A Theoretic System for Unified Linguistics: An Introduction

A major challenge for machine translation and cross-lingual applications is the lack of corresponding relationships among characters, words, and grammars of different languages, unless they are homological. This article reviews a theory based on semantic units (SUs) and their representation to address this challenge. SUs share the same meaning across languages, so the unified linguistics theory described here has the potential to greatly simplify natural language processing with rich semantics.