%0 Journal Article %1 cavacini2015database %A Cavacini, Antonio %D 2015 %I Springer %J Scientometrics %K dblp google-scholar inspec real scopus web-of-science %N 3 %P 2059-2071 %R 10.1007/s11192-014-1506-1 %T What is the best database for computer science journal articles? %U https://link.springer.com/article/10.1007/s11192-014-1506-1 %V 102 %X We compared general and specialized databases, by searching bibliographic information regarding journal articles in the computer science field, and by evaluating their bibliographic coverage and the quality of the bibliographic records retrieved. We selected a sample of computer science articles from an Italian university repository (AIR) to carry out our comparison. The databases selected were INSPEC, Scopus, Web of Science (WoS), and DBLP. We found that DBLP and Scopus indexed the highest number of unique articles (4.14 and 4.05 \% respectively), that each of the four databases indexed a set of unique articles, that 12.95 \% of the articles sampled were not indexed in any of the databases selected, that Scopus was better than WoS for identifying computer science publications, and that DBLP had a greater number of unique articles indexed (19.03 \%), when compared to INSPEC (11.28 \%). We also measured the quality of a set of bibliographic records, by comparing five databases: Scopus, WoS, INSPEC, DBLP and Google Scholar (GS). We found that WoS, INSPEC and Scopus provided better quality indexing and better bibliographic records in terms of accuracy, control and granularity of information, when compared to GS and DBLP. WoS and Scopus also provided more sophisticated tools for measuring trends of scholarly publications.

@article{cavacini2015database, abstract = {We compared general and specialized databases, by searching bibliographic information regarding journal articles in the computer science field, and by evaluating their bibliographic coverage and the quality of the bibliographic records retrieved. We selected a sample of computer science articles from an Italian university repository (AIR) to carry out our comparison. The databases selected were INSPEC, Scopus, Web of Science (WoS), and DBLP. We found that DBLP and Scopus indexed the highest number of unique articles (4.14 and 4.05 {\%} respectively), that each of the four databases indexed a set of unique articles, that 12.95 {\%} of the articles sampled were not indexed in any of the databases selected, that Scopus was better than WoS for identifying computer science publications, and that DBLP had a greater number of unique articles indexed (19.03 {\%}), when compared to INSPEC (11.28 {\%}). We also measured the quality of a set of bibliographic records, by comparing five databases: Scopus, WoS, INSPEC, DBLP and Google Scholar (GS). We found that WoS, INSPEC and Scopus provided better quality indexing and better bibliographic records in terms of accuracy, control and granularity of information, when compared to GS and DBLP. WoS and Scopus also provided more sophisticated tools for measuring trends of scholarly publications.}, added-at = {2019-11-16T00:26:11.000+0100}, author = {Cavacini, Antonio}, biburl = {https://www.bibsonomy.org/bibtex/2e19da0d89ea59df77dfed238d5434d25/jpmor}, day = 01, description = {What is the best database for computer science journal articles? | SpringerLink}, doi = {10.1007/s11192-014-1506-1}, interhash = {d0c5894af16c440fb6d714db3b57c8a3}, intrahash = {e19da0d89ea59df77dfed238d5434d25}, issn = {1588-2861}, journal = {Scientometrics}, keywords = {dblp google-scholar inspec real scopus web-of-science}, language = {English}, month = mar, number = 3, pages = {2059-2071}, publisher = {Springer}, school = {University of Milan (UniMi)}, timestamp = {2020-10-07T13:36:50.000+0200}, title = {What is the best database for computer science journal articles?}, url = {https://link.springer.com/article/10.1007/s11192-014-1506-1}, volume = 102, year = 2015 }