Determination of language families using deep learning

TL;DR

This paper employs a c-GAN neural network to analyze transliterated texts in order to identify linguistic relationships among various languages, including undeciphered scripts, without relying on translation or decipherment.

Contribution

It introduces a novel application of c-GANs for linguistic affinity detection across multiple languages and scripts, including undeciphered ones, without requiring translation.

Findings

Successful analysis of transliterated texts to establish language affinities

Potential for aiding decipherment of undeciphered scripts

Demonstrates neural networks' utility in comparative linguistics

Abstract

We use a c-GAN (convolutional generative adversarial) neural network to analyze transliterated text fragments of extant, dead comprehensible, and one dead non-deciphered (Cypro-Minoan) language to establish linguistic affinities. The paper is agnostic with respect to translation and/or deciphering. However, there is hope that the proposed approach can be useful for decipherment with more sophisticated neural network techniques.