| EPSRC Reference: |
GR/M43975/01 |
| Title: |
FARADAY PARTNERSHIP:INTELLIGENT DATA ANALYSIS & FUSION TECHNIQUES IN PHARMACEUTICALS,BIOPROCESSING & PROCESS |
| Principal Investigator: |
Buxton, Professor BF |
| Other Investigators: |
|
| Researcher Co-Investigators: |
|
| Project Partners: |
|
| Department: |
Computer Science |
| Organisation: |
UCL |
| Scheme: |
Faraday (PreFEC) |
| Starts: |
04 January 1999 |
Ends: |
03 July 2002 |
Value (£): |
211,733
|
| EPSRC Research Topic Classifications: |
| Information & Knowledge Mgmt |
|
|
| EPSRC Industrial Sector Classifications: |
| Manufacturing |
Pharmaceuticals and Biotechnology |
|
| Related Grants: |
|
| Panel History: |
|
|
Summary on Grant Application Form |
|
Abstract at time of the proposal:The aim of this project is to show how hybrid intelligent systems techniques recently employed for processing large data sets and data mining in retail and finance businesses can be developed for, and transferred to, some of the major industrial sectors participating in the NPL-Sira INTErSECT Faraday Partnership. The objectives are to develop and apply techniques derived from our previous work at UCL together with recent theoretical developments in neural networks and computational learning to the analysis and management of large data sets in pharmaceuticals, bioprocessing, and process control. Research will be driven by the needs of the Faraday partners, in particular, to address problems such as: understanding of structure-activity relationships in drug design (Glaxo-Wellcome); the optimization of high throughput screening processes (SmithKline Beecham); and understanding the relationships between production parameters, sensory assessments, product appearance, and consumer preferences (Unilever). The ultimate objective is to produce an integrated system that selects the most appropriate hybrid techniques for a particular problem and can use data fusion to combine their results in the most intelligent way. ISL, a leading developer and supplier of data mining and decision support software, such as Clementine, to this sector aims to incorporate the results in future products.New AbstractThe aim of this project is to show how hybrid intelligent systems techniques recently employed for processing large data sets and data mining in retail and finance businesses can be developed for, and transferred to, some of the major industrial sectors participating in the NPL-Sira INTErSECT Faraday Partnership. The objectives are to develop and apply techniques derived from our previous work at UCL together with recent theoretical developments in neural networks and computational learning, in particular, on the development of large margin classifiers and on classifier combination, to the analysis and management of large data sets in pharmaceuticals, bioprocessing, and process control. Research will be driven by the needs of the Faraday partners, in particular, to address problems such as: understanding of structure-activity relationships in drug design (Glaxo-Wellcome); the optimization of high throughput screening processes (SmithKline Beecham); and understanding the relationships between production parameters, sensory assessments, product appearance, and consumer preferences (Unilever). The ultimate objective was to produce a system that selects the most appropriate classifiers for a particular problem and can use data fusion techniques to combine their results in the most intelligent way. The Wilcoxon statistic derived from classifier ROC curves provides a robust measure of performance for the classifier combination which is carried out by genetic programming. The system developed has been test on publically available data sets and on drug screening data. It is currently being further evaluated in collaboration with GIaxoSmithKline.
|
|
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: |
|