Giovanni Colavizza

Characterizing in-text citations in scientific articles : A large-scale analysis.

We report characteristics of in-text citations in over five million full text articles from two large databases – the PubMed Central Open Access subset and Elsevier journals – as functions of time, textual progression, and scientific field. The purpose of this study is to understand the characteristics of in-text citations in a detailed way prior to pursuing other studies focused on answering more substantive research questions. As such, we have analyzed in-text citations in several ways and report many findings here. Perhaps most significantly, we find that there are large field-level differences that are reflected in position within the text, citation interval (or reference age), and citation counts of references. In general, the fields of Biomedical and Health Sciences, Life and Earth Sciences, and Physical Sciences and Engineering have similar reference distributions, although they vary in their specifics. The two remaining fields, Mathematics and Computer Science and Social Science and Humanities, have different reference distributions from the other three fields and between themselves. We also show that in all fields the numbers of sentences, references, and in-text mentions per article have increased over time, and that there are field-level and temporal differences in the numbers of in-text mentions per reference. A final finding is that references mentioned only once tend to be much more highly cited than those mentioned multiple times.

The references of references: a method to enrich humanities library catalogs with citation data

The advent of large-scale citation indexes has greatly impacted the retrieval of scientific information in several domains of research. The humanities have largely remained outside of this shift, despite their increasing reliance on digital means for information seeking. Given that publications in the humanities have a longer than average life-span, mainly due to the importance of monographs for the field, this article proposes to use domain-specific reference monographs to bootstrap the enrichment of library catalogs with citation data. Reference monographs are works considered to be of particular importance in a research library setting, and likely to possess characteristic citation patterns. The article shows how to select a corpus of reference monographs, and proposes a pipeline to extract the network of publications they refer to. Results using a set of reference monographs in the domain of the history of Venice show that only 7% of extracted citations are made to publications already within the initial seed. Furthermore, the resulting citation network suggests the presence of a core set of works in the domain, cited more frequently than average.

Clustering citation histories in the Physical Review

We investigate publications through their citation histories – the history events are the citations given to the article by younger publications and the time of the event is the date of publication of the citing article. We propose a methodology, based on spectral clustering, to group citation histories, and the corresponding publications, into communities and apply multinomial logistic regression to provide the revealed communities with semantics in terms of publication features. We study the case of publications from the full Physical Review archive, covering 120 years of physics in all its domains. We discover two clear archetypes of publications – marathoners and sprinters – that deviate from the average middle-of-the-roads behaviour, and discuss some publication features, like age of references and type of publication, that are correlated with the membership of a publication into a certain community.

The Closer the Better: Similarity of Publication Pairs at Different Co-Citation Levels

We investigated the similarities of pairs of articles that are cocited at the different cocitation levels of the journal, article, section, paragraph, sentence, and bracket. Our results indicate that textual similarity, intellectual overlap (shared references), author overlap (shared authors), proximity in publication time all rise monotonically as the cocitation level gets lower (from journal to bracket). While the main gain in similarity happens when moving from journal to article cocitation, all level changes entail an increase in similarity, especially section to paragraph and paragraph to sentence/bracket levels. We compared the results from four journals over the years 2010–2015: Cell, the European Journal of Operational Research, Physics Letters B, and Research Policy, with consistent general outcomes and some interesting differences. Our findings motivate the use of granular cocitation information as defined by meaningful units of text, with implications for, among others, the elaboration of maps of science and the retrieval of scholarly literature.

The structural role of the core literature in history

The intellectual landscapes of the humanities are mostly uncharted territory. Little is known on the ways published research of humanist scholars defines areas of intellectual activity. An open question relates to the structural role of core literature: highly cited sources, naturally playing a disproportionate role in the definition of intellectual landscapes. We introduce four indicators in order to map the structural role played by core sources into connecting different areas of the intellectual landscape of citing publications (i.e. communities in the bibliographic coupling network). All indicators factor out the influence of degree distributions by internalizing a null configuration model. By considering several datasets focused on history, we show that two distinct structural actions are performed by the core literature: a global one, by connecting otherwise separated communities in the landscape, or a local one, by rising connectivity within communities. In our study, the global action is mainly performed by small sets of scholarly monographs, reference works and primary sources, while the rest of the core, and especially most journal articles, acts mostly locally.

TimeRank: A dynamic approach to rate scholars using citations

Rating has become a common practice of modern science. No rating system can be considered as final, but instead several approaches can be taken, which magnify different aspects of the fabric of science. We introduce an approach for rating scholars which uses citations in a dynamic fashion, allocating ratings by considering the relative position of two authors at the time of the citation among them. Our main goal is to introduce the notion of citation timing as a complement to the usual suspects of popularity and prestige. We aim to produce a rating able to account for a variety of interesting phenomena, such as positioning raising stars on a more even footing with established researchers. We apply our method on the bibliometrics community using data from the Web of Science from 2000 to 2016, showing how the dynamic method is more effective than alternatives in this respect.

The Core Literature of the Historians of Venice

Over the past decades the humanities have been accumulating a growing body of literature at an increasing pace. How does this impact their traditional organization into disciplines and fields of research therein? This article considers history, by examining a citation network among recent monographs on the history of Venice. The resulting network is almost connected, clusters of monographs are identifiable according to specific disciplinary areas (history, history of architecture, history of arts) or periods of time (middle ages, early modern, modern history), and a map of the recent trends in the field is sketched. Most notably a set of highly-cited works emerges as the core literature of the historians of Venice. This core literature comprises a mix of primary sources, works of reference and scholarly monographs, and is important in keeping the field connected: monographs usually cite a combination of few core and a variety of less well-cited works. Core primary sources and works of reference never age, while core scholarly monographs are replaced at a very slow rate by new ones. The reliance of new publications on the core literature is slowly rising over time, as the field gets increasingly more varied.

Mapping the Early Modern News Flow: An Enquiry by Robust Text Reuse Detection

Early modern printed gazettes relied on a system of news exchange and text reuse largely based on handwritten sources. The reconstruction of this information exchange system is possible by detecting reused texts. We present a method to individuate text borrowings within noisy OCRed texts from printed gazettes based on string kernels and local text alignment. We apply our methods on a corpus of Italian gazettes for the year 1648. Beside unveiling substantial overlaps in news sources, we are able to assess the editorial policy of different gazettes and account for a multi-faceted system of text reuse.

The Labeling System: A New Approach to Overcome the Vocabulary Bottleneck

Shared controlled vocabularies are a prerequisite for collaborative annotation and semantic interchange. The creation and maintenance of such vocabularies is, however, time-consuming and expensive. The diversity of research questions in the humanities makes it virtually impossible to create shared controlled vocabularies that cover a wide range of potential applications and satisfy the needs of diverse stakeholders. In this paper we present a novel conceptual approach for mitigating these problems. We propose that projects define their own vocabularies as needed and link the vocabulary terms to one or more concepts in a reference thesaurus, so that the project-specific term effectively serves as a label for a set of shared concepts. We also describe the implementation of this approach in the Labeling System. The Labeling System is a Web application that allows users to easily import concepts or create SKOS vocabularies and link the vocabulary terms to concepts from one or more reference thesauri.

A diachronic study of historiography

The humanities are often characterized by sociologists as having a low mutual dependence among scholars and high task uncertainty. According to Fuchs’ theory of scientific change, this leads over time to intellectual and social fragmentation, as new scholarship accumulates in the absence of shared unifying theories. We consider here a set of specialisms in the discipline of history and measure the connectivity properties of their bibliographic coupling networks over time, in order to assess whether fragmentation is indeed occurring. We construct networks using both reference overlap and textual similarity. It is shown that the connectivity of reference overlap networks is gradually and steadily declining over time, whilst that of textual similarity networks is stable. Author bibliographic coupling networks also show signs of a decline in connectivity, in the absence of an increasing propensity for collaborations. We speculate that, despite the gradual weakening of ties among historians as mapped by references, new scholarship might be continually integrated through shared vocabularies and narratives. This would support our belief that citations are but one kind of bibliometric data to consider—perhaps even of secondary importance—when studying the humanities, while text should play a more prominent role.