State of the Art Machine Learning Techniques for Time Series Forecasting: A Survey

Nyein Naing, Wai Yan and Htike@Muhammad Yusof, Zaw Zaw (2015) State of the Art Machine Learning Techniques for Time Series Forecasting: A Survey. In: International Conference on Advances Technology in Telecommunication, Broadcasting, and Satellite, 26-27 September, 2015, Jakarta, Indonesia. (In Press)

PDF Restricted to Repository staff only Download (395kB) | Request a copy

Abstract

Time Series Forecasting is vital for wide range of domains such as financial market forecasting, earthquake forecasting, weather forecasting, electric power demand forecasting and etc. The past 25 years of time series forecasting research that has been reviewed in (Tinbergen Institute Discussion Paper: International Journal of Forecasting) for the period of 1985 to 2005. Therefore, the purpose of my paper is continue to review the recent 10 years of different state of the machine learning techniques for time series forecasting . The main contribution of this paper is to supply researchers with a cohesive overview of state of the art machine learning techniques (during the period of 2005 to 2015) and to identify possible opportunities for future research.

Item Type:	Conference or Workshop Item (UNSPECIFIED)
Additional Information:	6919/48055
Uncontrolled Keywords:	time series forecasting, machine learning techniques, evaluation models, FNN, RNN, SVM, SVR
Subjects:	T Technology > T Technology (General)
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button):	Kulliyyah of Engineering
Depositing User:	Mr. Zaw Zaw Htike
Date Deposited:	22 Jan 2016 15:22
Last Modified:	26 Jun 2018 10:59
URI:	http://irep.iium.edu.my/id/eprint/48055

Actions (login required)

View Item