@inproceedings{0ebaff7062c14af28e6830917be8e12f,
title = "Prediction of Category of Scientific Article by Graph Convolution",
abstract = "The convolution method that uses the IDs of citing paper and cited paper is known to improve the prediction performance of category of papers. This paper proposes a {"}word convolution{"}method that uses not only the IDs of the cited and citing papers, but also the words that appear in those papers. The proposed method improves the prediction performance (accuracy) 7% for the core dataset and 12% for the citeseer dataset and gives the same performance for the pubmed dataset compared with the state-of-the-art method.",
author = "Sachio Hirokawa and Takahiko Suzuki and Tetsuya Nakatoh",
note = "Funding Information: VI. CONCLUSION AND FURTHER WORK In this paper, we have proposed a vectorization of graph convolution that also uses keywords in citing and cited papers. Three datasets used in previous studies were used to evaluate the proposed method. The improvement was 7% for the cora dataset, 12% for the citeseer dataset, and the same level for the pubmed dataset. The use of graph convolution for literature information is expected to be applied to related research surveys and literature databases [1], [10]. ACKNOWLEDGMENT This work was supported by JSPS KAKENHI Grant Number 18K11990. Publisher Copyright: {\textcopyright} 2020 IEEE.; 9th International Congress on Advanced Applied Informatics, IIAI-AAI 2020 ; Conference date: 01-09-2020 Through 15-09-2020",
year = "2020",
month = sep,
doi = "10.1109/IIAI-AAI50415.2020.00023",
language = "English",
series = "Proceedings - 2020 9th International Congress on Advanced Applied Informatics, IIAI-AAI 2020",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "63--67",
editor = "Tokuro Matsuo and Kunihiko Takamatsu and Yuichi Ono and Sachio Hirokawa",
booktitle = "Proceedings - 2020 9th International Congress on Advanced Applied Informatics, IIAI-AAI 2020",
address = "United States",
}