Research Area:  Machine Learning
A time series represents a collection of data points captured over time.This type of data is actively studied in many domains of application, such as healthcare,finance,energy, or climate.The generalized interest in time series arises from the dynamic characteristics of many real-world phenomena, where events naturally occur and evolve over time.Uncertainty is a significant issue when analyzing time series, which complicates the accurate understanding of their future behavior.To cope with this problem, organizations engage in forecasting to drive their decision-making process. Forecasting denotes the process of predicting the future behavior of time series,which allows professionals to anticipate scenarios and take proactive measures.In this context,the aim of this thesis is to advance the state of the art of the literature in time series forecasting.Particularly,our research goalcan be divided into two main parts:(i) forecasting the future numeric valuesof time series; and (ii) the anticipation of interesting events in time series in atimely manner, a task that is commonly known as activity monitoring.In bothparts, we adopt an ensemble learning approach, the field of machine learningthat combines different predictive models to address a given predictive task.
Name of the Researcher:  Vitor Manuel Ara ́ujo Cerqueira
Name of the Supervisor(s):   Lu ́ıs Fernando Rainho Alves Torgo
Year of Completion:  2019
University:  À FACULDADE DE ENGENHARIA DA UNIVERSIDADE DO PORTO
Thesis Link:   Home Page Url