infrastructures: (1) the Scholarly Database (SDB) (http://sdb.slis.indiana.edu), which provides free online access to 22 million scholarly records—papers, patents, and funding awards which can be cross-searched and downloaded as dumps, and (2) Scientometrics- relevant plug-ins of the open-source Network Workbench (NWB) Tool (http://nwb.slis. indiana.edu). The utility of these infrastructures is then exemplarily demonstrated in three studies: a comparison of the funding portfolios and co-investigator networks of different universities, an examination of paper-citation and co-author networks of major network science researchers, and an analysis of topic bursts in streams of text. The article concludes with a discussion of related work that aims to provide practically useful and theoretically grounded cyberinfrastructure in support of computational scientometrics research, education and practice.

Complex Networks: An Engineering View

The last decade has seen an explosion of research in network science, a science whose initial work and subsequent developments are grounded in statistical physics applied to natural systems. In recent years researchers in the engineering discipline have also taken a keen interest in complex networks which has resulted in several application areas being investigated in engineering. In this paper, we focus on network optimization and network dynamics in the context of engineered (human made) networks.

Social network analysis: A useful tool for visualizing and evaluating forestry research

SUMMARY One of the foundational studies of social network analysis produced a depiction of scientific collaboration by tracing a network of scientific papers linked by co-authorships and co-citations, which has since spurred numerous studies on the typology, organization and dynamics of scientific research networks. This paper introduces social network analysis and its analytical measures of network structure. It then demonstrates the utility of social network analysis in forestry, in the evaluation of large research networks such as the Sustainable Forest Management Network (SFMN), and suggests other important uses of network visualization to facilitate exploring, discovering and selecting resources in a database.

From Human Capital to Social Capital: A Longitudinal Study of Technology–Based Academic Entrepreneurs

This article develops theory relating to how differences in the human capital of academic entrepreneurs influence their ability to develop social capital that can address the barriers to venture development. We examine the development of social capital by three types of academic entrepreneurs with differing levels of entrepreneurship experience: nascent, novice, and habitual entrepreneurs. Using a longitudinal study, critical differences are observed between the structure, content, and governance of their social networks. We propose that entrepreneurs with prior business ownership experience have broader social networks and are more effective in developing network ties. Less experienced entrepreneurs likely encounter structural holes between their scientific research networks and industry networks. Support initiatives help attract industry partners for novice entrepreneurs from engineering and the material sciences but academics based within biological sciences encounter greater difficulties building such ties. Regardless of academic discipline, business ownership experience appears essential to learn to build relationships with experienced managers and potential equity investors.

Making Sense of Mankind's Scholarly Knowledge and Expertise: Collecting, Interlinking, and Organizing What We Know and Different Approaches to Mapping (Network) Science

In this paper I discuss and compare different approaches to collecting, interlinking, organizing, and making sense of scholarly knowledge and expertise in a comprehensive and timely fashion. ‘Comprehensive’ refers to the need for collecting and interlinking multilingual, multidisciplinary records from multiple sources such as publications, patents, grants, and others to truly capture all relevant knowledge. By ‘timely’ I want to emphasize that there has to be a way to integrate the most recent—that is, today's—publications with existing holdings of scholarly knowledge and expertise. I then discuss the advantages and limitations of using search engines to access, and text mining and data mining to help extract, meaning from mankind's wisdom. Next I suggest the usage of semantic association networks as a viable and complementary alternative for interlinking and making sense of scholarly knowledge and expertise. The second part of the paper exemplifies and contrasts three approaches that can be used to delineate and make sense of scholarly knowledge. The first approach uses questionnaire data, the second uses citation data from a major digital library, and the third uses personal bibliography files. These approaches are exemplified by mapping the emerging research area of network science. A particular focus is the identification of major experts, papers, and research areas and geospatial locations in which network science research is conducted. The paper concludes with a summary and outlook.

PRINTO scholarships: the Italian experience.

The increasing availability of the internet allows physicians to access actualized medical information quickly and easily, but it is not comparable with the possibility of working in a well known international medical centre.International collaboration (scholarships, courses and research), such as the PRINTO alpha project, allows professionals not only to increase and share scientific knowledge and experiences but also to integrate into a working team in a foreign country which leads to an understanding among cultures.PRINTO has set up a scientific and technical collaborative research network in Paediatric Rheumatology for Latin American physicians.

RECIPROCITY AND THE EMERGENCE OF POWER LAWS IN SOCIAL NETWORKS

Research in network science has shown that many naturally occurring and technologically constructed networks are scale free, that means a power law degree distribution emerges from a growth model in which each new node attaches to the existing network with a probability proportional to its number of links (= degree). Little is known about whether the same principles of local attachment and global properties apply to societies as well. Empirical evidence from six ethnographic case studies shows that complex social networks have significantly lower scaling exponents γ ~ 1 than have been assumed in the past. Apparently humans do not only look for the most prominent players to play with. Moreover cooperation in humans is characterized through reciprocity, the tendency to give to those from whom one has received in the past. Both variables — reciprocity and the scaling exponent — are negatively correlated (r = -0.767, sig = 0.075). If we include this effect in simulations of growing networks, degree distributions emerge that are much closer to those empirically observed. While the proportion of nodes with small degrees decreases drastically as we introduce reciprocity, the scaling exponent is more robust and changes only when a relatively large proportion of attachment decisions follow this rule. If social networks are less scale free than previously assumed this has far reaching implications for policy makers, public health programs and marketing alike.

Preface: Volume 72, Issue 4

Software Evolution through TransformationsBusinesses, organisations and society at large are increasingly reliant on software at all levels. An intrinsic characteristic of software addressing a real-world application is the need to evolve. Such evolution is inevitable if the software is to remain satisfactory to its stakeholders. Changes to software artifacts and related entities tend to be progressive and incremental, driven, for example, by feedback from users and other stakeholders. Changes may be needed for a variety of reasons, such as bug reports, requests for new features or, more generally, changes of functional requirements, or by the need to adapt to new technology, e.g., to interface to other systems.In general, evolutionary characteristics are inescapable when the problem to be solved or the application to be addressed belongs to the real world. Transformations of artifacts like models, schemata, data, program code, or software architectures provide a uniform and systematic way to express and reason about the evolution of software systems. Literally, all activities that lead to the creation or modification of documents have a transformational aspect, i.e., they change a given structure into a new one according to pre-defined rules. In most cases, these rules manifest themselves as user-invoked operations in CASE tools or program editors. More abstract examples include rules for model refinement, for translating models to code, for recovering designs from legacy systems, or for refactoring and restructuring software. Workshop ObjectivesThis workshop aimed at providing a forum for the discussion of transformational techniques in software evolution with particular focus on approaches that are generally applicable throughout the software development life-cycle. Thereby, we have distinguished two co-existing, complementary and mutually reinforcing views of the evolution theme. The more widespread view focuses on the how of software evolution, emphasising the methods and means by which software is evolved. The other focuses on the what and why of the evolution phenomenon, its nature and underlying drivers. Being mutually supportive, both views are required. A better understanding of the phenomenon leads to more appropriate ways of achieving evolution. Both views are supported through the general concept of transformation, i.e., the manual, interactive, or automatic manipulation of artifacts according to pre-defined rules, either as a conceptual abstraction of human software engineering activities, or as the implementation of mappings on and between modelling and programming languages. Workshop ProgramThe workshop was scheduled for two half days and included one invited talk by Stuart Kent as well as presentations of contributed papers in two regular sessions on different Transformation Techniques and on the mutual Compatibility of Transformations. In addition, the workshop featured a special session on Case Studies for Visual Modelling Techniques held jointly with the Workshop on Graph Transformation and Visual Modelling Techniques (GT-VMT 2002) as part of the work carried out under the European research training network SegraVis (for Syntactic and Semantic Integration of Visual Modelling Techniques). AcknowledgementThis workshop has been supported by the following projects: the scientific research network Foundations of Software Evolution, funded by the Fund for Scientific Research - Flanders (Belgium), the scientific research network RELEASE (for REsearch Links to Explore and Advance Software Evolution) funded by the European Science Foundation, and the European research training network SegraVis.Reiko Heckel, Universität Paderborn, GermanyTom Mens, Vrije Universiteit Brussel, BelgiumMichel Wermelinger, Universidade Nova de Lisboa and ATX Software SA, Lisbon, Portugal

Research networks, scientific communication, and the personal computer

Personal-computer-based communications media-electronic mail, bulletin boards, and computer conferencing-have great potential for integrating scholarly and scientific research networks. Research networks, or information organizations of faculty who share an interest in a research area, are central to scholarly and scientific progress. They have been critized, however, for their exclusion of young researchers and of faculty at isolated or low prestige institutions. Studies show that computer networking opens network access by obliterating social barriers and status distinctions. It has often been argued that, if used as a medium for research network communication, computer networking could democratize research networks. Personal computer information services designed for personal computer uses, as well as personal-computer-based bulletin board systems, represent the most promising avenue for research network communication owing to their low cost, flexibility, and egalitarian ethos.