Determination of language families using deep learning
This work addresses the challenge of linguistic classification for researchers in computational linguistics and historical linguistics, though it appears incremental as it adapts existing deep learning methods to a new application.
The authors tackled the problem of determining language families by applying a convolutional generative adversarial network (c-GAN) to transliterated text fragments from various languages, including a non-deciphered one (Cypro-Minoan), and established linguistic affinities without relying on translation or deciphering.
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.