11:00am - 11:15amID: 181
/ PS-14: 1
Topics: Research MethodsKeywords: JASIST, bibliometrics, publication trends, authorship, citations
Journal of the Association for Information Science and Technology: Analysis of Two Decades of Published Research
1Simmons University, USA; 2University of Dhaka, Bangladesh
The Journal of the Association for Information Science & Technology (JASIST) is a leading information science journal, recognized by many ranking metrics. Understanding patterns in articles published would be helpful to new researchers seeking to publish with the journal. However, the last comprehensive bibliometric analysis on JASIST is more than a decade old. The objective of this paper is to analyze the bibliographic information of full-length research articles published in JASIST during the last two decades. This includes metrics such as article count, authorship, international collaboration, citations, and topical areas. Data was collected from SCOPUS, JASIST website, and Scimago. The findings show that JASIST published 3,052 articles during 2000-2020, which got cited 180,608 times (59.18 times per article) till date. Joint authorship and international collaboration has been increasing. Authors from institutions in 70 countries have published, with most articles from USA, with authorship from China steadily increasing in recent years. The detailed findings in the paper would help information science researchers and practitioners to assess the areas of focus, and various patterns in a top journal of the field. Such historical understanding is critical in charting agendas and directions for future research.
11:15am - 11:30amID: 283
/ PS-14: 2
Topics: Library and Information ScienceKeywords: citation analysis; bibliographic data; conceptual research strategy; disciplinary structure
The Evolution of LIS Research Topics and Methods from 2006 to 2018: A Content Analysis
Emporia State University, USA
Replicating a series of studies of LIS research trends performed by Järvelin and colleagues, this content analysis systematically examines the evolution and distribution of LIS research topics and methods at six-year increments from 2006 to 2018. Bibliographic data was collected for 3422 articles published in LIS journals for the years of 2006, 2012, and 2018. Using a conceptual research strategy, the researchers identified the central research topics and research method for each article. The findings indicate a shift towards greater emphasis on Scholarly Communications/Informetrics and Information Seeking/Behavior topics, along with a reduction in Information Systems and Library and Information Service topics. Quantitative-based approaches are predominant across the three years examined, with observed growth in the usage of questionnaires and informetric methods from 2006 to 2018. These findings indicate that LIS is a dynamic discipline, with quickly shifting interests/usage of research topics and methods.
11:30am - 11:45amID: 297
/ PS-14: 3
Topics: Social Media and Social ComputingKeywords: Author Profiling, Machine Learning, Deep Learning, Pre-trained Models, Twitter
Authorship Analysis of English and Spanish Tweets
1Jordan University of Science and Technology, Kingdom of Hashemite; 2Duquesne University, USA
With the countless advantages gained from the free, open, and ubiquitous nature of online social networks, they do come with their own set of problems and challenges. E.g., they represent a fertile ground for fake accounts and autonomous bots to spread fake news. Revealing whether some text content is written by a bot or a human would be of great value in the fight against the spreading of fake news and misinformation. In this paper, we address this problem using different Machine Learning (ML) techniques: conventional, Deep Learning (DL) based and Transfer Learning (TL) based. Using the dataset of the well-known PAN 2019 Author Profiling Task, we show how relatively simple conventional ML methods can outperform DL and TL based ones for different languages (English and Spanish). In fact, our simplest model performs closely to the state-of-the-art (SOTA) systems for the English language and even outperforms the SOTA systems for the Spanish language.
11:45am - 12:00pmID: 358
/ PS-14: 4
Topics: Human Computer Interaction (HCI)Keywords: cross-session search
Exploring Factors Affecting Renewal and Stopping Reasons in Cross-Session Search
University of North Carolina at Chapel Hill, USA
This study analyzes the relationships between search session renewal reasons and stopping reasons for everyday cross-session search, and their relationships with types of needed information, the number of search sessions, and task performance stage. We present results from an online survey questionnaire distributed on the Amazon Mechanical Turk platform. Our results validate the renewal reasons from Lin and Belkin’s MISE model and generalize the stopping reasons found by other empirical studies to a broader population and tasks. We found that participants in different renewal modes more often look for specific information than general information. Sessions which happened early or at the middle of a task were often renewed because a task spawned sub-problems or transmuted. Sessions renewed near the end or after the task was done often involved looking for updated information or were renewed due to search failure.
Our study has implications for predicting reasons that can cause searchers to start or stop a search session during cross session search, and on designing search support tools to provide more targeted, customized help based on the relations among those reasons.
12:00pm - 12:10pmID: 335
/ PS-14: 5
Topics: Library and Information ScienceKeywords: Funding Prediction, Bibliometrics, Scholarly Network
Initial Bibliometric Investigation of NIH Mentored K to R Transition
1School of Information Management, Wuhan University, People's Republic of China; 2Network Science Institute, Indiana University, USA; 3Department of Research, HealthPartners Institute, USA; 4Department of Preventive Medicine, Keck School of Medicine, USCLA, USA; 5School of Informatics, Computing and Engineering, Indiana University, USA
National Institutes of Health (NIH) is the world largest public funder of biomedical research, investing more than $30 billion dollars to achieve its mission to enhance health, lengthen life, and reduce illness and disability. Here, by leveraging individual-level characteristics and contextual/time-dependent features of professional scholarly network, we investigate the chance of NIH Mentored K (MK) to NIH R01 grant (independent research grant) or equivalent (R01-Eq) transition success. The aim of this work is to explore the relationship between investigator productivity (i.e., scholarly publication) and success (e.g, R01-Eq funding) during MK to R01-Eq transition using publicly available datasets and applying our machine learning techniques. The preliminary experiment based on PubMed data and NIH awardees database show that the proposed method is promising, and a number of interesting funding success factors can be located by utilizing statistical tools.