**PhD Project Topics 2023-2024**

The PhD topics I offer are in the area of neural information retrieval. Below are some ideas that can be tuned to match both our interests.

If you are a self-funded student considering a PhD in any of the topics below please email me to discuss before applying. Click here for information on applying for Research Degrees at Loughborough University.

Available topics:

Topic 1: Temporal and Relation Modelling, Retrieval and Ranking
This project is about the analysis and temporal modelling, retrieval and ranking of information found in articles and reports. Tasks may include:

(1) Temporal Knowledge Graph Reasoning based evolutional entity and relation representations.
(2) Design and develop methods for identifying obsolete/fake/incorrect data and for updating the knowledge of the network.
(3) Develop methods and strategies for continual learning with focus on temporal information updating.
(4) Extend the temporal information search, retrieval and ranking methods with query reformulation and relevance feedback functionality. Temporal modelling of search results.
(5) Develop methods for improving relevance of search results. These can include methods bridging vision and language with structured semantic representations with focus on information retrieval.
(6) The development of an interactive tool that would include the above functionality. Datasets can be research articles, papers, medical books, etc. The tool can be used to process, model, retrieve and visualise information.
Topic 2: Temporal Information Modelling for Retrieval and Summarisation
This project is about the analysis and summarisation of information found in articles and reports. Tasks may include:

(1) Design and develop summarisation models for text and/or multi-modal data (results of user-defined queries).
(2) Develop methods for identifying obsolete/fake/incorrect data from summaries and updating summaries.
(3) Develop methods for evaluating the quality, relevance, and correctness of summaries.
(4) Implementation of methods and strategies for continual learning with focus on summarisation and updating of summaries.
(5) Extend the information search, retrieval and summarisation methods with query reformulation and relevance feedback functionality.
(6) The development of an interactive information search, retrieval and summarisation tool that would also include the above functionality. Datasets can be research articles, papers, medical books, etc. The tool can be used to process, model, retrieve and visualise information.
Topic 3: Multi-modal Transfer Learning for Cross-Modal Information Retrieval
Cross-Modal Retrieval is the task of learning and retrieving data across different modalities, such as from images, video, audio, and text. It is a challenging task to learn representations from different modalities in the shared subspace. The project may focus on:

(1) Developing methods for learning relations of entities in cross-modal spaces comprising images and text. Identifying the limitations of existing cross-modal retrieval algorithms.
(2) Investigating bias in cross-modal transfer learning and developing strategies for detecting and mitigating bias
(3) Proposing and implementing strategies for cross-modal lifelong learning and for forgetting information when required.
(4) Development of a tool for the search, retrieval of multi-modal data. The tool will have query expansion and relevance feedback capabilities.