Page 165 - JOURNAL OF LIBRARY SCIENCE IN CHINA 2018 Vol. 43

165
Extended English abstracts of articles published in the Chinese edition of Journal of Library Science in China 2017 Vol.43 165

Evaluation of academic impact of open source software based on Altmetrics

ZHAO Rongying, WEI Mingkun〇 & WANG Shaozhen
〇a*
The development of Internet and social media has an increasingly stronger influence on academic
communication with the advent of Web 2.0 technology. Nowadays academic achievements
include not only papers but also non-paper forms of research results, such as blogs, videos, data
sets, software code etc. Academic influence is the impact on scientific research achievements
and academic status in academia. These non-paper forms of academic research should be also
included in the evaluation system of scholars or academic achievements. Open source software,
as a non-paper form of academic achievements, plays an important role in scientific research. The
development of Altmetrics not only enriches the development of metrology but also provides a
new perspective and method for the evaluation of academic influence. At present, there are lots
of studies on non-traditional evaluation indicators, but the researches on the evaluation object are
relatively few, especially those on open source software, which plays an important role in academic
research. The number of downloads, citations, and reuse between applications for academic
research are concrete reflections of its academic influence. At the same time, the open source
community and the development of computer technology make the research tools of software more
and more abundant.
Firstly, the index database of the website Depsy is obtained in the paper, and the open source
community of Python is researched and analyzed through the data provided by ImpactStory.
Then the third-party library Python pandas is used for extraction of the required attributes, i.e.
software library downloads, the cited academic literature and the relationship between the reuse of
software for further study of the academic influence of open source software, and finally the three
indicator data were analyzed. From the results of the software community analysis in the Python
community, three indicators-downloads, citation counts, and software reuse-successfully screened
out popular software in their communities. From the perspective of users, software reflects the
ordinary users to use more academic papers, the authors quoted for many times, and software
developers to reuse more times. There is a certain correlation between the three indicators, such
as software’s dependence that will lead to the increasing on the number of software downloads.
The increasing number of citations in academic papers has led to an increase in the use of the
software in the academic field. The correlation between download and reuse is relatively high,
which indicates that the dependency between software has a direct impact on the download;
the correlation between citation and reuse is low, which is attributed to the difference between
scholars and developers, and the citation index of the software in the academic literature reflects
the academic influence.
Finally, we can reach the following conclusions: 1)the evaluation of academic influence

* Correspondence should be addressed to WEI Mingkun, Email: weimingkun24@163.com, ORCID: 0000-0001-7689-6294