Early classification of time series data: overview, challenges, and opportunities

A time series is an ordered sequence of measurements, called data points, recorded over time. Generally, the term time series refers to univariate time series, where only one variable is involved in measurements, such as the temperature of a room, the electrical activity of a patient's heart (electrocardiography), etc. If two or more variables are measured, the time series is called a multivariate time series. For example, when monitoring a patient's health, multiple variables such as temperature, pulse rate, blood pressure, and oxygen rate may be analyzed. Patient information can be captured through multiple sensors attached to the body and decisions can be made by fuzzing data collected through these sensors. Usually, time series are classified when a complete data sequence becomes available. However, time-sensitive applications greatly benefit from early classification. For instance, if a patient's disease is detected early by a series of medical observations, the cost of therapy and the length of the recovery period will be reduced. Additionally, an early diagnosis could save the patient's life by giving health practitioners more time for treatment. Several approaches have been developed to solve early classification problems in various domains, including patient monitoring, human activity recognition, drought prediction, and industrial monitoring. This chapter reviews early classification approaches considering univariate and multivariate time series and guiding future research. It also highlights the importance of data fusion and its strategies with respect to early classification.