Skip to Main Content

CS425

Download as PDF

Generative AI for Natural Language Communication

SCIS Sch of Computing & Info Sys

Course (UG/PG)

Undergraduate

Offering Unit/Department

Course Description

Natural Language Communication (NLC) is the convergence of a diverse set of human language technologies that enable computer systems to interact reasonably with people in a natural and human-like way. NLC requires considering human language as the central part of communicative channel, where the computer should be able to perform a series of language processing actions:
- It should correctly process our written or spoken utterances as input in order to respond accordingly;
- It should allow technology to understand complex sentences, which may contain multiple pieces of information and many turns of requests;
- It can then react by reasoning and/or interrogating and synthesising various data from third-party systems or external knowledge, and use that information in generating sensible responses.

In this course, we will cover diverse fundamental methods and techniques across the themes of natural language processing, understanding and generation that are indispensable for constructing modern NLC systems. We will be focused on introducing and discussing the underlying computational models, data resources, toolkits, and practising them in developing interactive information seeking, dialogue, and cross-language communication systems. This includes but is not limited to exploration of a few conversational AI applications such as question answering, chatbots, virtual personal assistants, and dialogue management.

Course Learning Outcomes

  • Design and apply appropriate techniques, models, algorithms, toolkits and datasets built-in-class or from off-the-shelf resources for building up problem- solving solutions which are deliverable in due course.
  • Independently achieve the learning goals of the four lab tasks, by referring to the provided course materials and other relevant materials acquired by self-research and exploration.
  • Collaboratively achieve the learning goals of the final group project by making measurable and substantial contributions to the team's effort.

Discipline-Specific Competencies

Data Analytics, Algorithm Analysis, Pattern Recognition Systems, Text Analytics and Processing

SMU Graduate Learning Outcomes

Self-directed learning

Grading Basis

GRD - Graded

Course Units

1