Division of Linguistics and Multilingual Studies
Nanyang Technological University
14 Nanyang Dr, Singapore 637332
theoretical and computational pragmatics (especially conversation analysis), constructions and frames, machine learning
I am a PhD candidate in linguistics with a background in data science concerned with developing new tools for spoken dialog systems. My research focuses on describing and modeling pragmatic and phonological aspects of constructions (cxn) in human speech and their application in the field of spoken language understanding. More specifically, my PhD thesis is concerned with investigating and modelling the use of ‘causal’ discourse markers (janwai, yinwei, because) in a corpus of naturally-occurring everyday speech that I built. Drawing on insights from interactional and cognitive linguistics, the project aims to provide both (i) theoretical insights into how these discourse marker constructions are used in the real world and (ii) practical insights useful for designing open-domain spoken dialog systems and discourse parsers.
Part of this project is the compilation of a video corpus of Cantonese naturally-occurring speech that will be made available soon (see mycancor).
I mainly work with Chinese languages (Mandarin and Cantonese).
Tsundoku master, cyclist and hunter.
#CAS #CxG #ANT #NLP #ECG #EMCA #FCG #NLU #TCG #MDS #SIGSEM
Recent publications (refereed)
Liesenfeld, Andreas (2018). The use of janwai (yinwei) in environments of disagreement: Evidence from MYCanCor. 19th Chinese Lexical Semantics Workshop (CLSW 2018), National Chung Cheng University, Chiayi County, Taiwan.
Liesenfeld, Andreas (2018) MYCanCor: A Video Corpus of Malaysian Cantonese. 11th edition of the Language Resources and Evaluation Conference (LREC), Miyazaki, Japan
Liesenfeld, Andreas. (2017) Project notes on building a conversation parser on top of a text parser: Towards a causal language tagger for spoken Chinese. ISA-13: Thirteenth Joint ACL – ISO Workshop on Interoperable Semantic Annotation at 12th International Conference on Computational Semantics (IWCS), Laboratoire d’Informatique, de Robotique et de Microélectronique de Montpellier (LIRMM), 19-22 September 2017. Proceedings
Liesenfeld, Andreas. (2017) Extracting construction networks of Causality from Cantonese speech corpora using word embeddings. Corpus Linguistics Conference, University of Birmingham, 25-28 July 2017. Poster Abstract