This paper considers the status of information science as science through an exploration ofone of the leading journals in the field – the Journal of the American Society for InformationScience (JASIS) from its initial publication as American Documentation (AD) in 1950 through theclosing issue of its Silver Anniversary year in December 1999. It is a bibliometric examination ofAD/JASIS articles. Based on our analysis of articles published in AD and JASIS from 1950 to1999, we find that there has been a slow but perhaps inevitable shift based first on the single nonfundedresearcher and author to a much wider research and publishing participation amongauthors, regions, corporate authors, and countries. This suggests not only cross-fertilization ofideas, but also more complex research questions. A small trend toward greater external fundingfurther reinforces this hypothesis. Information may no longer be "little" science, but it is also not"big" science.

Information science as "Little Science":The implications of a bibliometric analysis of theJournal of the American Society for Information Science

This paper covers a total of 82 bibliometric studies on single journals (62 studies cover unique titles) published between 1998 and 2008 grouped into the following fields; Arts, Humanities and Social Sciences (12 items); Medical and Health Sciences (19 items); Sciences and Technology (30 items) and Library and Information Sciences (21 items). Under each field the studies are described in accordance to their geographical location in the following order, United Kingdom, United States and Americana, Europe, Asia (India, Africa and Malaysia). For each study, elements described are (a) the journal’s publication characteristics and indexation information; (b) the objectives; (c) the sampling and bibliometric measures used; and (d) the results observed. A list of journal titles studied is appended. The results show that (a)bibliometric studies cover journals in various fields; (b) there are several revisits of some journals which are considered important; (c) Asian and African contributions is high (41.4% of total studies; 43.5% covering unique titles), United States (30.4% of total; 31.0% on unique titles), Europe (18.2% of total and 14.5% on unique titles) and the United Kingdom (10% of total and 11% on unique titles); (d) a high number of bibliometrists are Indians and as such coverage of Indian journals is high (28% of total studies; 30.6% of unique titles); and (e) the quality of the journals and their importance either nationally or internationally are inferred from their indexation status.

/pdf/bibliometric-studies-on-single-journals-a-review-5by803wh37.pdf

Bibliometric studies on single journals: a review

The Journal of the American Society for Information Science (JASIS) is completing 50 years of publication. Aspects of authorship of papers in JASIS were studied by examining one volume from each decade of JASIS's existence. For each substantial paper in these volumes, data were collected regarding number of authors, type of affiliation of each author, author's gender, and author's country if it was not the United States. Also noted were data on length, content, and “colonicity” of the title, and data on the extent of citing and self-citing in the paper. Findings are presented, and are compared with findings of other studies of JASIS and related publications. Based on this survey of JASIS, the literature of information science has grown exponentially, as would be expected in a new or developing discipline. Authorship has been growing even a little faster, because multiple authorship of information science papers has become much more common. Representation of authors from different countries has increased greatly. But, compared to various library journals, JASIS is not outstanding in either multiple authorship or degree of foreign representation. Individual authors, at least in JASIS, are increasingly likely to produce multiple papers; the extent, explanation, and significance of this phenomenon warrant further inquiry. The percentage of authors who are female has grown, but is higher in many related journals than it is in JASIS. Trends in the titling of papers suggest that writings have become more informative, but also considerably wordier. “Scholarliness” of papers has increased on the basis of a rapid rise in use of colons in titles. More importantly, scholarship has increased greatly on the basis of the disappearance of papers that lack citations and the exponential growth in the average number of references per paper. It appears that the field of information science underwent an important transition in authorship characteristics after the 1950s. The proportion of authors with academic affiliations has grown so large that other types of affiliations, although significant in the 1950s, are now hardly represented at all. Contributions by authors whose professional concerns are primarily with applied aspects of information science have thus become rarer. Such changes may have serious implications for information science and for JASIS, and deserve study.

Aspects of JASIS authorship through five decades

This article reports findings from a study of patterns of foreign authorship of articles, and international composition of journal editorial boards in five leading journals in the field of information science, and scientometrics. The study covers an American journal and four European journals. Bibliographic data about foreign authors and their national affiliation from five selected years of publication were analyzed for all journals. The foreign input of articles were extremely high in Information Processing & Management, and Scientometrics, and were relatively low in the other three journals. The number of foreign countries contributing in all journals have increased rapidly since 1996. Canada, England, Belgium, Netherlands, China, and Spain were the countries with high contributions in JASIST. The authors from the USA have dominated the foreign-authored articles in all European journals. A simple linear regression analysis showed that 60% of variation in the proportion of foreign-authored articles in the set of five journals over the selected years could be explained by the percentage of foreign members on the editorial boards of the journals.

Assessing internationality of scholarly journals through foreign authorship patterns: the case of major journals in information science, and scientometrics

In academia, publication productivity, defined as the number of peer-reviewed articles published and the frequency of citations, is a primary factor in the assessment of tenure and promotion. One of the most cited gender differences in academia is the “productivity puzzle,” which suggests that women publish less than men, thereby affecting every aspect of a woman's academic career. Peer-reviewed articles published in the Annals of the Association of American Geographers (Annals) and The Professional Geographer (PG) between 1995 and 2006, and in four subdisciplinary journals between 2005 and 2009, as well as citation reports, were used to explore whether gender differences are present in publication productivity. Gender differences were evident in the proportion of women authors, the frequency of collaboration, and the number of citations across a broad range of prestigious geographic journals. For all journals studied, women were underrepresented, especially in the authorship positions that equate to noti...

/pdf/authorship-collaboration-and-gender-fifteen-years-of-4p9e7effct.pdf

Authorship, Collaboration, and Gender: Fifteen Years of Publication Productivity in Selected Geography Journals

Authorship in JASIS:A quantitative analysis

Translating training data into many languages has emerged as a practical solution for improving cross-lingual transfer. For tasks that involve span-level annotations, such as information extraction or question answering, an additional label projection step is required to map annotated spans onto the translated texts. Recently, a few efforts have utilized a simple mark-then-translate method to jointly perform translation and projection by inserting special markers around the labeled spans in the original sentence. However, as far as we are aware, no empirical analysis has been conducted on how this approach compares to traditional annotation projection based on word alignment. In this paper, we present an extensive empirical study across 57 languages and three tasks (QA, NER, and Event Extraction) to evaluate the effectiveness and limitations of both methods, filling an important gap in the literature. Experimental results show that our optimized version of mark-then-translate, which we call EasyProject, is easily applied to many languages and works surprisingly well, outperforming the more complex word alignment-based methods. We analyze several key factors that affect the end-task performance, and show EasyProject works well because it can accurately preserve label span boundaries after translation. We will publicly release all our code and data.

Frustratingly Easy Label Projection for Cross-lingual Transfer

Anaphora resolution is an important task for information extraction across a range of languages, text genres, and domains, motivating the need for methods that do not require large annotated datasets. In-context learning has emerged as a promising approach, yet there are a number of challenges in applying in-context learning to resolve anaphora. For example, encoding a single in-context demonstration that consists of: an anaphor, a paragraph-length context, and a list of corresponding antecedents, requires conditioning a language model on a long sequence of tokens, limiting the number of demonstrations per prompt. In this paper, we present MICE (Mixtures of In-Context Experts), which we demonstrate is effective for few-shot anaphora resolution in scientific protocols (Tamari et al., 2021). Given only a handful of training examples, MICE combines the predictions of hundreds of in-context experts, yielding a 30% increase in F1 score over a competitive prompt retrieval baseline. Furthermore, we show MICE can be used to train compact student models without sacrificing performance. As far as we are aware, this is the first work to present experimental results demonstrating the effectiveness of in-context learning on the task of few-shot anaphora resolution in scientific protocols.

Few-Shot Anaphora Resolution in Scientific Protocols via Mixtures of In-Context Experts

Large language models have demonstrated an emergent capability in answering knowledge intensive questions. With recent progress on web-scale visual and language pre-training, do these models also understand how to answer visual information seeking questions? To answer this question, we present InfoSeek, a Visual Question Answering dataset that focuses on asking information-seeking questions, where the information can not be answered by common sense knowledge. We perform a multi-stage human annotation to collect a natural distribution of high-quality visual information seeking question-answer pairs. We also construct a large-scale, automatically collected dataset by combining existing visual entity recognition datasets and Wikidata, which provides over one million examples for model fine-tuning and validation. Based on InfoSeek, we analyzed various pre-trained Visual QA systems to gain insights into the characteristics of different pre-trained models. Our analysis shows that it is challenging for the state-of-the-art multi-modal pre-trained models to answer visual information seeking questions, but this capability is improved through fine-tuning on the automated InfoSeek dataset. We hope our analysis paves the way to understand and develop the next generation of multi-modal pre-training.

Can Pre-trained Vision and Language Models Answer Visual Information-Seeking Questions?

We present Stanceosaurus, a new corpus of 28,033 tweets in English, Hindi, and Arabic annotated with stance towards 251 misinformation claims. As far as we are aware, it is the largest corpus annotated with stance towards misinformation claims. The claims in Stanceosaurus originate from 15 fact-checking sources that cover diverse geographical regions and cultures. Unlike existing stance datasets, we introduce a more fine-grained 5-class labeling strategy with additional subcategories to distinguish implicit stance. Pre-trained transformer-based stance classifiers that are fine-tuned on our corpus show good generalization on unseen claims and regional claims from countries outside the training data. Cross-lingual experiments demonstrate Stanceosaurus' capability of training multi-lingual models, achieving 53.1 F1 on Hindi and 50.4 F1 on Arabic without any target-language fine-tuning. Finally, we show how a domain adaptation method can be used to improve performance on Stanceosaurus using additional RumourEval-2019 data. We make Stanceosaurus publicly available to the research community and hope it will encourage further work on misinformation identification across languages and cultures.

Andrew A. Ritter

Papers

Authorship in JASIS:A quantitative analysis

Frustratingly Easy Label Projection for Cross-lingual Transfer

Few-Shot Anaphora Resolution in Scientific Protocols via Mixtures of In-Context Experts

Can Pre-trained Vision and Language Models Answer Visual Information-Seeking Questions?

Stanceosaurus: Classifying Stance Towards Multilingual Misinformation