Zpracování a vizualizace dat získaných ze senzorické sítě

Abstract

The thesis deals with the problem of analysis and forecasting of time series data. The aim of the thesis is to investigate existing methods of time series forecasting and their correlations, including the creation of an experimental sensor network, data collection and data preprocessing. Primary methods selected were SARIMA, triple exponential smoothing, clustering and correlation coefficients. Visualization of the obtained data and the results of the analytical methods in the web environment is also part of the work.