Feature Selection by Thematic and Temporal Distinction in Research Grant Applications
DOI:
https://doi.org/10.52731/lir.v001.019Keywords:
open data, institutional research, faculty development, text analysisAbstract
We propose an effective method for selecting feature documents from a
research grant database. The goal is to build a useful corpus
for analytical tasks. While grant applications adopted in the past contain abundant
information for institutional research, older applications are not
assigned newer category labels for research areas. It is
often difficult to apply unlabeled data to established techniques for data science
and text analysis. To deal with this issue, our method
automatically categorizes unlabeled grant applications into existing
research categories. Using a document-by-document search technique,
our method selects the best feature documents that are effective for
improving the classification accuracy. To confirm the effectiveness of
our proposed method, we conducted experiments using actual grant
applications. The useful findings obtained in this study are as
follows. (i) Using labeled grant applications, unlabeled grant applications are assigned labels to build a
well-assorted corpus that includes the same number of grant
applications from each research category of each year. (ii) By
selecting a certain number of best feature documents from each
research category of each year, the classification accuracy can be
improved compared to that obtained using the initial dataset
of labeled documents.
References
“KAKEN: Grants-in-Aid for Scientific Research Database (The National Institute of Informatics),” https://kaken.nii.ac.jp/
T. HAYASHI, “University Reform Policy and the New Image of University,” The Journal of Science Policy and Research Management, vol. 36, no. 3, pp. 257–270, 2021.
Y. Kikuchi, “Impact of university reform on research performance aggregated and disaggregated across research fields: a case study of the partial privatization of Japanese national universities,” The Japanese Economic Review, pp. 1–27, 2021.
S. Ito and T. Watanabe, “Multilevel Analysis of Research Management Professionals and External Funding at Universities: Empirical Evidence from Japan,” Science and Public Policy, vol. 47, no. 6, pp. 747–757, 2020.
M. Nishizawa, M. Negishi, M. Shibayama, Y. Sun, H. Nomura, M. Maeda, and Y. Mitsuda, “Evaluation of Japanese universities’ research activity based on the number of awards of Grants-in-Aid for Scientific Research from 1998 to 2002 and in 2003.”
Y. Mizunuma and K. Tsuji, “An Investigation on the Researchers Who Received Japanese Grant-in-Aid for Scientific Research (KAKENHI) with a Focus on Their Roles and Research Achievements,” in 2019 8th International Congress on Advanced Applied Informatics (IIAI-AAI). IEEE, 2019, pp. 103–108.
M. Fujita, T. Okudo, T. Terano, and H. Nagane, “Analyzing Two Approaches in Interdisciplinary Research: Individual and Collaborative,” Journal of Advanced Computational Intelligence and Intelligent Informatics, vol. 25, no. 3, pp. 301–309, 2021.
E. Ono and Y. Ikkatai, “Internet-based Services to Obtain Information on Science and Technology according to the Degree of Interest,” in 2020 9th International Congress on Advanced Applied Informatics (IIAI-AAI). IEEE, 2020, pp. 328–331.
S. Ozeki, T. Hayashi, M. Fukano, S. Yamazaki, A. L. Beach, and M. D. Sorcinelli, “Exploring the Future Trends of Faculty Development in Japanese Higher Education,” in 2021 10th International Congress on Advanced Applied Informatics (IIAI-AAI). IEEE, 2021, pp. 291–294.
C. Yin, Y. Tabata, and S. Hirokawa, “A“ Milky Way Research Trend ”system for Survey of Scientific Literature,” in International Conference on Web-Based Learning. Springer, 2012, pp. 90–99.
M. Igami and A. Saka, “Decreasing diversity in Japanese science, evidence from indepth analyses of science maps,” Scientometrics, vol. 106, no. 1, pp. 383–403, 2016.
K. Kurakawa, Y. Sun, and S. Ando, “Application of a Novel Subject Classification Scheme for a Bibliographic Database Using a Data-Driven Correspondence,” Frontiers in big Data, p. 48, 2020.
Y. Shimada, N. Tsukada, and J. Suzuki, “Promoting diversity in science in Japan through mission-oriented research grants,” Scientometrics, vol. 110, no. 3, pp. 1415 – 1435, 2017.
H. Kawashima and H. Tomizawa, “Accuracy evaluation of Scopus Author ID based on the largest funding database in Japan,” Scientometrics, vol. 103, no. 3, pp. 1061–1071, 2015.
T. Sugihara, S. Aihara, S. Hirokawa, and T. Nara, “An Analysis of Characteristics of Student-Athletes from Questionnaire by SVM,” in 2017 6th IIAI International Congress on Advanced Applied Informatics (IIAI-AAI). IEEE, 2017, pp. 163–166.
K. Ogashiwa, E. Takata, T. Oishi, M. Mori, and S. Hirokawa, “Automatic Estimation and Feature Word Analysis of Universities Using University Medium-term Plans,” in 2019 International Congress on Applied Information Technology (AIT). IEEE, 2019, pp. 1–6.
G. K. Zipf, Human Behavior and the Principle of Least Effort: An Introduction to Human Ecology. Addison-Wesley, 1949.
