Paper Info
Title
Structure and Dynamics of Research Collaboration in Computer Science
Abstract
Complex systems exhibit emergent patterns of behavior at different levels of organization. Powerful network analy- sis methods, developed in physics and social sciences, have been successfully used to tease out patterns that relate to community structure and network dynamics. In this paper, we mine the complex network of collaboration relationships in computer science, and adapt these network analysis meth- ods to study collaboration and interdisciplinary research at the individual, within-area and network-wide levels. We start with a collaboration graph extracted from the DBLP bibliographic database and use extrinsic data to de- fine research areas within computer science. Using topolog- ical measures on the collaboration graph, we find significant differences in the behavior of individuals among areas based on their collaboration patterns. We use community structure analysis, betweenness centralization, and longitudinal assor- tativity as metrics within each area to determine how central- ized, integrated, and cohesive they are. Of special interest is how research areas change with time. We longitudinally ex- amine the area overlap and migration patterns of authors, and empirically confirm some computer science folklore. We also examine the degree to which the research areas and their key conferences are interdisciplinary. We find that data mining and software engineering are very interdisciplinary while theory and cryptography are not. Specifically, it appears that SDM and ICSE attract authors who publish in many areas while FOCS and STOC do not. We also examine isolation both within and between areas. One interesting discovery is that cryptography is highly isolated within the larger computer science community, but densely interconnected within itself. 1 Background and Motivation Computer science is a diverse and growing area of schol- arly activity, with many subareas, such as artificial intel- ligence (AI), computational biology (CBIO), cryptography (CRYPTO), databases (DB), graphics (GRAPH), program- ming languages (PL), software engineering (SE), security (SEC), theory (THEORY), among others. Some of these areas are quite old, rooted in the earliest stirrings of the field (e.g., THEORY) and others started much later (e.g., GRAPH). Some are quite large, attracting a large number of researchers (e.g., DB and GRAPH) and others are smaller (e.g., CRYPTO and SE). Some are in a stable phase (e.g., THEORY); others are growing rapidly (e.g., SEC). There are other, more subtle differences in character and style between areas. These differences, although they are currently not rigorously quantified, nevertheless may have important implications for the future of these areas. These differences are recognized by researchers working in the respective (or closely allied) areas, but have not been rigorously studied. For example, some of these areas are considered intellectually unified, while others are said to include several distinct, thriving groups. Some areas tend to interact strongly with others, with a tradition of mutual enrichment, and others are more stand-alone. Some areas are dominated by a few researchers, while others have a more diffuse collaborative structure. In some areas, older and younger researchers frequently collaborate, while in others, researchers collaborate primarily with others like them. These informal, folkloric differences between areas are worthy of study, because such properties clearly can have a strong influence on the intellectual vibrancy and diver- sity of an area. In this paper, we begin to quantify and study these differences to produce data that may provide "actionable intelligence" for interested parties. For exam- ple, researchers (students, new faculty) might well consider these factors when deciding whether to enter (or leave) an area. Funding agencies (industries, government foundations) might consider the status and style of a field, choose to for- mulate Broad Area Announcements (BAAs) and Calls for Proposals to influence a field, for example, to become more interdisciplinary, or more intellectually diverse, or to spread their funding more broadly to increase centers of influence; contrariwise, they could design funding initiatives to reverse such trends if that seems appropriate.
Year
Venue
Keywords
2009
SDM
data mining,complex network,network analysis,software engineering,computational biology,betweenness centrality,social science,complex system,science communication,community structure,network dynamics
Field
DocType
Citations 
Data science,Assortativity,Network dynamics,Computer science,Centrality,Betweenness centrality,Collaboration graph,Complex network,Collaborative network,Special Interest Group
Conference
18
PageRank 
References 
Authors
1.06
15
6
Name
Order
Citations
PageRank
Christian Bird12469115.59
Earl Barr287140.79
a nash3181.06
Premkumar Devanbu44956357.68
Vladimir Filkov5150375.32
Zhendong Su63397175.76