Relationship discovery from heterogeneous data.

Many organisations are finding that the volume of information they need to analyse in order to make effective decisions is increasing. An important element in effective decision making is the ability to quickly and accurately prioritise information and identify relationships between important actors and entities.
Prior research on search and recommendation systems has extensively explored the problem of identifying similar documents based on syntax or semantics, but identifying non-similarity relationships, such as the capability to collaborate or fulfil a business need, from unstructured and semi-structured documents of varying formats, intents, and domains remains an open problem.
This project aims to apply recent advances in machine learning, data modelling, and statistical techniques to the problem of identifying the best candidates for formation of new relationships between entities in large heterogeneous datasets based on machine understanding of natural language text.
This project is jointly sponsored by KnowNow Information and the EPSRC as a CASE studentship.