|
|
|
|
|
Intelligent clustering of data based on TreeSOM
(2006-05-05) (»Add to Infobox)
|
To create and visualise clusters of data, and hence support the automatic structuring of data to thematic areas, an approach has been developed based on the TreeSOM algorithm. The approach extracts keywords from text in a semi-automatic manner, uses those keywords as quality dimension for training SOM:s through the TreeSOM algorithm, and finally utilizes the cluster tree created by the algorithm to visualize the clustering in an interactive real-time manner.
|
|
|
|
Attached below is a paper describing the essence of the research conducted. Attached below is also the source code for the software developed within the project (zip, approx. 300 kb). The code is developed in Java 5.0 with Eclipse Java 3.1 and the Visual Editor plug-in.
Install:
1.
| Download and install Eclipse 3.1 along with the Visual Editor plugin
| 2.
| Start Eclipse and create a new workspace
| 3.
| Choose to import existing projects into workspace
| 4.
| Download the archive with the source code, and import it as an archive
| 5.
| Make sure Eclipse is properly configured to compile for Java 5.0 |
|
| MEMBERS (1/1) |
|
Paper outlining the development work in BioCompetence based on the TreeSOM algorithm.
Source code to the developed interface for clustering.
News
|
|
|
|
|
