Project 52: Spoken Language Understanding with Graph Neural Network
Contact Information
Prof. Kai Yu
Email: kai.yu@sjtu.edu.cn
Project Description and Objectives
In the task-based human-machine dialogue system, it is important to convert the sentences of user input or speech recognition automatically into the semantic understanding of structured information. This topic aims at the structural and hierarchical definition of semantic objects (domain knowledge). It aims to use graph neural network to combine relevant domain knowledge with a general semantic understanding model to construct a robust and extensible semantic understanding system.
Eligibility Requirements
Interested students should have basic knowledge of machine learning.
Main Tasks
Development of a paradigm to employ knowledge graph in spoken language understanding.
Completion of a research report.
Website