INTELLIGENT HYBRID SPATIO-TEMPORAL DATA MINING FOR KNOWLEDGE DISCOVERY ON PROTEOMICS DATA
BibTeX
@MISC{Malone_intelligenthybrid,
author = {James Malone and Ken Mcgarry and Chris Bowerman},
title = {INTELLIGENT HYBRID SPATIO-TEMPORAL DATA MINING FOR KNOWLEDGE DISCOVERY ON PROTEOMICS DATA},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
The need to extract and subsequently represent meaningful knowledge from biomedical data sets is a rapidly growing area of research. Often, such data sets offer further challenges than more traditional analysis, since many domains contain data that is inherently multidimensional and contains spatial and/or temporal elements. This may be further complicated still by the need to incorporate expert’s heuristics in order to perform any meaningful analysis which may not be detected by data driven techniques alone. We present a novel hybrid architecture to perform knowledge discovery on spatiotemporal proteomics data. This architecture employs a combination of data and goal driven elements in order to allow the integration of expert’s opinions within the process. This approach is able to automatically identify proteins exhibiting interesting behaviour from large proteomics data sets. 1.
