Don't miss our weekly PhD newsletter | Sign up now Don't miss our weekly PhD newsletter | Sign up now

  Towards Machine Understanding of Human Languages: AI for Storytelling Across Languages.

   Graduate School

  Dr Javad Zarrin, Dr Kean Lee Kang, Dr Darshana Jayemanne  Thursday, August 15, 2024  Funded PhD Project (Students Worldwide)

About the Project

In the realm of AI and computer science, disciplines such as Natural Language Processing (NLP), Human-Computer Interaction (HCI), and AI-centric Language Models are critical for fostering communication and collaboration between humans and machines in the future. Despite the progress made in AI and Machine Learning, existing models still struggle to achieve a deep and precise grasp of natural languages. Although there has been remarkable progress in developing more efficient models for various NLP tasks, Deep NLP models, transformers, and large-scale language models like ChatGPT still face a significant challenge in generating engaging and coherent stories. The existing storytelling models suffer from a lack of coherence and contextuality, which can be attributed to the narrow focus on horizontal sequence-based processing. Storytelling models face challenges in coherence and contextuality due to their emphasis on horizontal sequence-based processing and struggle with narrative intricacies, demanding the incorporation of innovative AI techniques for a deeper grasp of narrative components.

Enhancing modern Deep NLP techniques requires exploring and capturing the complex interactions among words, considering diverse categories of orthographic and phonological relations, as well as the relationships between words and their contextual surroundings across a broad spectrum of texts. Through meticulous scrutiny of the multifarious ways in which words are employed and their interconnectivity, a more profound comprehension of the underlying implications and concepts can be attained. Utilising these insights enables the enhancement of current approaches. The project aims to improve machine understanding of natural languages by developing novel methods for AI-driven storytelling. The work will be focused on designing and developing methods and techniques to analyse and explore large textual datasets and lexical graph databases, discover, and extract the hidden pattern and relationship between words within a language and automatically generate stories describing and explaining those relationships. The project will combine and incorporate approaches from various disciplines including NLP, Deep Neural Networks, Computational Reasoning, Computational Linguistics, Generative AI, Graph Learning and Graph Neural Networks.

Supervisory Team: The candidate will be supervised within the School of Design and Informatics by Dr Javad Zarrin, Dr Kean Lee Kang, and Dr Darshana Jayemanne. Queries on this project should be directed to Dr. Javad Zarrin ().

Entry Requirements: Candidates must have, or expect to obtain, a first-class or upper second-class honours degree or international equivalent (preferably with a recognised master's degree) in Artificial Intelligence, Data Science, Computer Science, Mathematics, or a closely related discipline.

The successful candidates will have a solid background in AI, with experience spanning machine learning, graph learning, natural language processing, and generative AI. Experience in designing and conducting AI-driven research, coupled with a thorough grasp of research methodologies, AI-related mathematical concepts, and practical skills in Python and data analysis techniques, is highly regarded. Familiarity with storytelling and gaming is a plus but not a mandatory requirement.

For applicants who are non-native speakers of English, the University requires an IELTS of 6.5 (with no band less than 6.0) or an equivalent qualification accepted by the Home Office.

Computer Science (8) Mathematics (25)

Funding Notes

R-LINCS2 funded. The Studentship is available for a June, October 2024, or February 2025 start.

A PhD studentship that comprises tax-free stipend of £17,668 (increasing in line with UKRI per annum) per year over 3.5 years, tuition fees paid, and a generous study package (e.g. limited research consumables, travel budget, and training when appropriate).  

Register your interest for this project

Search Suggestions
Search suggestions

Based on your current searches we recommend the following search filters.