|Teaching Staff||V. Megalooikonomou|
The evolution of technology has contributed significantly to the accumulation of huge volumes of data. The course aims to study techniques of management, and analysis of large databases that have, among others, spatial and temporal components. The purpose of analyzing this data is to understand patterns, find similarities, identify correlations, normalities, and anomalies. Management techniques are essential for efficient data processing and storage. This kind of data is collected daily by organizations, research centers, hospitals, businesses, etc.Due to the nature of this data, applications vary, such as diagnostics in medicine, business / stock forecasting and decision support, etc. The course examines, among other things, the applications of these techniques to biomedical databases.
Data preprocessing, data cleansing, feature extraction, feature selection; Singular Value Decomposition; introduction to basic signal processing methods (DFT, wavelets), data compression (scalar and vector quantization, lossless and lossy compression); extraction of knowledge from spatial and temporal databases; clustering, classification, prediction, decision trees, association mining, Bayesian networks; spatial access methods (k-d trees, quadtrees, z-ordering, space filing curves, R-trees); general purpose multimedia indexing, GEMINI; spatial and temporal databases; techniques for searching by content in multimedia databases (time series, images, videos); fractals in databases; self-similarity of data; fractal dimension; applications in biomedical databases; data stream management and analysis.