1.Decomposingsocial and semantic networks in emerging''big data''research

Journal of Informetrics, Volume 7, Issue 3, July 2013, Pages 756-765

Park,H.W.; Leydesdorff, L.

Highlights

•Thisstudy conducts a social network analysis of international co-authorship relationshipsin “big data” science and analyzes semantic networks.

•Theresults indicate that the U.S. was the most central country, followed byGermany, the U.K., and France, in that order.

•SomeEuropean countries engaged in international collaboration to an extent greaterthan expected.

Abstract

Thispaper examines the structural patterns of networks of internationallyco-authored SCI papers in the domain of research driven by big data andprovides an empirical analysis of semantic patterns of paper titles. Theresults based on data collected from the DVD version of the 2011 SCI databaseidentify the U.S. as the most central country, followed by the U.K., Germany,France, Italy, Australia, the Netherlands, Canada, and Spain, in that order.However, some countries (e.g., Portugal) with low degree centrality occupiedrelatively central positions in terms of betweenness centrality. The results ofthe semantic network analysis suggest that internationally co-authored paperstend to focus on primary technologies, particularly in terms of programming andrelated database issues. The results show that a combination of words andlocations can provide a richer representation of an emerging field of sciencethan the sum of the two separate representations.

Keywords

Datascience; Big data; International co-authorship; Social network analysis; SCI;Semantic network

2.Correlationbetween Journal Impact Factor and Citation Performance: An experimental study

Journal of Informetrics, Volume 7, Issue 2, Pages 357-370

Finardi,Ugo

 Cited by Scopus (3)

Highlights

►Journal Impact Factors are widely diffused, but also widely misused.

►Further insights into the nature of Journal Impact Factor are offered.

►Time evolution of correlation between Journal Impact Factor and Citedness isstudied.

►Correlation coefficients are calculated and tested from a sample of Journal data.

►The studied correlation is a weaker one rather than a strong one.

Abstract

Sinceits introduction, the Journal Impact Factor has probably been the mostextensively adopted bibliometric indicator. Notwithstanding its well-knownstrengths and limits, it is still widely misused as a tool for evaluation, wellbeyond the purposes it was intended for. In order to shed further light on its nature,the present work studies how the correlation between the Journal Impact Factorand the (time-weighed) article Mean Received Citations (intended as a measureof journal performance) has evolved through time. It focuses on a sample ofhard sciences and social sciences journals from the 1999 to 2010 time period.Correlation coefficients (Pearson's Coefficients as well as Spearman'sCoefficients and Kendall's τα) are calculated and then tested against severalnull hypotheses. The results show that in most cases Journal Impact Factors andtheir yearly variations do not display a strong correlation with citedness.Differences also exist among scientific areas.

Keywords

Garfield;Journal Impact Factor; Article citations; Correlation coefficients; Timeevolution; Impact factor misuses

3.Sentiment analysis: A combined approach

Journal of Informetrics, Volume 3, Issue 2, April 2009, Pages 143-157

Prabowo,Rudy; Thelwall, Mike

 Cited by Scopus (63)

Abstract

Sentimentanalysis is an important current research area. This paper combines rule-basedclassification, supervised learning and machine learning into a new combinedmethod. This method is tested on movie reviews, product reviews and MySpacecomments. The results show that a hybrid classification can improve theclassification effectiveness in terms of micro- and macro-averaged F1. F1 is ameasure that takes both the precision and recall of a classifier’seffectiveness into account. In addition, we propose a semi-automatic,complementary approach in which each classifier can contribute to otherclassifiers to achieve a good level of effectiveness.

Keywords

Sentimentanalysis; Unsupervised learning; Machine learning; Hybrid classification

4.Exploring scientists' workingtimetable: Do scientists often work overtime?

Journal of Informetrics, Volume 6, Issue 4, October 2012, Pages 655-660

Wang,Xianwen; Xu, Shenmeng; Peng, Lian; Wang, Zhi; Wang, Chuanli; Zhang, Chunbo;Wang, Xianbing

 Cited by Scopus (6)

Highlights

►A novel methodology is proposed to analyze scientists’ working timetable.

►Scientists are still engaged in their research after working hours.

►Scientists’ working time curves in different countries show different features

Abstract

Anovel method is proposed to monitor and record scientists’ working timetable.We record the downloads information of scientific papers real-timely fromSpringer round the clock, and try to explore scientists’ working habits. As ourobservation demonstrates, many scientists are still engaged in their researchafter working hours every day. Many of them work far into the night, even tillnext morning. In addition, research work also intrudes into their weekends.Different working time patterns are revealed. In the US, overnight work is moreprevalent among scientists, while Chinese scientists mostly have busy weekendswith their scientific research.

Keywords

Scientist;Realtime; Working habits; Springer; Work–family conflict

5.Measuringcontextual citation impact of scientific journals

Journal of Informetrics, Volume 4, Issue 3, July 2010, Pages 265-277

Moed,Henk F.

 Cited by Scopus (106)

Abstract

Thispaper explores a new indicator of journal citation impact, denoted as sourcenormalized impact per paper (SNIP). It measures a journal's contextual citationimpact, taking into account characteristics of its properly defined subjectfield, especially the frequency at which authors cite other papers in theirreference lists, the rapidity of maturing of citation impact, and the extent towhich a database used for the assessment covers the field's literature. Itfurther develops Eugene Garfield's notions of a field's ‘citation potential’defined as the average length of references lists in a field and determiningthe probability of being cited, and the need in fair performance assessments tocorrect for differences between subject fields. A journal's subject field isdefined as the set of papers citing that journal. SNIP is defined as the ratioof the journal's citation count per paper and the citation potential in itssubject field. It aims to allow direct comparison of sources in differentsubject fields. Citation potential is shown to vary not only between journalsubject categories – groupings of journals sharing a research field – ordisciplines (e.g., journals in mathematics, engineering and social sciencestend to have lower values than titles in life sciences), but also betweenjournals within the same subject category. For instance, basic journals tend toshow higher citation potentials than applied or clinical journals, and journalscovering emerging topics higher than periodicals in classical subjects or moregeneral journals. SNIP corrects for such differences. Its strengths andlimitations are critically discussed, and suggestions are made for furtherresearch. All empirical results are derived from Elsevier's Scopus.

Keywords

Journalmetrics; Journal impact factor; Reference practices; Database coverage; Scopus;Source normalization

6.Identifyingthe effects of co-authorship networks on the performance of scholars: Acorrelation and regression analysis of performance measures and social networkanalysis measures

Journal of Informetrics, Volume 5, Issue 4, October 2011, Pages 594-607

Abbasi,Alireza; Altmann, Jörn; Hossain, Liaquat

 Cited by Scopus (24)

Highlights

►We use SNA measures for examining the social network effect on scholar performance.

►Scholars connected to many distinct scholars have a good citation-based performance.

►Scholars with high average number of repeated co-authorships show a goodg-index.

►Scholars with links to only one co-author of a linked co-author group perform well.

►Scholars should work with many students instead of other well-performing scholars.

Abstract

Inthis study, we develop a theoretical model based on social network theories andanalytical methods for exploring collaboration (co-authorship) networks ofscholars. We use measures from social network analysis (SNA) (i.e., normalizeddegree centrality, normalized closeness centrality, normalized betweenness centrality,normalized eigenvector centrality, average ties strength, and efficiency) forexamining the effect of social networks on the (citation-based) performance ofscholars in a given discipline (i.e., information systems). Results from ourstatistical analysis using a Poisson regression model suggest that researchperformance of scholars (g-index) is positively correlated with four SNAmeasures except for the normalized betweenness centrality and the normalizedcloseness centrality measures. Furthermore, it reveals that only normalizeddegree centrality, efficiency, and average ties strength have a positivesignificant influence on the g-index (as a performance measure). The normalizedeigenvector centrality has a negative significant influence on the g-index.Based on these results, we can imply that scholars, who are connected to manydistinct scholars, have a better citation-based performance (g-index) thanscholars with fewer connections. Additionally, scholars with large average tiesstrengths (i.e., repeated co-authorships) show a better research performancethan those with low tie strengths (e.g., single co-authorships with manydifferent scholars). The results related to efficiency show that scholars, whomaintain a strong co-authorship relationship to only one co-author of a groupof linked co-authors, perform better than those researchers with manyrelationships to the same group of linked co-authors. The negative effect ofthe normalized eigenvector suggests that scholars should work with many studentsinstead of other well-performing scholars. Consequently, we can state that theprofessional social network of researchers can be used to predict the futureperformance of researchers.

Keywords

Collaboration;Citation-based research performance; g-index; Co-authorship networks; Socialnetwork analysis measures; Regression; Correlation

7.Anew approach to the metric of journals' scientific prestige: The SJR indicator

Journal of Informetrics, Volume 4, Issue 3, July 2010, Pages 379-391

González-Pereira,Borja; Guerrero-Bote, Vicente P.; Moya-Anegón, Félix

 Cited by Scopus (57)

Abstract

Asize-independent indicator of journals’ scientific prestige, the SCImagoJournal Rank (SJR) indicator, is proposed that ranks scholarly journals basedon citation weighting schemes and eigenvector centrality. It is designed foruse with complex and heterogeneous citation networks such as Scopus. Itscomputation method is described, and the results of its implementation on theScopus 2007 dataset is compared with those of an ad hoc Journal Impact Factor,JIF(3y), both generally and within specific scientific areas. Both the SJRindicator and the JIF distributions were found to fit well to a logarithmiclaw. While the two metrics were strongly correlated, there were also majorchanges in rank. In addition, two general characteristics were observed. On theone hand, journals’ scientific influence or prestige as computed by the SJRindicator tended to be concentrated in fewer journals than the quantity ofcitation measured by JIF(3y). And on the other, the distance between thetop-ranked journals and the rest tended to be greater in the SJR ranking thanin that of the JIF(3y), while the separation between the middle and lowerranked journals tended to be smaller.

Keywords

SJRindicator; Academic journals; Journal prestige; Eigenvector centrality;Citation networks

8.Approaches to understanding andmeasuring interdisciplinary scientific research (IDR): A review of theliterature

Journal of Informetrics, Volume 5, Issue 1, January 2011, Pages 14-26

Wagner,Caroline S.; Roessner, J. David; Bobb, Kamau; Klein, Julie Thompson; Boyack,Kevin W.; Keyton, Joann; Rafols, Ismael; Börner, Katy

 Cited by Scopus (59)

Abstract

Interdisciplinaryscientific research (IDR) extends and challenges the study of science on anumber of fronts, including creating output science and engineering (S&E)indicators. This literature review began with a narrow search for quantitativemeasures of the output of IDR that could contribute to indicators, but theauthors expanded the scope of the review as it became clear that differingdefinitions, assessment tools, evaluation processes, and measures all shedlight on different aspects of IDR. Key among these broader aspects is (a) theimportance of incorporating the concept of knowledge integration, and (b)recognizing that integration can occur within a single mind as well as among ateam. Existing output measures alone cannot adequately capture this process.Among the quantitative measures considered, bibliometrics (co-authorships,co-inventors, collaborations, references, citations and co-citations) are themost developed, but leave considerable gaps in understanding of the socialdynamics that lead to knowledge integration. Emerging measures in networkdynamics (particularly betweenness centrality and diversity), and entropy arepromising as indicators, but their use requires sophisticated interpretations.Combinations of quantitative measures and qualitative assessments being appliedwithin evaluation studies appear to reveal IDR processes but carry burdens ofexpense, intrusion, and lack of reproducibility year-upon-year. This review isa first step toward providing a more holistic view of measuring IDR, althoughresearch and development is needed before metrics can adequately reflect theactual phenomenon of IDR.

Keywords

Interdisciplinary;Science; Research; Indicators; Bibliometrics; Evaluation

9.Informetrics at the beginningof the 21st century-A review • Review article

Journal of Informetrics, Volume 2, Issue 1, January 2008, Pages 1-52

Bar-Ilan,Judit

 Cited by Scopus (78)

Abstract

Thispaper reviews developments in informetrics between 2000 and 2006. At thebeginning of the 21st century we witness considerable growth in webometrics,mapping and visualization and open access. A new topic is comparison betweencitation databases, as a result of the introduction of two new citationdatabases Scopus and Google Scholar. There is renewed interest in indicators asa result of the introduction of the h-index. Traditional topics like citationanalysis and informetric theory also continue to develop. The impact factordebate, especially outside the informetric literature continues to thrive.Ranked lists (of journal, highly cited papers or of educational institutions)are of great public interest.

Keywords

Informetrics;Bibliometrics; Scientometrics; Webometrics

10.   Quantifying the interdisciplinarity of scientific journals andfields

Journal of Informetrics, Volume 7, Issue 2, Pages 469-477

Silva,F.N.; Rodrigues, F.A.; Oliveira, O.N.; da F. Costa, L.

 Cited by Scopus (1)

Highlights

►The use of temporal journal citation networks provides very good insights aboutthe evolution of scientific fields.

►The introduced entropy-based measurement of interdisciplinarity correlates wellwith impact factor and citation count.

►It also confirms quantitatively that science is becoming moreinterdisciplinary.

Abstract

Thereis an overall perception of increased interdisciplinarity in science, but thisis difficult to confirm quantitatively owing to the lack of adequate methods toevaluate subjective phenomena. This is no different from the difficulties inestablishing quantitative relationships in human and social sciences. In thispaper we quantified the interdisciplinarity of scientific journals and sciencefields by using an entropy measurement based on the diversity of the subjectcategories of journals citing a specific journal. The methodology consisted inbuilding citation networks using the Journal Citation Reports® database, inwhich the nodes were journals and edges were established based on citationsamong journals. The overall network for the 11-year period (1999–2009) studiedwas small-world and followed a power-law with exponential cutoff distributionwith regard to the in-strength. Upon visualizing the network topology anoverall structure of the various science fields could be inferred, especiallytheir interconnections. We confirmed quantitatively that science fields arebecoming increasingly interdisciplinary, with the degree of interdisplinarity(i.e. entropy) correlating strongly with the in-strength of journals and withthe impact factor.

Keywords

Scientometry;Complex networks; Subjective phenomena quantification; Interdisciplinarity;Citation networks

11. A general method forgenerating parametric Lorenz and Leimkuhler curves

Journalof Informetrics, Volume 4, Issue 4, October 2010, Pages 524-539

Sarabia,José María; Gómez-Déniz, Emilio; Sarabia, María; Prieto, Faustino

 Cited by Scopus (3)

Abstract

LetL₀ consider an initial Lorenz curve. In this paper we propose ageneral methodology for obtaining new classes of parametric Lorenz orLeimkuhler curves that contain the original curve as limiting or special case.The new classes introduce additional parameters in the original family,providing more flexibility for the new families. The new classes are built froman ordered sequence of power Lorenz curves, assuming that the powers aredistributed according to some convenient discrete random variable. Using thismethod we obtain many of the families proposed in the literature, including theclassical proposal of Bradford, 1934 and Kakwani and Podder, 1973 and others.We obtain some inequality measures and population functions for the proposedfamilies.

Keywords

Productivity;Cumulative distribution function; Probability generating function; Gini index

12.Some modifications to the SNIPjournal impact indicator

Journal of Informetrics, Volume 7, Issue 2, Pages 272-285

Waltman,Ludo; van Eck, Nees Jan; van Leeuwen, Thed N.; Visser, Martijn S.

 Cited by Scopus (7)

Highlights

►A number of modifications are explained that were recently made to the SNIPjournal impact indicator.

►Some counterintuitive properties of the original SNIP indicator are discussed.

►The revised SNIP indicator is shown not to have these properties.

►Empirically, the differences between the original and the revised SNIPindicator are relatively small.

Abstract

TheSNIP (source normalized impact per paper) indicator is an indicator of thecitation impact of scientific journals. The indicator, introduced by Henk Moedin 2010, is included in Elsevier's Scopus database. The SNIP indicator uses asource normalized approach to correct for differences in citation practicesbetween scientific fields. The strength of this approach is that it does notrequire a field classification system in which the boundaries of fields areexplicitly defined.

Inthis paper, a number of modifications that were recently made to the SNIPindicator are explained, and the advantages of the resulting revised SNIPindicator are pointed out. It is argued that the original SNIP indicator hassome counterintuitive properties, and it is shown mathematically that therevised SNIP indicator does not have these properties. Empirically, thedifferences between the original SNIP indicator and the revised one turn out tobe relatively small, although some systematic differences can be observed.Relations with other source normalized indicators proposed in the literature arediscussed as well.

Keywords

Journalimpact; Journal indicator; Scopus; Source normalization; SNIP

13.   A longitudinal comparison of citation rates and growth among openaccess journals

Journal of Informetrics, Volume 7, Issue 3, July 2013, Pages 642-650

Solomon,D.J.; Laakso, M.; Bjork, B.C.

 Cited by Scopus (2)

Highlights

•The number of OAjournals/articles published in the Scopus reached 11%/8% by 2010.

•6.4% of Scopusjournals are subscription journals that have made at least their digitalversions OA.

•Citation rates for APCfunded OA journals equaled subscription journals by 2005.

Abstract

Thestudy documents the growth in the number of journals and articles along withthe increase in normalized citation rates of open access (OA) journals listedin the Scopus bibliographic database between 1999 and 2010. Longitudinalstatistics on growth in journals/articles and citation rates are broken down byfunding model, discipline, and whether the journal was launched or hadconverted to OA. The data were retrieved from the websites of SCIMago Journaland Country Rank (journal/article counts), JournalM3trics (SNIP2 values),Scopus (journal discipline) and Directory of Open Access Journals (DOAJ) (OAand funding status). OA journals/articles have grown much faster thansubscription journals but still make up less that 12% of the journals inScopus. Two-year citation averages for journals funded by Article ProcessingCharges (APCs) have reached the same level as subscription journals. Citationaverages of OA journals funded by other means continue to lag well behind OA journalsfunded by APCs and subscription journals. We hypothesize this is less an issueof quality than due to the fact that such journals are commonly published inlanguages other than English and tend to be located outside the four majorpublishing countries.

Keywords

Openaccess; Citation rate; Scopus; Article Processing Charge

14. h Index: A review focused in its variants, computation andstandardization for different scientific fields • Review article

Journal of Informetrics, Volume 3, Issue 4, October 2009, Pages 273-289

Alonso,S.; Cabrerizo, F.J.; Herrera-Viedma, E.; Herrera, F.

 Cited by Scopus (112)

Abstract

Theh-index and some related bibliometric indices have received a lot of attentionfrom the scientific community in the last few years due to some of their goodproperties (easiness of computation, balance between quantity of publicationsand their impact and so on). Many different indicators have been developed inorder to extend and overcome the drawbacks of the original Hirsch proposal. Inthis contribution we present a comprehensive review on the h-index and relatedindicators field. From the initial h-index proposal we study their mainadvantages, drawbacks and the main applications that we can find in theliterature. A description of many of the h-related indices that have beendeveloped along with their main characteristics and some of the works thatanalyze and compare them are presented. We also review the most up to datestandardization studies that allow a fair comparison by means of the h-indexamong scientists from different research areas and finally, some works thatanalyze the computation of the h-index and related indices by using differentcitation databases (ISI Citation Indexes, Google Scholar and Scopus) areintroduced.

Keywords

h-Index;Bibliometric indicators

15.   A bibliometric analysis of academic publication and NIH funding

Journal of Informetrics, Volume 7, Issue 2, Pages 318-324

Yang,Jiansheng; Vannier, Michael W.; Wang, Fang; Deng, Yan; Ou, Fengrong; Bennett,James; Liu, Yang; Wang, Ge

 Cited by Scopus (1)

Highlights

►We present an axiomatic approach for co-authors’ credit-sharing and associatedbibliometric measures.

►We revisit a recent study published in Science on the relation between NIHawards and race/ethnicity.

►In contrast to the Science paper, our results suggest that there is nosignificant racial bias in the NIH review.

►Our axiomatic approach has a potential to be widely used for scientificassessment and management.

Abstract

Academicproductivity and research funding have been hot topics in biomedical research.While publications and their citations are popular indicators of academicproductivity, there has been no rigorous way to quantify co-authors’ relativecontributions. This has seriously compromised quantitative studies on therelationship between academic productivity and research funding. Here we applyan axiomatic approach and associated bibliometric measures to revisit a recentstudy by Ginther et al. (Ginther et al., 2011a and Ginther et al., 2011b) in whichthe probability of receiving a U.S. National Institutes of Health (NIH) R01award was analyzed with respect to the applicant's race/ethnicity. Our resultsprovide new insight and suggest that there is no significant racial bias in theNIH review process, in contrast to the conclusion from the study by D. K.Ginther et al. Our axiomatic approach has a potential to be widely used forscientific assessment and management.

Keywords

Bibliometricanalysis; Citation analysis; h-Index; a-Index; Research funding; Race/ethnicity

16.   Estimating the diffusion models of crisisinformation in micro blog

Journal of Informetrics, Volume 6, Issue 4, October 2012, Pages 600-610

Wei,Jiuchang; Bu, Bing; Liang, Liang

 Cited by Scopus (1)

Highlights

►Three information release patterns in micro blog are proposed according to theduration of crisis information released.

►Three respective diffusion models are constructed using the Logistic function.

►The diffusion of crisis information can be described by Logistic model, thegrowth curve of NMCI is S-shaped.

Abstract

Thestudy tries to construct the diffusion models of crisis information in microblog. We propose three information release patterns in micro blog according tothe duration of crisis information released, namely concentrated release, continuousrelease, and pulse release. Based on Logistic function, three respectivediffusion models are constructed. We choose three crisis events to test thediffusion models using the variables of the number of micro blogs with thecrisis information (NMCI) and the increment of NMCI. The estimate results showthat the diffusion of crisis information in micro blogs can be described byLogistic function, and the growth curve of NMCI is S-shaped.

Keywords

Diffusionmodel; Crisis information; Micro blog; Logistic model

17.Communication network dynamicsduring organizational crisis

Journal of Informetrics, Volume 7, Issue 1, January 2013, Pages 16-35

Hossain,Liaquat; Murshed, Shahriar Tanvir; Uddin, Shahadat

Highlights

►Investigate patterns of changing communications structure associated withorganizational crisis using communications dataset.

►Observe that reciprocity within the organisational communication networkincreases during crisis period.

►Organizational communication network becomes less transitive as organizationsexperience crisis.

►Number of cliques increases in a communication network as organizations aregoing through crisis.

►Communication network becomes increasingly centralised as organizations gothrough crisis.

Abstract

Communicationnetwork is a personal or professional set of relationships between individualsor organizations. In other words, it is a pattern of contacts which are createddue to the flow of information among the participating actors. The flow ofinformation establishes various types of relationships among the participatingentities. These relationships eventually form an overall pattern that couldform a gestalt of the total structure within organizational context. In thispaper, we analyze the changing communications structure in order to investigatethe patterns associated with the final stages of organizational crisis.Organizational crisis has been defined as organizational mortality,organizational death, organizational exit, bankruptcy, decline, retrenchmentand failure to characterize various forms of organizational crisis. We draw ontheoretical perspectives on organizational crisis proposed by social networkanalysts and other sociologists to test 5 key propositions on the changes inthe network communication structure associated with organizational crisis: (1)a few actors, who are prominent or more active, will become central during theorganizational crisis period; (2) reciprocity within the organizationalcommunication network will increase during crisis period; (3) organizationalcommunication network becomes less transitive as organizations experiencecrisis; (4) number of cliques increases in a communication network asorganizations are going through crisis; and (5) communication network becomesincreasingly centralized as organizations go through crisis.

Keywords

Communicationnetwork; Organizational crisis; Social network analysis; Actor centrality;Reciprocity; Transitivity; Cliques; Centralization

18.   Web traffic and organization performancemeasures: Relationships and data sources examined

Journal of Informetrics, Volume 7, Issue 3, July 2013, Pages 699-711

Vaughan,L.; Yang, R.

Highlights

•Found significantcorrelations between Web traffic and academic/business performance.

•Alexa Internet is thebest among the three Web traffic data sources examined.

•Traffic data from thethree sources are highly correlated.

Abstract

UnlikeWeb hyperlink data, Web traffic data have not yet been the focus ofconsiderable study in Webometrics research. The relationships between Webtraffic data and academic/business performance measures have not been as firmlyestablished as the relationships between Web hyperlink data and suchperformance measures. Although various traffic data sources exist, few studieshave examined and compared their relative merits. We carried out a study thataimed to address this lack. We selected groups of universities and businessesfrom the U.S. and China and collected their Web traffic data from threesources: Alexa Internet, Google Trends for Websites, and Compete. We foundsignificant correlations between Web traffic data and organizationalperformance measures, specifically academic quality for universities andfinancial variables for businesses. We also examined the characteristics of thethree data sources and compared their usefulness. We found that Alexa Internetoutperformed the others.

Keywords

Webtraffic analysis; Webometrics; Performance measurements; Data qualityassessment

19.   Everything is plentiful-Except attention. Attention data ofscientific journals on social web tools

Journal of Informetrics, Volume 6, Issue 4, October 2012, Pages 661-668

Kortelainen,Terttu; Katvala, Mari

Highlights

►Scientific communication is studied from attention economy viewpoint.

►The social media tools have a clear role of their own in scientificcommunication.

►Attention data concerning published items of scientific journals are compiledfrom social media.

Abstract

Onehundred scientific and scholarly journal web sites were investigated to findout their use of social media tools and to examine attention data revealed bythem. Seventy-eight scientific journals used social media tools, RSS being themost common. Interactive social media tools – Facebook, Twitter and blogs –were present on 19 journal web sites. Attention data were operationalised asliking, commenting or sharing postings on Facebook, Twitter or blog texts orlinking to articles, liking a YouTube entry or following a journal on Twitter.Facebook and blog sites of the journals had varying roles with respect tocontent generated by readers and the journal, and the amount of attention datareceived by the journals’ Facebook, Twitter and blog sites also showed greatvariation. In scientific communication, social media have a role of their own,complementing that of scientific journals, and their active use indicates theclear demand for them. Attention is difficult to measure also by social media,but their interactive features obviously indicate one part of it, and attentioneconomy presents a fruitful viewpoint for studying scientific communication byproviding relevant and useful concepts that describe its characteristics andfactors that influence the attention it receives.

Keywords

Attention;Scientific publishing; Social media; Facebook; Twitter

20.   Scientific collaboration and endorsement: Network analysis ofcoauthorship and citation networks

Journal of Informetrics, Volume 5, Issue 1, January 2011, Pages 187-203

Ding,Ying

 Cited by Scopus (28)

Research highlights

▶This study applied the combination of a topic modeling algorithm and apath-finding algorithm to mine research topics of scientists based on theirpublications and identified their semantic associations based on coauthorshipnetworks and author citation networks.

▶This paper was able to address the collaboration patterns and citation pattersat the topic level rather than at the domain/disciplinary level.

Abstract

Scientificcollaboration and endorsement are well-established research topics whichutilize three kinds of methods: survey/questionnaire, bibliometrics, andcomplex network analysis. This paper combines topic modeling and path-findingalgorithms to determine whether productive authors tend to collaborate with orcite researchers with the same or different interests, and whether highly citedauthors tend to collaborate with or cite each other. Taking informationretrieval as a test field, the results show that productive authors tend todirectly coauthor with and closely cite colleagues sharing the same researchinterests; they do not generally collaborate directly with colleagues havingdifferent research topics, but instead directly or indirectly cite them; andhighly cited authors do not generally coauthor with each other, but closelycite each other.

Keywords

Scientificcollaboration; Scientific endorsement; Topic modeling; Path-finding algorithm

21.   Community detection: Topological vs. topical

Journal of Informetrics, Volume 5, Issue 4, October 2011, Pages 498-514

Ding,Ying

 Cited by Scopus (8)

Highlights

►This paper applied the topology-based and the topic-based community detectionapproaches to the coauthorship networks of the information retrieval.

►The results are consistent with the hypothesis 1: Communities detected by thetopology-based community detection approaches tend to contain topically-diversesub-communities within each community.

►The results are consistent with the hypothesis 2: Communities detected by the topic-basedcommunity detection approaches tend to contain topologically-diversesub-communities within each community.

►It proposes that community detection should consider both the topological andtopical features of the networks.

Abstract

Theevolution of the Web has promoted a growing interest in social networkanalysis, such as community detection. Among many different community detectionapproaches, there are two kinds that we want to address: one considers thegraph structure of the network (topology-based community detection approach);the other one takes the textual information of the network nodes intoconsideration (topic-based community detection approach). This paper conductedsystematic analysis of applying a topology-based community detection approachand a topic-based community detection approach to the coauthorship networks ofthe information retrieval area and found that: (1) communities detected by thetopology-based community detection approach tend to contain different topicswithin each community; and (2) communities detected by the topic-basedcommunity detection approach tend to contain topologically-diversesub-communities within each community. The future community detectionapproaches should not only emphasize the relationship between communities and topics,but also consider the dynamic changes of communities and topics.

Keywords

Communitydetection; Topics; Communities; Coauthor network

22.   The distribution of references across texts: Some implications forcitation analysis

Journal of Informetrics, Volume 7, Issue 3, July 2013, Pages 583-592

Ding,Y.; Liu, X.; Guo, C.; Cronin, B.

 Cited by Scopus (1)

Highlights

•This paper applied thecontent-based citation analysis on the large scale data (866 full-text JASISTpapers) with the focus on the location of the references and how many times onereference appearing in one citing article.

•It compared thedifference between two counting methods: CountOne vs. CountX.

•It also compared thelocation of the references in different sections of one citing article.

•The major findingsare: (1) the most highly cited works appear in the Introduction and Literature Review sections of citing papers, and (2) the citation rankings produced by Count One and Count X differ.

Abstract

Incitation network analysis, complex behavior is reduced to a simple edge,namely, node A cites node B. The implicit assumption is that A is giving creditto, or acknowledging, B. It is also the case that the contributions of allcitations are treated equally, even though some citations appear multiply in atext and others appear only once. In this study, we apply text-miningalgorithms to a relatively large dataset (866 information science articlescontaining 32,496 bibliographic references) to demonstrate the differentialcontributions made by references. We (1) look at the placement of citationsacross the different sections of a journal article, and (2) identify highlycited works using two different counting methods (CountOne and CountX). We findthat (1) the most highly cited works appear in the Introduction and LiteratureReview sections of citing papers, and (2) the citation rankings produced byCountOne and CountX differ. That is to say, counting the number of times abibliographic reference is cited in a paper rather than treating all referencesthe same no matter how many times they are invoked in the citing articlereveals the differential contributions made by the cited works to the citingpaper.

Keywords

Content-basedcitation analysis; Citation; Mentioning; Citation analysis

23.   Do open access articles have greater citation impact? • Reviewarticle

Journal of Informetrics, Volume 1, Issue 3, July 2007, Pages 239-248

Craig,Iain D.; Plume, Andrew M.; McVeigh, Marie E.; Pringle, James; Amin, Mayur

 Cited by Scopus (85)

Abstract

The last few years have seen the emergence of several open access options inscholarly communication which can broadly be grouped into two areas referred toas ‘gold’ and ‘green’ open access (OA). In this article we review the literatureexamining the relationship between OA status and citation counts of scholarlyarticles. Early studies showed a correlation between the free onlineavailability or OA status of articles and higher citation counts, and impliedcausality without due consideration of potential confounding factors. Morerecent investigations have dissected the nature of the relationship betweenarticle OA status and citations. Three non-exclusive postulates have beenproposed to account for the observed citation differences between OA and non-OAarticles: an open access postulate, a selection bias postulate, and an earlyview postulate. The most rigorous study to date (in condensed matter physics)showed that, after controlling for the early view postulate, the remaining differencein citation counts between OA and non-OA articles is explained by the selectionbias postulate. No evidence was found to support the OA postulate per se; i.e.article OA status alone has little or no effect on citations. Further studiesusing a similarly rigorous approach are required to determine the generality ofthis finding.

Keywords

Openaccess; Citation analysis; Early view; Quality bias

24.   Counting publications and citations: Is more always better?

Journal of Informetrics, Volume 7, Issue 3, July 2013, Pages 635-641

Waltman,L.; van Eck, N.J.; Wouters, P.

 Cited by Scopus (1)

Highlights

•Is more always better when counting highly cited publications?

•A model of the relationship between scientific impact and citations is introduced.

•Having more highly cited publications need not coincide with having more impact.

•An improved way of counting highly cited publications is suggested.

Abstract

Ismore always better? We address this question in the context of bibliometricindices that aim to assess the scientific impact of individual researchers bycounting their number of highly cited publications. We propose a simple modelin which the number of citations of a publication depends not only on thescientific impact of the publication but also on other ‘random’ factors. Ourmodel indicates that more need not always be better. It turns out that the mostinfluential researchers may have a systematically lower performance, in termsof highly cited publications, than some of their less influential colleagues.The model also suggests an improved way of counting highly cited publications.

Keywords

Bibliometricindex; Citation; Highly cited publication; Modeling; Scientific impact

25.   Finding scientific gems with Google's PageRank algorithm

Journal of Informetrics, Volume 1, Issue 1, January 2007, Pages 8-15

Chen,P.; Xie, H.; Maslov, S.; Redner, S.

 Cited by Scopus (93)

Abstract

We apply the Google PageRank algorithm to assess the relative importance of allpublications in the Physical Review family of journals from 1893 to 2003. Whilethe Google number and the number of citations for each publication arepositively correlated, outliers from this linear relation identify someexceptional papers or “gems” that are universally familiar to physicists.

Keywords

GooglePageRank algorithm; Scientific gems; Physical Review; Citations

数据来源：http://top25.sciencedirect.com/subject/decision-sciences/8/journal/journal-of-informetrics/17511577/archive/47/

检索时间：2014.1.28

注：从word粘过来格式上有问题，有兴趣可下载journal of informetrics top25.docx。