Time Series Forecasting Techniques for Infectious Disease Prediction

Abstract

This chapter provides an overview of the Time Series Forecasting techniques for infectious disease prediction with a focus on enabling technologies, protocols, and implementation issues. Recent developments in RFID, smart sensors, communications technologies, and Internet protocols enable the latest technological advances in the field of epidemiology. The basic idea of this chapter is to have smart sensors working directly to achieve a new class of autonomous applications without human involvement. One of the significant challenges in the medical domain is the effective handling of large volumes of time series data and the associated algorithms. Apart from the sheer volume of data, the diversity of data sources gives rise to considerable complexity: without comprehensive semantic meaning data on signals, objects, and their location, users cannot comprehend the raw time-series data. This further augments the complexity by considering predictive models for capturing regularities and time series statistical patterns, often as a function of other time series influenced through co-location and physical processes. 270This chapter outlines a detailed survey of the time series forecasting techniques in the context of medical applications.