javialonsaso/TFM2020: Master Thesis
javialonsaso · Zenodo (CERN European Organization for Nuclear Research) · 2020
Time series analysis is a fundamental field for information processing. In the real world, the context of the data is often unknown and the experimenter must make an effort to extract limited information. In this project, we study a set of unsupervised algorithms for the analysis of time series corresponding to the water consumption of a residential area. In particular, the algorithms focus on anomaly detection, dimensionality reduction and data clustering. In a series of experiments, the results obtained by the different algorithms in each category are compared.