Morphological Inflection Generation Using Character Sequence to Sequence Learning

TL;DR

This paper introduces a neural encoder-decoder model for morphological inflection generation, capable of handling multiple languages and trained in supervised or semi-supervised settings, achieving competitive results.

Contribution

It presents a language-independent character sequence to sequence neural model for inflection generation, improving or matching state-of-the-art performance.

Findings

Achieved superior or comparable results on seven language datasets.

Model is effective in both supervised and semi-supervised training.

Demonstrates versatility across morphologically rich languages.

Abstract

Morphological inflection generation is the task of generating the inflected form of a given lemma corresponding to a particular linguistic transformation. We model the problem of inflection generation as a character sequence to sequence learning problem and present a variant of the neural encoder-decoder model for solving it. Our model is language independent and can be trained in both supervised and semi-supervised settings. We evaluate our system on seven datasets of morphologically rich languages and achieve either better or comparable results to existing state-of-the-art models of inflection generation.