Page 163 - JOURNAL OF LIBRARY SCIENCE IN CHINA 2018 Vol. 42

162 Journal of Library Science in China, Vol. 8, 2016

altmetrics indicator etc., which emphasizes the cormon information behavior. The second
perspective is altmetrics source, such as Twitter altmetrics indicator (usually refers to Twitter
mention), Mendeley altmetrics indicator (usually refers to Mendeley readership) and F1000
altmetrics indicator (usually refers to F1000 rating) etc., which emphasizes the characteristics of
different data sources. The research adopted Sina Weibo altmetrics indicator, in order to highlight
characteristics of Sina Weibo, which refers to Sina Weibo mention.

1.2 Data source

Altmetric LLP is worldwide leading provider of altmetrics data and services, providing various
altmetrics services for institutions, libraries, scholars and publishers, as well as supporting
researches of altmetrics topic for the academia. Data sources tracked by the company cover public
policy document, mainstream media, online reference management platforms, post peer review
platforms, Wikipedia, research highlights, blogs, social media, multimedia and other online
platforms. Types of tracked scholarly products include mainly scholarly articles, and increasing
volume of books, dataset and news. Altmetric LLP began tracking in Oct 2011 from publishers
including Nature Publishing Group, Elsevier Publishing Group etc. However, most Chinese journal
articles are not within tracking scope. Among all the data sources, Sina Weibo is the main Chinese
data source. Upon the request of altmetrics research group of UNC Chapel Hill, Altmetric LLP
provided full dataset from Oct 2011 to Nov 2015 consisting of 4,415,638 records, which became
the basis of this research.

1.3 Data processing

Data were stored in JSON file. We used python program to extract all files with Weibo altmetrics
indicator. These files were parsed into CSV documents and then statistically analyzed. The
structure of JSON file is shown in Figure 1 and Figure 2. Weibo is labeled as “weibo” under the
“posts” level.

Figure1. Inner structure Figure 2. Weibo data in the JSON file.
of JSON file.