Introduction
This website provides several demos for recently completed or ongoing projects supported by TUBITAK 113E517 ``Big Data Adaptive Signal Processing Algorithms" under the supervision of Assoc. Prof. Dr. Suleyman Serdar Kozat at Bilkent University, Ankara, Turkey. You can select one of the projects from the list on the left hand side to access a brief explanation about the project and an executive simulation file for download. The simulation files are provided with a user friendly graphical user interface. You can set parameters for both the algorithm and the data, and run the simulation.
About Us
We are a research team, working under Dr. Kozat supervision in Bilkent University, Ankara, Turkey. Dr. Kozat conducts research broadly, but not limited to, in the following topics: Machine Learning for Signal Processing, Big Data and Analytics, Anomaly Detection, Signal processing theory, Adaptive signal processing, Our recent researches focuses on the online nonlinear regression problem. We provide several efficient algorithms for online nonlinear regression, which provide satisfactory results with significantly reduced computational complexity.
Please feel free to visit Assoc. Prof. Dr. Suleyman Serdar Kozat web page.Project1 description
In this section, we demonstrate the performance of Lexicographical Model Weighting algorithm, over a synthetic piecewise linear data set. The following program gets the parameters of the data set and the algorithm from user, runs LMW algorithm and two of its natural competitors over the data set, and plots the error curves. Detailed description of the model is provided in the executive file.
Click here to download the program for Lexicographical Project!Contact
Bilkent University, Ankara, Turkey