Clustering time-series. An overview about different application contexts of time-series clustering

R ecp package

Abstract

Time-series are becoming more and more important in the digitized industry 4.0. from forecasting of sales to increase the profit in retail industry, to real time streamed analysis for fraud detection, intrusion-detection, to medical applications e.g. combination of different time series (ECG, Blood, …) for improvement of diagnoses, to applications in stock market. This work presents an overview of different application contexts of time-series clustering as a very hands-on, tutorial-like approach. Clustering time-series is often used to gain insight into the generating mechanism of the data in order to predict future values.

Publication
Clustering time-series, BSc Thesis
Avatar
Georg Heiler
PhD candidate & data scientist

My research interests include large geo-spatial time and network data analytics.