My areas of interest include Data Mining and Machine Learning in general. More specifically I am interested in their application for time series analysis. Some sub-areas in which I have worked and I am particularly excited about are:

Mining motifs and discords from large time series data sets

Manifold learning from high dimensional data (time series in particular) and in the presence of noise

Nonlinear dimensionality reduction techniques, again with focus on time series data

Nearest neighbor approaches for classification, anomaly detection and forecasting

Ensemble methods

Dissertation

Learning from Time Series in the Presence of Noise: Unsupervised and Semi-supervised Approaches [pdf]

D. Yankov, E. Keogh, L. Wei, X. Xi: Fast best-match shape searching in rotation invariant metric spaces (extended version). To appear in IEEE Transactions on Multimedia, Special Issue on Data Mining, 2007

D. Yankov, E. Keogh: Manifold clustering of shapes. Proceedings of the 6th IEEE International Conference on Data Mining (ICDM 2006), p. 1167-1171 [Powerpoint]

D. Yankov, D. DeCoste, E. Keogh: Ensembles of nearest neighbors forecasts. 17th European Conference on Machine Learning (ECML 2006), Proceedings. Lecture Notes in Computer Science, p. 545-556 [Powerpoint]

D. Yankov, E. Keogh, S. Lonardi, A. Fu: Dot plots for time series analysis. 17th IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2005), p. 159-168 [Powerpoint]

Projects

A list for some of the projects that I have completed during the last few years can be found here.