Home |  DEUTSCH |  Contact |  Imprint |  Data Protection |  Login |  KIT

Inproceedings3768

Aus Aifbportal

Wechseln zu: Navigation, Suche


Finding Temporal Trends of Scientific Concepts




Published: 2019

Buchtitel: Proceedings of the 8th International Workshop on Bibliometric-enhanced Information Retrieval (BIR’19)
Verlag: CEUR

Referierte Veröffentlichung

BibTeX

Kurzfassung
Science evolves very rapidly, and researchers have studied the evolution of coarse-grained research topics. However, to our knowledge, no analysis of the temporal trends of fine-grained scientific concepts has been performed based on papers' full texts. For this paper, we extract noun phrases as concepts from all computer science papers ofarXiv.org. We then identify positive and negative trends by means of simple linear regression, Mann-Kendall test, and Theil-Sen estimate. In our experiments, we obtain noteworthy findings about trends using the Mann-Kendall test, while the Theil-Sen estimate and simple linear regression lead to many non-scientific concepts. Our findings are potentially relevant for both ordinary researchers and researchers working in bibliometrics and scientometrics.

Download: Media:Science_Trends_BIR2019.pdf
Weitere Informationen unter: Link



Forschungsgruppe

Web Science


Forschungsgebiet

Information Retrieval, Informationsextraktion, Natürliche Sprachverarbeitung