Unsupervised Morphology Induction Using Word Embeddings

Citations Over TimeTop 1% of 2015 papers

Abstract

We present a language agnostic, unsupervised method for inducing morphological transformations between words. The method relies on certain regularities manifest in highdimensional vector spaces. We show that this method is capable of discovering a wide range of morphological rules, which in turn are used to build morphological analyzers. We evaluate this method across six different languages and nine datasets, and show significant improvements across all languages.

Related Papers

• → Template morphology and inflectional morphology(1997)105 cited
• What do overregularizations tell us about morphological knowledge?(2006)
• Surface morphology of crystalline materials(1991)
• → Supplementary material from "Effect of captivity on morphology: negligible changes in external morphology mask significant changes in internal morphology"(2018)
• → Morphology 2(2002)