Project notes on building a conversation parser on top of a text parser: Towards a causal language tagger for spoken Chinese

Abstract

This ongoing doctoral study examines cause and effect relationships in Chinese spoken language corpora and aims to build a tagger (Cause-Chi) that automatically annotates linguistic patterns used to express these relationships. Drawing on insights from Construction Grammar (CxG), Cause-Chi is a tool to detect explicit causal language and automatically parse constructions of causation and their slot-fillers for Chinese conversational corpus data. Built on top of an existing tagger for text corpora, Cause-Chi is designed to not only detect lexical constructions but also conversation-specific causal language such as multi-segment causal expressions and the usage of temporal constructions to express causal relation. Cause-Chi is currently under development and will be released in 2018 together with MYCanCor, a small corpus of spoken Chinese, and a mini-constructicon of causal constructions based on the corpus.

Publication
In ISA-13 Thirteenth Joint ACL - ISO Workshop on Interoperable Semantic Annotation at 12th International Conference on Computational Semantics (IWCS)
Andreas Liesenfeld
Andreas Liesenfeld
Postdoc in language technology

I am a social scientist working in Conversational AI. My work focuses on understanding how humans interact with voice technologies in the real world. I aspire to produce useful insights for anyone who designs, builds, uses or regulates talking machines. Late naturalist under pressure. “You must collect things for reasons you don’t yet understand.” — Daniel J. Boorstin