
|
|
| Position | Professor of Computer Science |
| Research Area | Compilers |
| Telephone | +44 1784 443427 |
| Publications |
Professor Scott's current research interests lie in Theoretical Computer Science, particularly in language design and compiler theory. Her recent work centres on the formal properties of, and applicability of, generalised parsing. There has been a resurgence of interest in general parsing techniques because modern computer hardware is sufficiently powerful to enable such techniques to be used. Together with Adrian Johnstone, Professor Scott has developed correct, efficient algorithms for parsing general context free languages, extending Tomita's epsilon-free algorithm to include epsilon rules and Aycock and Horspool's algorithm to grammars with hidden left recursion. She is presently investigating their application to computer language parsing; compiler code generation; reverse compilation and bioinformatics, as well as the traditional domain of natural language parsing.
Professor Scott's early research in Computer Science was in the area of formal methods and automated theorem proving. She has worked on the general theory of well founded orderings in order to address the problem of proving process termination. In particular she has given a classification for the case of strings on two letters in terms of the order type of the ordering, and she has given an automated proof of the correctness of a compiler as part of the ESPRIT PROCOS initiative.
Professor Scott also has an extensive research history in Pure Mathematics, carried out while she was at Oxford University and at the Institute for Advanced Studies in Canberra. Her work involved computability and unsolvability results in group theory. She has constructed what is still the only known example of a finitely presented simple group with unsolvable conjugacy problem. She has also published an Oxford University Press monograph on Existentially Closed Groups, which addresses an area on the borders between algebra and logic and uses algebraic techniques to prove results which could not be derived using a purely logic based approach.