Evaluating Contrasting Research of Academic Institutions through AI and Deep Learning Model: A Use Case of Scite
Year: 2026
Type: article
Abstract: Objective The study endeavours to analyse the context of the citations using Scite. In addition to analysing citation statements related to the research productivity of selected organisations, the study also investigates the relationship between Lifetime SI and total cites, total cites and contrasting, total cites and supporting, and contrasting and supporting using Scite. Methodology The data was obtained from Scite's organisation-level citation data, including organisation names, Lifetime Scite Index (SI), total citations, supporting citations, mentioning citations, and contradicting citations. Scite is a web tool that utilises AI and deep learning models. It helps researchers to better understand the context, location, and classification of citation statements. The data was processed using Jamovi. The Pearson's correlation test was used to examine the relationship between Lifetime SI and total citations, total citations and contrasting, total citations and supporting, as well as contrasting and supporting. Findings A positive correlation was observed between lifetime SI (LSI) and total cites (TC) (r= 0.145). Contrasting (C) and Total Cites (TC) (r = 0.984), Contrasting (C) and Supporting (S) (r = 0.988), and Total Cites (TC) and Supporting (S) (r = 0.995) exhibited a strong positive correlation. A significant and strong positive relationship was found between the parameters constituting citation statements. Originality This study analyses the contextual nature of citations using Scite. Specifically, it seeks to examine citation statements related to the research productivity of selected organisations.
Language: English
Subfield: Library and Information Sciences
Field: Social Sciences
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W7220479811
