Transformer based Model for Coherence Evaluation of Scientific Abstracts: Second Fine-tuned BERT

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 aforeme...

Descripción completa

Detalles Bibliográficos
Autores: Gutiérrez Choque, Anyelo Carlos, Medina Mamani, Vivian, Castro Gutiérrez, Eveling, Núñez Pacheco, Rosa, Aguaded, José Ignacio
Tipo de recurso: artículo
Fecha de publicación:2022
País:España
Institución:Universidad de Huelva (UHU)
Repositorio:Arias Montano. Repositorio Institucional de la Universidad de Huelva
Idioma:inglés
OAI Identifier:oai:ariasmontano.uhu.es:10272/22012
Acceso en línea:https://hdl.handle.net/10272/22012
Access Level:acceso abierto
Palabra clave:Coherence evaluation
Inconsistent sentences detection
BERT
Second fine-tuned
5701 Lingüística Aplicada
Descripción
Sumario: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 abstracts and mixing them when performing Second Fine-Tuned does not provide better results than using a single technique.