Title: TsP-SA: usage of time series techniques on healthcare data

Authors: Soheila Mehrmolaei; Mohammad Reza Keyvanpour

Addresses: Department of Computer Engineering, Data Mining Laboratory, Alzahra University, Tehran, 0098, Iran ' Department of Computer Engineering, Alzahra University, Tehran, 0098, Iran

Abstract: In the recent years, there has been an increase in the usage of time series techniques on healthcare data. Although much effort has been made to develop techniques of time series, there is a lack of comprehensive categorisation of such techniques to make the possibility of exact study, comparison and assessment of techniques in terms of the ability predicting. We proposed time series prediction-strategy ahead (TsP-SA), a systematic framework which consists of the three main components: categorisation of time series prediction techniques in the context of healthcare, defining general evaluation criteria and analytical evaluation to illustrate a qualitative comparison between each category of techniques which is a proof of the understanding of their supremacy to one another. We believe that using proposed framework as a stimulus can help in the proper selecting technique, efficiency improvement, and development of techniques in researcher's future activities and can provide a helpful platform for the comparative study.

Keywords: time series prediction; TsP-SA; time series prediction-strategy ahead; categorisation of techniques; analytical evaluation; healthcare.

DOI: 10.1504/IJEH.2018.094044

International Journal of Electronic Healthcare, 2018 Vol.10 No.3, pp.190 - 230

Accepted: 27 Mar 2018
Published online: 13 Aug 2018 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article