Clarivate Analytics’s Web of Science (WoS) is the world’s leading scientific citation search and analytical information platform. It used as both a research tool supporting broad array tasks across diverse knowledge domains well dataset for large-scale data-intensive studies. WoS has been in thousands published academic studies over past 20 years. also most enduring commercial legacy Eugene Garfield. Despite central position holds contemporary research, quantitative impact not previously...

Abstract This paper studies how varied damping factors in the PageRank algorithm influence ranking of authors and proposes weighted algorithms. We selected 108 most highly cited information retrieval (IR) area from 1970s to 2008 form author co‐citation network. calculated ranks these based on with factor ranging 0.05 0.95. In order test relationship between different measures, we compared results citation ranking, h‐index, centrality measures. found that our network, rank is correlated also...

Abstract Many studies on coauthorship networks focus network topology and statistical mechanics. This article takes a different approach by studying micro‐level properties with the aim of applying centrality measures to impact analysis. Using data from 16 journals in field library information science (LIS) time span 20 years (1988–2007), we construct an evolving calculate four (closeness centrality, betweenness degree PageRank) for authors this network. We find that are significantly...

This study explores the similarity among six types of scholarly networks aggregated at institution level, including bibliographic coupling networks, citation cocitation topical coauthorship and coword networks. Cosine distance is chosen to measure similarities The authors found that have lowest similarity; high similarity. In addition, through multidimensional scaling, two dimensions can be identified networks: D imension 1 interpreted as citation‐based versus noncitation‐based, 2 social...

Abstract This study comprises a suite of analyses words in article titles order to reveal the cognitive structure Library and Information Science (LIS). The use title elucidate LIS has been relatively neglected. present addresses this gap by performing (a) co‐word analysis hierarchical clustering, (b) multidimensional scaling, (c) determination trends usage terms. is based on 10,344 articles published between 1988 2007 16 journals. Methodologically, novel aspects are: its large scale,...

Abstract Ranking scientific productivity and prestige are often limited to homogeneous networks. These networks unable account for the multiple factors that constitute scholarly communication reward system. This study proposes a new informetric indicator, P‐Rank, measuring in heterogeneous containing articles, authors, journals. P‐Rank differentiates weight of each citation based on its citing papers, journals, authors. Articles from 16 representative library information science journals...

This paper proposes entitymetrics to measure the impact of knowledge units. Entitymetrics highlight importance entities embedded in scientific literature for further discovery. In this paper, we use Metformin, a drug diabetes, as an example form entity-entity citation network based on related Metformin. We then calculate features and compare centrality ranks biological with results from Comparative Toxicogenomics Database (CTD). The comparison demonstrates usefulness detect most outstanding...

This paper empirically studies the effect of Open Access on journal CiteScores. We have found that general is positive but not uniform across different types journals. In particular, we investigate two heterogeneous treatment effect: (1) differential among journals grouped by academic field, publisher, and tier; (2) effects as a function propensity to be treated. The results are robust number sensitivity checks falsification tests. Our findings shed new light can help stakeholders in...

Understanding semantic relatedness and similarity between biomedical terms has a great impact on variety of applications such as information retrieval, extraction, recommender systems. The objective this study is to examine word2vec's ability in deriving from large publication data. Specifically, we focus the effects recency, size, section data performance word2vec.We download abstracts 18,777,129 articles PubMed 766,326 full-text Central (PMC). datasets are preprocessed grouped into subsets...

We propose using the technique of weighted citation to measure an article's prestige. The allocates a different weight each reference by taking into account impact citing journals and time intervals. Weighted captures prestige, whereas counts capture popularity. compare value variances for popularity prestige articles published in Journal American Society Information Science Technology from 1998 2007, find that majority have comparable status.

The field of LIS continues to face a vexing paradox. Its longstanding ideal and concomitant commitment serving diverse communities users equally has failed translate into diversity, equity, inclusion (DEI) in the profession or education. This article analyzes efforts promote North American programs between 1971 2018. First, it addresses change continuity diversity-centered scholarship over time. Second, unpacks key terms such as inclusion. Third, underlines importance mandate for adducing...

Abstract The authors propose using the technique of weighted citation to measure an article's prestige. allocates a different weight each reference by taking into account impact citing journals and time intervals. Weightedcitation captures prestige, whereas counts capture popularity. They compare value variances for popularity prestige articles published in Journal American Society Information Science Technology from 1998 2007, find that majority have comparable status.

Exploring community is fundamental for uncovering the connections between structure and function of complex networks practical applications in many disciplines such as biology sociology. In this paper, we propose a TTR-LDA-Community model which combines Latent Dirichlet Allocation (LDA) Girvan-Newman detection algorithm with an inference mechanism. The then applied to data from Delicious, popular social tagging system, over time period 2005-2008. Our results show that 1) users same tend be...

Two commonly used ideas in the development of citation-based research performance indicators are idea normalizing citation counts based on a field classification scheme and recursive weighing (like PageRank-inspired indicators). We combine these two single indicator, referred to as mean normalized score we study validity this indicator. Our empirical analysis shows that proposed indicator is highly sensitive used. The also has strong tendency reinforce biases caused by scheme. Based...