Welcome to the xSDR page. xSDR is a prototype designed specifically for the deployment and assessment of any dimensionality reduction technique.
xSDR comprises an integrated environment for dimensionality reduction and knowledge discovery that addresses the whole life cycle of data mining. In the current version, it supports communication with a variety of database management systems and integrates a large number of dimensionality reduction algorithms including the MATLAB toolbox for dimensionality reduction. Additionally, it interacts with Weka thus enabling the exploitation of a large number of knowledge mining algorithms.
Finally, xSDR provides an API that enables the integration and evaluation of any dimensionality reduction algorithm. Through this page you may download the application or retrieve additional information related to this effort.
xSDR frontend is implemented in C# while all algorithms are implemented in MATLAB. The interfacing of this technologies is accomplished with the use of the COM Automation Server. DRC's deployment requires MATLAB and .NET framework (version 3.5 SP1) together with Microsoft Chart Controls library.
xSDR is Anastasios Kapernekas' master thesis. It is the evolution of DRC, which was Alexandros Biratsis' master thesis. Both of the projects were developed under the supervision of Panagis Magdalinos.
For any information or question feel free to contact one of us at the following mail addresses:
Tassos Kapernekas
kapernekas-at-aueb-dot-gr
Alexandros Biratsis
alexbir-at-gmail-dot-com
Panagis Magdalinos
pmagdal-at-aueb-dot-gr