Context-sensitive Classification for Scientific Keywords in Grant Reports
DOI:
https://doi.org/10.52731/lir.v004.308Abstract
In the task of institutional research (IR), it is important for each university to identify the latest trends in cutting-edge scientific research and to understand its own strengths. The Grantin-Aid for Scientific Research (KAKENHI), the largest research grant in Japan, makes publicly available the research outline, progress, and keywords of adopted research projects. These open data can be used to analyze research information in IR tasks. Our study in this paper focuses specifically on keyword analysis in research grant reports. Technical terms that describe scientific projects are important clues in analyzing research information. However, state-of-the-art terminology is not easy to process on computers because word occurrences and usages are often polysemous and unpredictable. To deal with this issue, we propose a method for disambiguating keywords by attaching a prefix to each keyword that takes into account the context in which the keyword appears. Such contextual prefixes are expected to enable useful searches for relevant keywords and automatic classification of keywords. Evaluation experiments on real data confirmed the effectiveness of our proposed method.
References
The National Institute of Informatics, “KAKEN: Grants-in-Aid for Scientific Research Database,” https://kaken.nii.ac.jp/ .
Japan Society for the Promotion of Science, “The Review Section Table for Basic Sections (lists),” https://www-kaken.jsps.go.jp/kaken1/shoukubunListEn.do .
M. Yasukawa and K. Yamazaki, “Entity Linking among Categorized Knowledge Resources for Computer Science Curricula,” IIAI Letters on Institutional Research, vol. 3, no. LIR152, pp. 1–14, 2023.
T. Tsumagari, N. Nakazato, and T. Tsumagari, “Student’ Interests and Career Understanding: A Topic Analysis of First-year Career Courses,” IIAI Letters on Institutional Research, vol. 1, no. LIR013, pp. 1–8, 2022.
A. Itoh, H. Ito, S. Matsumoto, I. Noda, K. Bannaka, K. Nishiyama, T. Kirimura, T. Kunisaki, K. Mitsunari, K. Murakami, R. Kozaki, A. Kishida, M. Kondo, S. Imai, M. Mori, Y. Nakata, M. Omori, and K. Takamatsu, “A Two-Step Approach for Syllabus Development and Evaluation using Machine Learning such as Doc2Vec based on Eduinformatics,” IIAI Letters on Institutional Research, vol. 3, no. LIR128, pp. 1–11, 2023.
M. Yasukawa and K. Yamazaki, “Detecting Transition of Research Themes using Time-oriented Attributes in Governmental Funding,” International Journal of Institutional Research and Management, vol. 7, no. 1, pp. 1–17, 2023.
F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V. Dubourg, J. Vanderplas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and E. Duchesnay, “Scikit-learn: Machine learning in Python,” Journal of Machine Learning Research, vol. 12, pp. 2825–2830, 2011.
K. Robershaw and B. Wolf, “Research analytics: A systematic literature review,” Social Science Research Network (SSRN), 2023. [Online]. Available: http://dx.doi.org/10.2139/ssrn.4363262
M. Abeysiriwardana and D. Sumanathilaka, “A survey on lexical ambiguity detection and word sense disambiguation,” IEEE International Colloquium on Signal Processing and its Applications (CSPA 2024), 2024. [Online]. Available https://arxiv.org/abs/2403.16129
M. Bevilacqua, T. Pasini, A. Raganato, and R. Navigli, “Recent trends in word sense disambiguation: A survey,” in International Joint Conference on Artificial Intelligence (IJCAI-21), 2021, pp. 4330–4338. [Online]. Available: https://www.ijcai.org/proceedings/2021/0593.pdf
The PostgreSQL Global Development Group, “Chapter 9. Functions and Operators,” https://www.postgresql.org/docs/current/functions.html .
