Semantic Role Labeling: Case studies and What’s Next?

PhD Qualifying Examination


Title: "Semantic Role Labeling: Case studies and What’s Next?"

by

Mr. Serkan KUMYOL


Abstract:

In contrast to sparse bag-of-words and syntactic features, semantic roles 
provide meaning bearing features for the NLP tasks such that synonymous 
sentences are labelled the same, even if they differ syntactically. Therefore, 
its natural to say that semantic role labeling (SRL) is a linguistic 
information that reduces the domain dependency and sparsity in downstream NLP 
tasks. In this paper, we will first establish the foundations of the SRL such 
as meaning representation and role of syntax in meaning, then conduct a 
comparative analysis between different computational models. Finally, address 
even more abstract problem of computational metaphor analysis.


Date:			Tuesday, 24 August 2021

Time:                  	2:00pm - 4:00pm

Zoom meeting:
https://hkust.zoom.us/j/91048958612?pwd=Mnl0VFNDSG1rMjNTVTNFOS9qVXo0UT09

Committee Members:	Prof. Dekai Wu (Supervisor)
 			Prof. Nevin Zhang (Chairperson)
 			Prof. Pedro Sander
 			Dr. Yangqiu Song


**** ALL are Welcome ****