Generating Combination of Biblical Baby Names using Recurrent Neural Network (RNN) and Optimization Comparison

Sanjaya, Samuel Ady (2023) Generating Combination of Biblical Baby Names using Recurrent Neural Network (RNN) and Optimization Comparison. 4th International Conference on Big Data Analytics and Practices (IBDAP).

Text Generating Combination of Biblical Baby Names using Recurrent Neural Network (RNN) and Optimization Comparison.pdf Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (10MB)

Abstract

Giving names to newborns in every country has its own uniqueness based on ethnicity, religion, race, culture, and names from past history. All parents in the world should give their baby a name with a good meaning because there is a saying that a name is a prayer for the child. However, these names have been used repeatedly based on trends every year, so it is very possible that humans on Earth have the same name. In this research, an approach will be made to generate names based on names from the Bible to make combinations of characters into new names. This Bible-based name is used because it is the most common source of names for Christians around the world. Names will be generated using a Recurrent Neural Network (RNN) with a comparison of several Optimizers such as Adadelta, Adagard, SGD, Adam, Adamax, and RMSProp. This comparison found that the 3 optimizers with the best loss function were Adam at 1,755, RMSProp at 1,814, and Adamax at 2,064. The optimizer with the best loss function can produce data with a good combination of consonants and vowels, but the optimizer with a bad loss function will produce a large number of combinations of consonants. This research resulted in a new approach for generating names from the RNN model generation with the best optimizer that can be used as inspiration for baby names.

Item Type:	Article
Keywords:	Biblical names, loss function, RNN, optimizer algorithm
Subjects:	000 Computer Science, Information and General Works > 000 Computer Science, Knowledge and Systems > 000 Computer Science, Information and General Works
Divisions:	Faculty of Engineering & Informatics > Information System
Depositing User:	Administrator UMN Library
Date Deposited:	05 Aug 2025 09:35
Last Modified:	05 Aug 2025 09:35
URI:	https://kc.umn.ac.id/id/eprint/39848

Actions (login required)

View Item