Journal of Applied Sciences and Applications in Engineering
Research Article Volume: 2 & Issue: 2
Journal Menu
- Home
- Aims and Scope
- Author Guidelines
- Editorial Board
- Reviewer Board
- Current Issue
- Archive
- Journal Contact
Indexed & Abstracted In
Abstract
This study investigates automatic identification of the English ditransitive construction by integrating LoRA fi ne-tuning of a large language model with a Retrieval-Augmented Generation (RAG) framework. A binary classification task was conducted on annotated data from the British National Corpus. Results show that a LoRA-fi ne-tuned Qwen3-8B model significantly outperformed both a native Qwen3-MAX model and a theory-only RAG system, achieving an accuracy of 0.936 and an F1 score of 0.874. Error analysis reveals that fi ne-tuning shifts the model’s judgment from surface-form matching towards semantically grounded understanding. The RAG system’s poor performance underscores the necessity of combining theoretical knowledge with instance-based learning. This work provides an effective, low-cost computational pathway for implementing constructionist theories and offers a practical tool for linguistic research and second language pedagogy.
Keywords: Construction Grammar; Ditransitive Construction; Large Language Models; LoRA Fine-tuning; Retrieval-Augmented Generation.
AUTHOR TESTIMONIALS
Trusted by Researchers
Experiences from authors who have published with SCIVOLVE.
I had a very positive experience working with the Journal of Case Reports and Reviews in Medicine. The publication process was quick and well organized, every step was clearly explained, and the editorial team was always responsive and helpful.
I really enjoyed communicating with the editorial staff. The editorial team provided clear and efficient communication throughout the process. The manuscript submission process was straightforward, and the editorial team made it easy to address the necessary corrections. I appreciated the professional, patient, and supportive approach of the editorial staff, which has encouraged me to consider publishing my future work.
I had an excellent experience publishing my manuscript with the journal. The communication with the editorial team was clear, professional, and efficient throughout the entire publication process. I truly appreciated their support and responsiveness.
As I pursue a Doctor of Juridical Science degree, I am pleased to have worked with Scivolve Publishers on publishing my article on allergies. In my work to advance women’s human rights and help humanity, it is vital to have publishing opportunities that respect research integrity. From the first contact with the Editorial Manager to the final stage, the process was precise, transparent, and exemplary. Nothing was overpromised, and the handling of sources and references was also excellent. I warmly recommend Scivolve Publishers for their integrity, professionalism, and the respect with which they conduct their work.
From submission to final publication, the process was organized and easy to follow. The editorial communication made the overall experience straightforward and professional.