Abstract
The definition of a complex question is a long sentence with many focal points that may have many answers, involve many entities, or use non-standard linguistic constructs. Complex questions can be obtained from voice commands. It will help some people with blindness or visual impairments. Voice commands have become a mandatory feature of applications, including the basic one to access databases. In this work, the Generate Structured Query Language Configuration (GSC) approach was developed to answer complex questions in the Indonesian language. The non-learning approach was chosen as the learning process needs costly computation and is inappropriate for low-resource languages. GSC can be plugged into any database system without any learning process and is implemented on Android devices. GSC succeeded in generating about 93.33% of all the questions. An accuracy of 80.01% was achieved based on experiments with 196 questions submitted to the Mondial database. The accuracy of the truth in the 114-question experiment using the IMDB database was 78.07%. The fundamental framework of GSC does not have a very high score but is still promising. Indeed, this work must be elaborated on shortly by adding a few methods to handle some limitations.
Keywords
Assistive technology, Complex query, Database, SQL, Voice question answering
Subject Area
Computer Science
Article Type
Article
First Page
2352
Last Page
2363
Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.
How to Cite this Article
Prabaswara, P. and Wardani, D.
(2026)
"Complex Question Analysis in Voice Commands Using Generate SQL for Database Searching,"
Baghdad Science Journal: Vol. 23:
Iss.
6, Article 29.
DOI: https://doi.org/10.21123/2411-7986.5343
