Abstract
Existing researches have proved that both academic reviews and library holdings can be alternative sources to assess impacts of academic books. This paper endeavors to identify correlations between the two sources, which may be useful for book ordering of libraries. Specifically, 69,263 academic reviews in Choice {Choice: Current Reviews for Academic Libraries) are collected with four metrics: recommendation levels, readership levels, numbers of interdisciplinary subjects and review contents. Then, a topic model is used to extract topics from review contents. Meanwhile, library holdings of each book are identified, including the total number of library holdings, holding regions and holding distributions based on an entropy method. Finally, correlation analysis between Choice reviews and library holdings are conducted. Experimental results reveal that books with higher recommendation levels or extensive readerships tend to be ordered more by academic libraries. Meanwhile, books with extensive readerships will be collected more uniformly by academic libraries. In conclusion, metrics derived from Choice academic book reviews can be used as indicators to recommend book ordering of academic libraries.
| Original language | English |
|---|---|
| Pages | 31-43 |
| Number of pages | 13 |
| Publication status | Published - 2017 |
| Externally published | Yes |
| Event | 16th International Conference on Scientometrics and Informetrics, ISSI 2017 - Wuhan, China Duration: 16 Oct 2017 → 20 Oct 2017 |
Conference
| Conference | 16th International Conference on Scientometrics and Informetrics, ISSI 2017 |
|---|---|
| Country/Territory | China |
| City | Wuhan |
| Period | 16/10/17 → 20/10/17 |
Keywords
- Altmetric
- Book impact assessment
- Library holdings
- Review mining
- Topic extraction