TY - GEN
T1 - Author-Topic Modelling for Reviewer Assignment of Scientific Papers in Bahasa Indonesia
AU - Kusumawardani, Renny Pradina
AU - Khairunnisa, Siti Oryza
N1 - Publisher Copyright:
© 2018 IEEE.
PY - 2018/7/2
Y1 - 2018/7/2
N2 - Assigning reviewers to paper submissions is a knowledge-intensive and time-consuming work, especially since committee members have to sufficiently understand the paper and have a broad knowledge of reviewers' expertise to be able to match a submission to multiple reviewers. In this paper, we apply the author-topic modelling to a corpus of scientific papers in the Indonesian language, and reviewers as represented by the papers they author. We explore the use of stemming and POS-tagging to address some of the issues arising from the morphologically agglutinative nature of Bahasa Indonesia. We also use bigrams to capture multiword terms, as these are often found in Bahasa Indonesia as capturing semantically atomic concepts. We found that stemming does improve the performance of the author-topic model in the reviewer assignment task, while POS-tagging might not. Our results show that upon inspecting papers and the reviewers suggested by the model, there indeed exists a sound and reasonable relationship between these, revealing prospective reviewers which might not previously been suspected of being a good match.
AB - Assigning reviewers to paper submissions is a knowledge-intensive and time-consuming work, especially since committee members have to sufficiently understand the paper and have a broad knowledge of reviewers' expertise to be able to match a submission to multiple reviewers. In this paper, we apply the author-topic modelling to a corpus of scientific papers in the Indonesian language, and reviewers as represented by the papers they author. We explore the use of stemming and POS-tagging to address some of the issues arising from the morphologically agglutinative nature of Bahasa Indonesia. We also use bigrams to capture multiword terms, as these are often found in Bahasa Indonesia as capturing semantically atomic concepts. We found that stemming does improve the performance of the author-topic model in the reviewer assignment task, while POS-tagging might not. Our results show that upon inspecting papers and the reviewers suggested by the model, there indeed exists a sound and reasonable relationship between these, revealing prospective reviewers which might not previously been suspected of being a good match.
KW - author-topic model
KW - probabilistic vector similarity
KW - reviewer assignment
KW - topic modeling
UR - http://www.scopus.com/inward/record.url?scp=85062784552&partnerID=8YFLogxK
U2 - 10.1109/IALP.2018.8629124
DO - 10.1109/IALP.2018.8629124
M3 - Conference contribution
AN - SCOPUS:85062784552
T3 - Proceedings of the 2018 International Conference on Asian Language Processing, IALP 2018
SP - 351
EP - 356
BT - Proceedings of the 2018 International Conference on Asian Language Processing, IALP 2018
A2 - Dong, Minghui
A2 - Bijaksana, Moch.
A2 - Sujaini, Herry
A2 - Negara, Arif Bijaksana Putra
A2 - Romadhony, Ade
A2 - Ruskanda, Fariska Z.
A2 - Nurfadhilah, Elvira
A2 - Aini, Lyla Ruslana
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 22nd International Conference on Asian Language Processing, IALP 2018
Y2 - 15 November 2018 through 17 November 2018
ER -