We investigate the comparison of two bibliometric approaches to measure the origin of science: one is disruption indexes based on bibliographic coupling analysis, the other is seed algorithm based on co-citation analysis. In particular, we compare the publication characteristics across the stages of in-process-out for disruptive discoveries and scientific ‘seeds’ across different subjects, and explore their relationships. Datasets from Web of Science core collection, consisting of 2,922,689 papers across six subjects from Arts & Humanities, Life Sciences & Biomedicine, Physical Sciences, Social Sciences, Technology, are analyzed using seed algorithm and disruption index family. Our results show that stem papers belong to highly cited papers. We also find that both algorithms have different strengths in identifying disruptive discoveries. Disruption index family can measure the disruption between focal papers and their knowledge base, but it is difficult to ensure that highly disruptive papers receive high numbers of citations, while the 'seeds’ overcome the above shortcomings but lose the degree of distinction between focal papers and their references. Besides, our results suggest that, stem papers with higher h-index across different subjects tend to cite fewer references and involve smaller affiliations and countries, which is contrary to top 10 disruptive papers with high citation, but both of them receive higher citations. Specifically, both of them in Computer Science are characterized by high disruption and citations, along with a central role in the scientific community.
This repository provide easy access to supplemental information in this study.