Recruit Co., Ltd. April 2023 - Present
Machine Learning Engineer & Data Scientist
Ayaka Ueyama
Machine Learning Engineer / Data Scientist at Recruit Co., Ltd.
Master of Informatics
Work Experiences
LINE Corporation November 2021 - March 2022
AI Company, Machine Learning & Server-side Engineer ( Part-time job )
Recruit Co., Ltd. October 2021 - November 2021
Machine Learning Engineer & Data Scientist ( Part-time job )
LINE Corporation August 2021 - September 2021
AI Company, Machine Learning & Server-side Engineer ( Internship )
[blog]
Akatsuki Inc. July 2021
Server-side Engineer ( Internship )
Publications
Journal
Ayaka Ueyama, Yoshinobu Kano.
DCZAR: Dialogue Response Generation Using Completion of Omitted Predicate Arguments Based on Zero
Anaphora Resolution.
Transactions of the Japanese Society for Artificial Intelligence, Vol. 39, No. 3, 2024.
[paper]
Masaki Fujita, Takaaki Onaga, Ayaka Ueyama, Yoshinobu Kano.
Legal Textual Entailment using Ensemble of Rule-based and BERT-based method with Data Augmentation
by Related Article Generation.
Lecture Notes in Artificial Intelligence, Cham: Springer Nature Switzerland, 2023.
[paper]
Ayaka Ueyama, Yoshinobu Kano.
Diverse Dialogue Generation by Deep Learning Methods Using Loss Function Incorporating Word
Statistics.
Transactions of the Japanese Society for Artificial Intelligence, Vol. 37, No. 2, 2022.
[paper]
International Conferences (referred)
Ayaka Ueyama, Yoshinobu Kano.
Dialogue Response Generation Using Completion of Omitted Predicate Arguments Based on Zero Anaphora
Resolution.
In Proceedings of the 24th Meeting of the Special Interest Group on Discourse and Dialogue (SIGDIAL2023,
Main),
Prague Czechia, 2023.
[paper]
Masaki Fujita, Takaaki Onaga, Ayaka Ueyama, Yoshinobu Kano.
Legal Textual Entailment using Ensemble of Rule-based and BERT-based method with Data Augmentations
including Generation without Excess or Deficiency.
In Proceedings of the 16th International Workshop on Juris-informatics (JURISIN2022), Kyoto Japan, 2022.
Ayaka Ueyama, Yoshinobu Kano.
Diverse Dialogue Generation with Context Dependent Dynamic Loss Function.
In Proceedings of the 28th International Conference on Computational Linguistics (COLING2020, Main),
pp.4123–4127, Barcelona Spain (Online), 2020. (Acceptance rate:26.2%)
[paper]
Domestic Conference / Others
Awards