Context-Aware Prediction of Derivational Word-forms
This addresses the problem of automated derivational morphology prediction for natural language processing applications, but it is incremental as it builds on existing neural methods for morphological tasks.
The paper tackled the task of predicting context-appropriate derivational word-forms from base-form lemmas, using an encoder-decoder neural network that generates derived forms character-by-character. The result showed the model could generate valid context-sensitive derivations from known base forms but was less accurate in a lexicon-agnostic setting.
Derivational morphology is a fundamental and complex characteristic of language. In this paper we propose the new task of predicting the derivational form of a given base-form lemma that is appropriate for a given context. We present an encoder--decoder style neural network to produce a derived form character-by-character, based on its corresponding character-level representation of the base form and the context. We demonstrate that our model is able to generate valid context-sensitive derivations from known base forms, but is less accurate under a lexicon agnostic setting.