Degree: Graduate, M.S.
Nationality: International Students
Application deadlines: 30 June 2021
This studentship is based in one of our Doctoral Training Centres, the Centre for Digital Entertainment (CDE).
AI and machine learning has been used for natural language understanding and generation for many years, which has reshaped the technology landscape. Continuous meaningful conversation generation brings huge challenges which require understanding the dynamic context to keep the conversation flowing. In the long-term, AI driven devices with sophisticated network structures and sensors will grow continuous learning ability to make such devices versatile and adoptive for natural conversations.
The proposed research is about the development of novel technologies for maintaining and evolving a dynamic conversation database to assist the development of conversational AI. The dialog context, including the back and forth dialogs, is preserved in the database for further analysis. The database will help automatically (or semi-automatically) train and drive the conversational AI conversations. Based on a given scenario and common senses, the agents can retrieve dialog information from the database, search for content in the knowledge base system, or generate new conversation threads dynamically, leading to natural conversations.
The key is to effectively generate the conversational databases which support human-like conversations to process and understand multiple intents within one query, including features such as multiple intents per entity and nested intents. A machine learning framework will be developed to offer end-to-end language transforming and attention mechanisms when generating dialog contents. The databases will provide the virtual agents with the capacity of building up understanding, making corrections and updating the context while dialogues continue Such a database will support the interpretation and understanding of emotions, allowing creating dialogs that can read and respond naturally during a conversation.
Project description pdf:
AI and machine learning
Outstanding academic potential as measured normally by either a 1st class honours degree or equivalent Grade Point Average (GPA), or a Master’s degree with distinction or equivalent. If English is not your first language you’ll need IELTS (Academic) score of 6.5 minimum (with a minimum 6.0 in each component, or equivalent).
For more information check out our full entry requirements