Page 130 - Journal of Library Science in China, Vol.45, 2019

ZHANG Chengzhi, LI Zhuo, ZHAO Mengyuan, LIU Jiahao & ZHOU Qingqing / Citing behavior of Chinese books based on citation content 129

method (Abu-Jbara et al., 2013). Firstly, we counted neutral citations and subjective citations, and
the results are shown in Table 4. Secondly, we subdivided the subjective citations into positive and
negative citations, and the statistical results are shown in Table 5.

Table 4. Proportion of books’ citation sentiments in different disciplines (%)
Discipline Computer science Law Literature Medicine Sport science

Neutral ratio 84.2 83.6 84.9 85.1 93.4

Subjective ratio 15.8 16.4 15.1 14.9 6.6

Table 5. Proportion of positive and negative citations in different disciplines (%)
Discipline Computer science Law Literature Medicine Sport science

Positive ratio 91.1 76.7 91.4 87.4 94.1

Negative ratio 8.9 23.3 8.6 12.6 5.9

The proportion of positive citations in sport science is relatively high, reaching 93.4%. The
sentiment citations in other disciplines only account for about 15%, which is close to the statistical
results of citation content of 310 articles in natural language processing by Athar (accounting for
14%) (Athar, 2011). From Table 5, it can be seen that in computer science, literature, medicine
and sport science, the proportions of positive citations are about 90%, while the negative citations
in law are 23.3%. In general, authors tend to express positive sentiments when citing Chinese
books.

4 Discussion

Books are important communication resources in academic research, and analyzing citing
behaviors over Chinese books has important theoretical and practical significance. Research on
the citing behaviors of books can not only be used to evaluate the academic impacts of books but
also to explore the discipline differences, thus providing a reference for book evaluation analysis
in different disciplines. Currently, relevant researches on citation content in academic articles have
aroused wide attention in academia (Y. Ding et al., 2014; Hu, 2014; S.B. Liu & K. Ding, 2013;
W. Lu et al., 2014). However, few of the existing academic research on citation content analysis
distinguished the types of references corresponding to citation content (e.g. journal articles,
conference papers, books, etc.). We refer to these references as non-specific types of references.
This paper analyzed the differences of feature distribution between books and non-specific types of
references.