| 2005 |
|
YEAR BOOK |
Dublin Institute of Technology
|
Postgraduate Research in the School of Computing
|
Knowledge Management and Database Applications in Healthcare: In the K-Camp, Healthcare Informatics research focuses on developing computing approaches and techniques for supporting data, information and knowledge management by using advanced databases, XML technologies and a hybrid of knowledge management technologies. In an on-going research project led by Bing Wu, Kudakwashe Dube's work focuses on supporting clinical practice guideline (CPG)-based clinical event monitors, alerts and reminders that are integrated into electronic healthcare records and clinical workflow. The project investigates computer-based support for the management of CPGs that are implemented as clinical event monitors, alerts and reminders over an active database-based electronic patient record.
Distributed Management of Business Rules: In distributed information systems, there is a requirement for supporting distributed management of business rules that span participating organisations. Essam Mansour, supervised by Bing Wu within K-Camp, is working on developing the support for the distributed management of business rules that can be specified by using the event-condition-action (ECA) rule paradigm. This project focuses on incorporating ECA rule paradigm support into XML and providing the management support mechanism for the business rules within a distributed and collaborative environment.
Text Classification: SoC research in the area of text classification focuses on using case-based reasoning (CBR) for text classification, a fundamental task in many applications. Text classification is applied to the problem of spam filtering. Sarah-Jane Delany is working on developing a case-based technique for tracking concept drift in spam filtering while Matt Healy, under Sarah-Jane's supervision, is working on the problem of multi-category classification of text using CBR as a machine learning paradigm.
Ontology Versioning: Existing uni-directional mapping between ontology versions cannot satisfy bi-directional mapping requirements of distributed data-sharing and reuse. Siyang Zhao, supervised by Brendan Tierney, is developing a bi-directional ontology version (BOV) mapping that makes it possible to bi-directionally map one ontology version onto another for distributed data sharing and reuse. The BOV mapping will relate similar concepts in two ontology versions in bi-directions and provide the bi-directional transformations between the mapping concepts. This BOV mapping is useful in such situations as that of data sharing between subsidiary companies of an international corporation.
Contact: Dr Bing Wu,
Computer Science Department, School of Computing,
Dublin Institute of Technology, Kevin Street, Dublin 8;
Tel: +353 1 402 4819; E-mail: [email protected]