A avaliação acadêmica (de docentes, departamentos, programas de pós-graduação ou instituições) é uma atividade rotineira e utilizada para, por exemplo, a concessão de recursos para um projeto, definição da verba que será liberada a um dado programa de pós-graduação ou mesmo para a autorização do oferecimento de um curso de doutorado para um programa de pós-graduação. Contudo, determinar a qualidade do ensino não é uma tarefa trivial, com metodologias comparativas ainda em estudo. Dois destes critérios são informações bibliométricas e análise das redes de coautorias dos programas. Assim, o desenvolvimento de métodos e ferramentas para automatizar parte da avaliação ou mesmo para identificar a importância de diferentes métricas podem tanto auxiliar no processo de avaliação como servir para auxiliar as pessoas ou os grupos que serão avaliados. Neste trabalho foram desenvolvidas ferramentas para a realização automática de análise bibliométrica e de redes sociais para a caracterização e ranqueamento de grupos acadêmicos, bem como investigar a relação entre diferentes medidas e alguns ranqueamentos acadêmicos existentes. As ferramentas foram avaliadas considerando departamentos internacionais e programas de pós-graduação da área de Ciência da Computação. Os resultados de nosso estudo indicaram quais métricas contribuíram positivamente e negativamente para a posição dos ranqueamentos dos programas / The academic evaluation (of researchers, departments, graduate programs or institutions) is a routine activity and used for the granting of resources for a project, definition of the funds that will be released to a given graduate program or even for the authorization of the offering of a doctorate course. However, this activity is complex with comparative methodologies still in evaluation. Two criteria used are bibliometrics information and scientific social network analysis. Thus, the development of methods and tools to automate part of the evaluation or even to identify the importance of different metrics for the evaluation can both aid the evaluation process and serve as a guide for the researchers or groups that will be evaluated. In this work, tools were developed for performing automatically bibliometric and social networks analysis for the characterization and ranking of academic groups, as well as to investigate the relationship between different measures and some existing academic rankings. The tools were evaluated considering international departments and graduate programs in the Computer Science area. Our results showed which metrics contributed positively and negatively in the rank positioning of the programs

A pesquisa-ação, desde a sua origem, associada às Ciências Sociais, sofreu diferentes incorporações teóricas, tanto relacionadas ao seu conceito quanto à sua prática, que resultaram em muitas interpretações distintas que, às vezes, podem dificultar o seu emprego em produções acadêmicas da área de Ciências da Saúde. Objetivo: identificar e caracterizar as teses de doutorado e as dissertações de mestrado produzidas nos programas de pós-graduação de Ciências da Saúde da Universidade de São Paulo (USP), entre 2000 a julho de 2012, em que foi utilizada a metodologia de pesquisa-ação, além de analisar os conceitos de pesquisa- ação que foram empregados nestes trabalhos. Metodologia: utilizamos o método bibliográfico, a bibliometria e a Teoria Comunicativa da Terminologia (TCT), através da abordagem qualitativa e quantitativa. Usamos como fonte de dados, a Biblioteca Digital de Teses e Dissertações da USP (BDTD/USP) e como instrumento, usamos uma ficha de coleta de dados validada por especialistas. Realizamos a análise dos dados de forma manual. Resultados: a pesquisa-ação foi utilizada em 49 trabalhos (31 de doutorado e 18 de mestrado), de autoria de 45 diferentes alunos/autores que, em sua maioria, tinham formação em Enfermagem (54,17%), atuando na linha de pesquisa de Educação em Saúde e Formação de Recursos Humanos (51,02%), do Programa de Pós-graduação em Enfermagem Psiquiátrica (55,10%), da Escola de Enfermagem de Ribeirão Preto (EERP) (61,22%), sob a orientação de uma mesma docente (48,98%). Predominou trabalhos defendidos entre 2007 e 2011 (61,22%), em que o pesquisador atuava, previamente, no campo de aplicação do estudo (59,19%), desenvolvidos sem apoio financeiro de órgãos de fomento (69,39%). Os trabalhos apresentaram abordagem qualitativa, com temáticas sobre pesquisa-ação (5,91%), enfermagem (5,38%) e educação em saúde (3,76%), tendo como foco principal a formação/prática do profissional de saúde (44,90%). Nestes, prevaleceu o emprego da pesquisa-ação crítica (89,80%), com a realização de todas as etapas da pesquisa-ação indicada por Thiollent. Identificamos 124 citações relacionadas ao conceito de pesquisa-ação (média=2,82 citações/trabalho), provenientes de 50 referências e 48 autores diferentes. Em sua maioria, eram livros (58,06%), publicados em português e, na década de 2000. O livro Metodologia da pesquisa-ação foi empregado para conceituar a pesquisa-ação por 88,64% dos trabalhos. Em ordem decrescente, os autores Thiollent, Bueno e Freire foram os mais citados. Identificamos 356 excertos de citações relacionados ao conceito de pesquisa-ação (média=8,09 excertos/trabalho). A pesquisa-ação foi conceituada mediante a descrição dos tipos de pesquisas relacionadas a essa metodologia, sua origem, aplicação e princípios básicos. Considerações finais: a aplicação da pesquisa-ação em teses de doutorado indica que esta metodologia apresenta rigor científico adequado às exigências de estudos considerados complexos. A pesquisa-ação foi conceituada, principalmente, através de fontes específicas sobre o tema em questão, com a utilização de um núcleo principal de fontes e de atores utilizados como referências para conceituar a temática em foco. / Action research, since its origin, linked to the Social Sciences, presented various forms of theoretical inclusions, related to its concept and practice, which resulted in many different interpretations that can sometimes hinder its use in academic productions in the area of Health Sciences. Objective: to identify and characterize the dissertation of masters and doctorate theses developed in the post-graduate program in Health Sciences of the University of São Paulo (USP), between 2000 and July 2012, in which was used the action research methodology, apart from analyzing the concepts of action research were used in these studies. Methodology: We used the review of literature, the bibliometrics and Communicative Theory of Terminology (TCT), through qualitative and quantitative approach. It was used as a data source, the Digital Library of Theses and Dissertations of USP (BDTD/USP) and it was used a data collection form, validated by experts. We conducted data analysis manually. Results: action research was used in 49 studies (31 doctorates and 18 masters), which were written by 45 different students/authors, who mostly had training in Nursing (54.17%), working in line research in Health Education and Training Human Resources (51.02%) of the Graduate Program in Psychiatric Nursing (55.10%), of the Nursing School of Ribeirão Preto (EERP) (61.22%), under the supervision of the same teacher (48.98%).Studies defended predominated between 2007 and 2011 (61.22%), in which the researcher worked previously in the field of study (59.19%), which were developed without the financial support of funding agencies (69.39% ). The studies had qualitative approach, presenting themes about action research (5.91%), nurses (5.38%) and health education (3.76%), with the main objective of formation/practice of the health care professionals (44.90%). In these studies, the prevailing use of critical action research (89.80%), with the completion of all stages of the research- action indicated by Thiollent. In the studies analyzed, we identified 124 citations related to the concept of action research (mean = 2.82 citations/study), from 50 authors and 48 different references. Most part of it was books (58.06%), published in Portuguese in the 2000s. The book Action research methodology was employed to conceptualize action research by 88.64% of the studies. In descending order, the authors Thiollent, Bueno and Freire were the most cited. We identified 356 excerpts of citations related to the concept of action research (mean = 8.09 excerpts/study). Action research was conceptualized by describing the types of research related to this methodology, its origin, and applying basic principles. Final considerations: the application of action research in the theses indicates that this methodology has adequate scientific rigor to the demands of the studies considered complex. Action research was conceptualized mainly through specific sources on the topic in question, with the existence of a core group of actors and sources used as references to conceptualize the thematic in focus.

Economia e Gestão Internacional / Master in International Economics and Management / Os estudos que se debruçam sobre a influência e o impacto de organizações geradores de conhecimento (e.g., universidades ou instituições de I&D) têm sido, normalmente, abordados por meio da análise estritamente económica, enfatizando o seu impacto económico no âmbito local, regional ou nacional. No presente estudo, avança-se com uma metodologia alternativa de modo a avaliar a influência e o impacto científico internacional de uma instituição geradora e difusora de conhecimento. São assim estudados dois ramos da literatura que tratam, por um lado, da mensurabilidade do impacto económico de organizações de I&D, e por outro lado, dos fluxos de conhecimento: nomeadamente, estudos económicos tradicionais e análises cienciométricas e bibliométricas. Consequentemente, apresentamos aqui uma metodologia complementar, baseada na cienciometria e bibliometria, por considerar a influência de uma instituição de I&D através da análise da produção científica desenvolvida e por via do reconhecimento da sua relevância pela comunidade científica internacional. Concretamente, tendo como caso de estudo o INESC Porto, analisamos a dinâmica da sua produção científica durante os últimos doze anos, dando especial relevo à evolução das suas co-autorias científicas internacionais, delineando a arquitectura da sua network de conhecimento, bem como da sua estrutura (provavelmente) mutável. Adicionalmente, dando um enfoque especial às suas áreas científicas mais prolíficas, e atendendo ao trabalho científico registado no Science Citation Index (SCI), mapeamos as citações e inferimos sobre a sua influência e o seu impacto científico internacional. Desta forma, somos capazes de quantificar e mapear a rede científica internacional de uma organização produtora de conhecimento, através do uso de métodos estatísticos descritivos e geográficos, bem como por meio de modelos logit, que permitem a visualização do âmbito e a avaliação da importância da estrutura de influência internacional do INESC Porto. Os resultados demonstram que o INESC Porto tem vindo a expandir a sua rede científica internacional. De facto, a sua rede de influência ao nível do conhecimento chega aos cinco continentes. Para além disso, as estimações econométricas levam-nos a concluir que a influência geográfica alargada da investigação científica do INESC Porto não é resultado do seu posicionamento internacional em termos de co-autorias, mas antes sim da qualidade intrínseca da sua produção científica.

本研究以擇定之檢索策略，自ERIC資料庫取得38,230筆領導主題文獻，利用書目計量分析方法，探討ERIC資料庫時間範圍是「最早至2009年」之領導主題文獻特性，包括文獻之年代、語言、教育階段、出版類型，以及適合讀者的分布；再以布萊德福定律和布萊德福─齊夫定律圖解，驗證領導主題期刊文獻的分布情形，並分析領導主題核心期刊之特色；最後統計核心期刊1980年至2009年間刊載的文章中，於摘要明顯提及的領導理論或名詞，以找出新近領導主題的研究焦點，進而掌握未來領導研究的趨勢。 根據研究結果與分析，本研究之研究結論歸納如下： 一、領導主題文獻呈現穩定成長，其成長模式為指數模式。 二、英文是領導主題文獻的主要使用語言。 三、中小學教育乃是領導主題文獻最為關注的教育階段。 四、研究報告為領導主題文獻的主要傳播媒介。 五、實踐者是領導主題文獻最主要的適合讀者。 六、領導主題期刊文獻分布不符合布萊德福定律，但與布萊德福─齊夫定律圖解法相符，求得之核心期刊共有十三種。 七、核心期刊有出版國家和學科主題集中的情形。 八、教學領導、分散領導、教師領導、轉型領導、道德領導，以及交易領導為新近領導主題的焦點。 最後，本研究依據研究結果，針對資料庫製作者、圖書資訊服務界、研究者，以及未來研究等四方面，提出相關建議。 / The purpose of this study is to explore the characteristics of leadership topics literature from earliest to 2009 based on the theoretical perspectives of the bibliometrics. Five fields are used to compile statistics, including publication year, language, education level, publication type, and target audience. The scattering of leadership topics journal articles is examined by Bradford’s Law and Bradford-Zipf law, and the characteristics of core journals are analyzed. Finally, the study counts the leadership theories or ideas explicitly mentioned in abstracts of articles published by core journals from 1980 to 2009 to find out the focuses of the lately leadership topics, and have a better understanding of the developments and trends in leadership research. A total of 38,230 bibliographic records are retrieved from ERIC database. The study concludes as follows: 1.The overall growth of the leadership topics literature is stable, and the growth curve fits exponential growth. 2.English is the most often used language in leadership topics literature. 3.Elementary Secondary Education is the education level that leadership topics literature most concerns. 4.Reports are the major publication type of leadership topics literature. 5.Practitioners are the primary target audience of leadership topics literature. 6.The scattering of leadership topics journal articles doesn’t fit Bradford’s Law, but it fits the S-shaped curve of Bradford-Zipf law. According to Bradford’s verbal expression, there are 13 core journals in this study. 7.The core journals have the publication country and the discipline subject centralized situation. 8.The focuses of the lately leadership topics are instructional leadership, distributed leadership, teacher leadership, transformational leadership, moral leadership, and transactional leadership. According to the results, some suggestions were given for database provider, libraries or information service institutions, researchers, and further research.

This thesis provides an in-depth study of the possibilities of applying bibliometric methods to the research field of literary studies. The four articles that constitute the backbone of this thesis focus on different aspects of references and citations in literary studies: from the use of references in the text to citation patterns among 34 literature journals. The analysis covers both an Anglo-Saxon context as well as research in Swedish literary studies, and the materials used include Web of Science data, references in the Swedish literature journal TFL (Tidskrift för Litteraturvetenskap) and applications to the Swedish Research Council (Vetenskapsrådet). A study is also made of the influence of one single publication—Walter Benjamin’s Illuminations—and its impact in literary studies and in wider academia. The results from the four articles are elaborated upon using a theoretical framework that focuses on differences in the social and intellectual organization of research fields. According to these theories literary studies can be described as a fragmented, heterogenic, interdisciplinary and ‘rural’ field with a diverse audience. The fragmented and rural organization of the field is reflected in low citation frequencies as well as in the difficulties in discerning research specialities in co-citation mappings, while the analysis of the intellectual base (highly cited authors) is an example of the heterogenic and interdisciplinary character of the field, as it includes authors from many fields across the humanities and the social sciences. The thesis emphasizes that bibliometric studies of research fields in the humanities need to incorporate non-English and non-journal publications in order to produce valid and fair results. Moreover, bibliometric methods must be modified in accordance with the organization of research in a particular field, and differences in referencing practices and citation patterns ought to be considered. Consequently, it is advised that bibliometric measures for evaluating research in these fields should, if used at all, be applied with great caution. / © Björn Hammarfelt 2012

La Universidad Nacional Autónoma de México tiene las siguientes funciones sustantivas: la docencia, la investigación y la difusión de la cultura. Sin embargo, hasta la fecha la Universidad no dispone de indicadores que den a conocer el grado de desarrollo al que ha llegado la investigación que se lleva a cabo en ella. Por tanto, se intentó determinar el desempeño de los investigadores universitarios en el periodo 1995-2003 a través de un análisis bibliométrico. Para la construcción de los datos empíricos se recurrió a la Web of Knowledge. Se identificaron 9903 artículos que lograron acumular 65291 citas, publicados la mayoría de ellos en disciplinas correspondientes a las ciencias duras. Se concluye que los investigadores de la UNAM tienen que esforzarse aún más para que la investigación se refleje en publicaciones que lleven a su incorporación a un mercado de la ciencia visible. / The National Autonomous University of Mexico has the following functions: teaching, research and popularization of culture. The University has been recognized by its outstanding performance and has been considered the most relevant institution of higher education at the national level and one of the best at the international level. However, so far, there are no indicators that show the direction into which the research activity is moving. Therefore, this investigation intended to determine the University's research performance from 1995 to 2003 using bibliometric techniques. The Web of Knowledge was used to identify cited papers and citations attracted. In order to determine the visibility of the communication channels used for the publication of research results, we used the Journal Citation Reports (JCR). We also tried to determine the global publication rate (GPR) of the researchers and an approximation of the cost of published articles by university departments.We identified 9903 articles that accumulated 65291 citations. We found that scientists in the hard sciences are more active and more cited than their colleagues in other areas. Three disciplines were the more active: space sciences, physics and microbiology.We concluded that researchers at the University have to work harder in order to be taken into account by the science that counts, i.e. the science that is visible.

As the Web has become more and more important to our daily lives, algorithms that can effectively utilize the link structure have become more and more important. One such task has been to find communities in social network data. Recently, however, there has been increased interest in augmenting links with additional semantic information. We examine link classification from the point of view of scientometrics, with an eye towards applying what has been learned about scientific citation to Web linking. Some community detection algorithms are reviewed, and one that has been developed for topical community finding on the Web is adapted to typed scientific citations.

As the Web has become more and more important to our daily lives, algorithms that can effectively utilize the link structure have become more and more important. One such task has been to find communities in social network data. Recently, however, there has been increased interest in augmenting links with additional semantic information. We examine link classification from the point of view of scientometrics, with an eye towards applying what has been learned about scientific citation to Web linking. Some community detection algorithms are reviewed, and one that has been developed for topical community finding on the Web is adapted to typed scientific citations.

Science and Technology (S and T) information presents a rich resource, essential for managing research and development (R and D) programs. Management of R and D has long been a labor-intensive process, relying extensively on the accumulated knowledge of experts within the organization. Furthermore, the rapid pace of S and T growth has increased the complexity of R and D management significantly. Fortunately, the parallel growth of information and of analytical tools offers the promise of advanced decision aids to support R and D management more effectively. Information retrieval, data mining and other information-based technologies are receiving increased attention. In this thesis, a framework based on text mining techniques is proposed to discover useful intelligence implicit in large bodies of electronic text sources. This intelligence is a prime requirement for successful R and D management. This research extends the approach called Technology Opportunities Analysis (developed by the Technology Policy and Assessment Center, Georgia Institute of Technology, in conjunction with Search Technology, Inc.) to create the proposed framework. The commercialized software, called VantagePoint, is mainly used to perform basic analyses. In addition to utilizing functions in VantagePoint, this thesis also implements a novel text association rule mining algorithm for gathering related concepts among text data. Two algorithms based on text association rule mining are also implemented. The first algorithm called tree-structured networks is used to capture important aspects of both parent-child (hierarchical structure) and sibling relations (non-hierarchical structure) among related terms. The second algorithm called concept-grouping is used to construct term thesauri for data preprocessing. Finally, the framework is applied to Thai S and T publication abstracts toward the objective of improving R and D management. The results of the study can help support strategic decision-making on the direction of S and T programs in Thailand.

The Institute for Scientific Information's (ISI) citation databases have been used for decades as a starting point and often as the only tools for locating citations and/or conducting citation analyses. ISI databases (or Web of Science [WoS]), however, may no longer be sufficient because new databases and tools that allow citation searching are now available. Using citations to the work of 25 library and information science faculty members as a case study, this paper examines the effects of using Scopus and Google Scholar (GS) on the citation counts and rankings of scholars as measured by WoS. Overall, more than 10,000 citing and purportedly citing documents were examined. Results show that Scopus significantly alters the relative ranking of those scholars that appear in the middle of the rankings and that GS stands out in its coverage of conference proceedings as well as international, non-English language journals. The use of Scopus and GS, in addition to WoS, helps reveal a more accurate and comprehensive picture of the scholarly impact of authors. WoS data took about 100 hours of collecting and processing time, Scopus consumed 200 hours, and GS a grueling 3,000 hours.