| EPSRC Reference: |
GR/J54956/01 |
| Title: |
SNAP- A SIMPLE NATURAL LANGUAGE ACCESS PLATFORM TO STRUCTURED AND TEXTUAL DATA. IED4/1/5837 |
| Principal Investigator: |
De Roeck, Professor A |
| Other Investigators: |
|
| Researcher Co-Investigators: |
|
| Project Partners: |
|
| Department: |
Computer Sci and Electronic Engineering |
| Organisation: |
University of Essex |
| Scheme: |
Standard Research (Pre-FEC) |
| Starts: |
01 September 1994 |
Ends: |
31 August 1997 |
Value (£): |
188,067
|
| EPSRC Research Topic Classifications: |
| Human Communication in ICT |
|
|
| EPSRC Industrial Sector Classifications: |
|
| Related Grants: |
|
| Panel History: |
|
|
Summary on Grant Application Form |
|
Often information about a single domain (e.g. personnel management) is held both in databases and text, which must be retrieved through different means. SNAP aims to design and develop a portable platform from which both structured and textual data pertaining to a domain can be accessed uniformly through natural language queries. This will be achieved by integrating a Natural Language Front End (SQUIRREL) and a Text Retrieval System (STATUS/IQ) behind a suitable Graphical User Interlace, and developing appropriate customisation tools. Findings may impact on Open Systems Design. A very high degree of user involvement is envisaged. Progress:Traditionally, database interrogation and text retrieval rely on fundamentally different language processing techniques. For database interrogation, it is imperative that the query is analysed in depth to help ensure users retrieve the information they intended. Text retrieval on the other hand is usually a hit and miss operation using shallow processing techniques. The problems are (i) how to design a platform from which two systems with different functionality can operate and (ii) how to instil cohesion in the overall systems behaviour w.r.t. the domain covered by the data. The former issue is a question of design - the latter also touches on terminology and appropriate customisation tools. Heavy user involvement is envisaged to evaluate the platform and its interface, and the adequacy of the customisation tools. In order to fit in at least two user evaluation cycles, the whole project is based around successive generations of demonstrators called Frankenstein . The project has just passed the 6 month mark and the first phase has been completed. Frankenstein Mark I is in place and we are assessing the design. We are currently drafting specifications of customisation tools before executing a test port to a real domain. The panel of users is being assembled and we have started preparing for the first round of user evaluation.
|
|
Key Findings |
|
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
|
|
Potential use in non-academic contexts |
|
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
|
|
Impacts |
|
| Description |
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk |
| Summary |
|
| Date Materialised |
|
|
|
|
Sectors submitted by the Researcher |
|
This information can now be found on Gateway to Research (GtR) http://gtr.rcuk.ac.uk
|
| Project URL: |
|
| Further Information: |
|
| Organisation Website: |
http://www.sx.ac.uk |