RT Journal Article
T1 Transformer based Model for Coherence Evaluation of Scientific Abstracts: Second Fine-tuned BERT
A1 Gutiérrez Choque, Anyelo Carlos
A1 Medina Mamani, Vivian
A1 Castro Gutiérrez, Eveling
A1 Núñez Pacheco, Rosa
A1 Aguaded, José Ignacio
AB Coherence evaluation is a problem related to the area of natural language processing whose complexity lies mainly in the analysis of the semantics and context of the words in the text. Fortunately, the Bidirectional Encoder Representation from Transformers (BERT) architecture can capture the aforementioned variables and represent them as embeddings to perform Fine-tunings. The present study proposes a Second Fine-Tuned model based on BERT to detect inconsistent sentences (coherence evaluation) in scientific abstracts written in English/Spanish. For this purpose, 2 formal methods for the generation of inconsistent abstracts have been proposed: Random Manipulation (RM) and K-means Random Manipulation (KRM). Six experiments were performed; showing that performing Second Fine-Tuned improves the detection of inconsistent sentences with an accuracy of 71%. This happens even if the new retraining data are of different language or different domain. It was also shown that using several methods for generating inconsistent abstractsand mixing them when performing Second Fine-Tuned does not provide better results than using a single technique.
PB The Science and Information Organization
SN 2158-107X
SN 2156-5570 (electrónico)
YR 2022
FD 2022
LK https://hdl.handle.net/10272/22012
UL https://hdl.handle.net/10272/22012
LA eng
NO Gutiérrez-Choque, A.-C., Medina-Mamani, V., Castro-Gutiérrez, E., Nuñez-Pacheco, R., & Aguaded, I. (2022). Transformer based Model for Coherence Evaluation of Scientific Abstracts: Second Fine-tuned BERT. In International Journal of Advanced Computer Science and Applications (Vol. 13, Issue 5). The Science and Information Organization. https://doi.org/10.14569/ijacsa.2022.01305105
NO To the Universidad Nacional de San Agust´ın de Arequipa for the funding granted to the project ”Transmedia, Gamificationand Video games to promote scientific writing in Engineering students”, under Contract No. IBA-IB-38-2020-UNSA. We would like to thank to the ”Research Center, Transfer of Technologies and Software Development R+D+i” -CiTeSoft-EC-0003-2017-UNSA, for their collaboration in the use of their equipment and facilities, for the development of this research work.
DS Repositorio Institucional de la Universidad de Huelva
RD 13 jun 2026