Co-authorship networks are a particular sort of social networks representing authors collaborating on joint publications. Such networks are studied within the fields of bibliometrics and scientometrics. While it is possible to analyze co-authorship networks in their entirety, certain analytical tasks would benefit from representing such networks as dynamic graphs, which incorporate a temporal dimension and capture structural transformations unfolding over time. The importance of dynamic graphs has emerged in recent years, in graph theory at large as well as within application domains such as social sciences, for instance.Research regarding dynamic graphs has been identified as one of the major challenges within network theory since they are particularly useful for describing real-world systems.This thesis project revolves around dynamic co-authorship network analysis algorithms, which aim to extract various temporal aspects regarding author collaborations.It is the result of a proposal by the ISOVIS group at Linnaeus University, which is active within the fields of exploratory data analysis and information visualization, including the problem of visual analysis of scientific publication data. The algorithms developed in this project extract analytical data such as (1) joint publications among pairs of authors, (2) temporal trends on connected components (groups of authors) along with network centrality measurements, and (3) major events regarding emergence, mergers, and splits of connected components over time. Together with domain experts, the analysis regarding usability, performance, and scalability of the algorithms took place as part of the evaluation process to assure that the result met the needs which instigated this thesis project. The application of the algorithms on real data sets provided by the ISOVIS group was useful concerning the evaluation of the usability domain. In contrast, customized synthetic data sets was an excellent tool for evaluating performance and scalability.

<p>Almost any large-scale system can be viewed as a network that shows interac-tions among entities which are constituent parts of the system. The focus of this<br />dissertation is on complex networks from three domains: (1) networks extracted<br />from source code of computer programs that represent design of software systems,<br />(2) networks extracted from semantic web ontologies that describe the structure<br />of shared and reusable knowledge, and (3) networks extracted from bibliographic<br />records that depict collaboration in science. We proposed new methods for the<br />extraction of networks from mentioned domains. Secondly, on several case stud-ies we demonstrated benets of network-based analysis of concrete systems from<br />those domains. In contrast to the previous work on the subject, analyses pre-sented in this dissertation are not purely topological, but combine techniques and<br />metrics developed under the framework of complex network theory with domain-dependent metrics.</p> / <p>Skoro svaki kompleksan sistem se može predstaviti mrežom koja opisuje interakcije izmedju entiteta od kojih je sistem komponovan. Fokus ove disertacije je na&nbsp;kompleksnim mrežama iz tri domena: (1) mreže ekstrahovane iz izvornog koda&nbsp;računarskih programa koje reprezentuju dizajn softverskih sistema, (2) mreže ekstrahovane iz ontologija semantičkog web-a koje opisuju strukturu deljenog znanja&nbsp;pogodnog za vi&scaron;ekratnu upotrebu, i (3) mreže ekstrahovane iz bibliografskih zapisa koje opisuju saradnju istraživača. U okviru disertacije predložene su nove&nbsp;metode za ekstrakciju mreža iz pomenutih domena. Drugo, na nekoliko studija&nbsp;slučaja ilustrovani su beneti mrežno orjentisane analize konkretnih sistema iz&nbsp;domena obuhvaćenih disertacijom. U poredjenju sa prethodnim relevantim istraživanjima, analize prezentovane u disertaciji nisu čisto topolo&scaron;ke, nego kombinuju tehnike i metrike razvijene u okviru teorije kompleksnih mreža sa metrikama iz konkretnog domena.</p>

Comment les structures sociales et sémantiques d'une communauté scientifique guident-elles les dynamiques de collaboration à venir ? Dans cette thèse, nous combinons des techniques de traitement automatique des langues et des méthodes provenant de l'analyse de réseaux complexes pour analyser une base de données de publications scientifiques dans le domaine de la linguistique computationnelle : l'ACL Anthology. Notre objectif est de comprendre le rôle des collaborations entre les chercheurs dans la construction du paysage sémantique du domaine, et, symétriquement, de saisir combien ce même paysage influence les trajectoires individuelles des chercheurs et leurs interactions. Nous employons des outils d’analyse du contenu textuel pour extraire des textes des publications les termes correspondant à des concepts scientifiques. Ces termes sont ensuite connectés aux chercheurs pour former un réseau socio-sémantique, dont nous modélisons la dynamique à différentes échelles. Nous construisons d’abord un modèle statistique, à base de régressions logistiques multivariées, qui permet de quantifier le rôle respectif des propriétés sociales et sémantiques de la communauté sur la dynamique microscopique du réseau socio-sémantique. Nous reconstruisons par la suite l’évolution du champ de la linguistique computationelle en créant différentes cartographies du réseau sémantique, représentant les connaissances produites dans le domaine, mais aussi le flux d’auteurs entre les différents champs de recherche du domaine. En résumé, nos travaux ont montré que la combinaison des méthodes issues du traitement automatique des langues et de l'analyse des réseaux complexes permet d'étudier d'une manière nouvelle l'évolution des domaines scientifiques. / How are the social and semantic structures of a scientific communitydriving future research dynamics? In this thesis we combine naturallanguage processing techniques and network theory methods to analyzeavery large dataset of scientific publications in the field of computationallinguistics,i.e.the ACL Anthology. Ultimately, our goal is to understandthe role of collaborations among researchers in building and shaping thelandscape of scientific knowledge, and, symmetrically, to understand howthe configuration of this landscape influences individual trajectories ofresearchers and their interactions. We use natural language processingtools to extract the terms corresponding to scientific concepts from thetexts of the publications. Then we reconstruct a socio-semantic networkconnecting researchers and scientific concepts, and model the dynamicsof its evolution at different scales. To achieve this, we first build astatistical model, based on multivariate logistic regression, that quantifiesthe role that social and semantic features play in the evolution of thesocio-semantic network, namely in the emergence of new links. Then, wereconstruct the evolution of the field through different visualizations of theknowledge produced therein, and of the flow of researchers across thedifferent subfields of the domain. To summarize, we have shown throughour work that the combination of natural language processing techniqueswith complex network analysis makes it possible to investigate in a novelway the evolution of scientific fields.

Made available in DSpace on 2015-04-16T15:23:21Z (GMT). No. of bitstreams: 1 arquivototal.pdf: 3269377 bytes, checksum: 9ba367f6ed340606cb82a4fc21abdd5a (MD5) Previous issue date: 2012-04-16 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / Scientific production is an exteriorization way of scientific knowledge and it reaches its summit with its communication once produced knowledge, besides enhancing the area, is socialized. In this context, the research aims at analyzing quantitatively the professors´ production of the Post-graduation Program in Information Science (PPGCI), of the Federal University of Paraíba, from 2007 and 2011, in order to map the production of scientific articles in journals, complete studies and abstracts published in conference proceedings. It is a documental research and quanti-qualitative procedures, for data analysis, were used. The sources of the analysis are the Lattes curricula from the sixteen professors of the Post-graduation Program in Information Science, of the Federal University of Paraíba. A questionnaire was also employed in order to complement information, present in the Lattes curricula, about the professors´ profile and to collect their opinions about production in co-authorship. Data analysis shows that the professors´ production, in the studied period, totalized 418 publications being 178 articles, 192 communications in events and 48 abstracts, which refer to 175 different themes. For such production, there was collaboration of 317 co-authors. Professors, independently from their research line, are used to publish in partnership with their colleagues from the PPGI and other post-graduation programs, with students under their counseling service, master s students and PIBIC scholarship students, among others, taking part either as authors or co-authors. It can be concluded that the production of PPGCI professors was based, in the last five years, on the following aspects: co-authorship production was higher than the individual one; production of communications in events was larger than the article and abstracts one; production related to present issues and those which promote the contact with their colleagues, debates and discussions, such as communications in events. PPGCI professors are favorable to co-authorship production, but they assume that it does not provide high visibility and acknowledgement for the author. / A produção científica é uma forma de exteriorização do conhecimento científico, cujo ápice é atingido com a sua comunicação, uma vez que o conhecimento produzido, além de fortalecer a área onde foi produzido, é socializado. Nesse contexto, realizou-se a pesquisa, que tem como objetivo analisar quantitativamente a produção dos docentes do Programa de Pós-graduação em Ciência da Informação (PPGCI), da Universidade Federal da Paraíba, no período de 2007 a 2011, visando mapear a produção de artigos de periódicos, trabalhos completos e resumos publicados em anais de eventos. A pesquisa é do tipo documental e utilizou procedimentos quanti-qualitativo para a análise dos dados. As fontes de análise são os Currículos Lattes dos 16 docentes do Programa de Pós-graduação em Ciência da Informação, da Universidade Federal da Paraíba. Também foi empregado um questionário para complementar as informações constantes nos Currículos Lattes sobre o perfil dos docentes e para coletar suas opiniões sobre a produção em coautoria. A análise dos dados mostra que a produção dos docentes, nos cinco anos estudados, foi de 418 publicações: 178 artigos, 192 comunicações em eventos e 48 resumos que versam sobre 175 temáticas. Para essa produção, houve a colaboração de 317 coautores. Os docentes, independentemente da linha de pesquisa de que fazem parte, costumam publicar em parceria com seus pares do PPGCI e de outros programa de pós-graduação, com seus orientandos, mestrandos e com bolsistas PIBIC, entre outros, assumindo a posição ora de autores, ora de coautores. Conclui-se que a produção dos docentes do PPGCI guiou-se, nos últimos cinco anos, pelos seguintes pontos: produção mais em coautoria que individual; maior produção de comunicações em eventos do que de artigos e resumos, produção voltada para questões atuais e produções que proporcionam o contato com seus pares, debates e discussões, como as comunicações em eventos. Os docentes do PPGCI são favoráveis à produção em coautoria, mas não consideram que ela proporciona mais visibilidade e reconhecimento do autor.

This thesis documents the development of a web-application attacking the problem of visualization of co-authorship networks. The visualization encompasses several views.Each of them shows different aspects of the data which is loaded from Academic Archive Online (DiVa), a library system which holds all publications released in the Linnaeus University.  To detect relationships among authors, a new interactive layout for Node-Link Diagrams was developed which shows publications, authors and corresponding organizations (faculties, departments) in a radial manner. This Network-View is connected to another view showing the attributes (year, type) of the publications. In development, particular emphasis was placed on a rich support of user interaction in order to equip the user with a tool that allows graphical and explorative analysis of the underlying data.