Fusion of statistical and machine learning approaches for time series prediction using earth observation data

被引：15

作者：

Agrawal K.P. ^{[1
]}

Garg S. ^{[1
]}

Sharma S. ^{[2
]}

Patel P. ^{[1
]}

Bhatnagar A. ^{[3
]}

机构：

[1] Department of Computer Science and Engineering, Institute of Technology, Nirma University, Ahmedabad, Gujarat

[2] Space Application Centre (ISRO), Ahmedabad, Gujarat

[3] Department of Computer Science and Engineering, Indian Institute of Technology (IIT) Bombay, Mumbai, Maharashtra

来源：

International Journal of Computational Science and Engineering | 2017年 / 14卷 / 03期

关键词：

ARIMA; Auto regressive integrated moving average; Prediction; Scalability; Support vector regression; SVR; Time series;

D O I：

10.1504/IJCSE.2017.084159

中图分类号：

学科分类号：

摘要：

This paper focuses on fusion of statistical and machine learning models for improving the accuracy of time series prediction. Statistical model like integration of auto regressive (AR) and moving average (MA) is capable to handle non-stationary time series but it can deal with only single time series. The machine learning approach [i.e. support vector regression (SVR)] can handle dependency among different time series along with nonlinear separable domains, however it cannot incorporate the past behaviour of time-series. This led us to hybridise auto regressive integrated moving average (ARIMA) with SVR model, where focus has been given on minimisation of forecast error using residuals. Issue related to scalability has been handled by taking suitable representative samples from each sub-areas which helped in reducing number of models and drastic reduction in time for tuning different parameters. Results obtained show that the performance of proposed hybrid model is better than individual models. Copyright © 2017 Inderscience Enterprises Ltd.

引用

页码：255 / 266

页数：11

共 18 条

[11]

Munawar M.A., Ward P.A.S., Leveraging many simple statistical models to adaptively monitor software systems, Int. J. of High Performance Computing and Networking, 7, 1, pp. 29-39, (2011)

[12]

Munoz-Mari J., Bruzzone L., Camps-Vails G., A support vector domain description approach to supervised classification of remote sensing images, Geoscience and Remote Sensing, IEEE Transactions on, 45, 8, pp. 2683-2692, (2007)

[13]

Niazmardi S., Shang J., A new classification method based on the support vector regression of NDVI time series for agricultural crop mapping, Second International Conference on Agro-Geoinformatics, pp. 361-364, (2013)

[14]

Omuto C.T., Vargas R.R., Alim M.S., Paron P., Mixed-effects modeling of time series NDVI-rainfall relationship for detecting human-induced loss of vegetation cover in dry lands, Journal of Arid Environments, 74, 11, pp. 1552-1563, (2010)

[15]

Smola A., Scholkopf B., A tutorial on support vector regression, Statistics and Computing, 14, 3, pp. 199-222, (2004)

[16]

Vapnik V., The Nature of Statistic Learning Theory, (1995)

[17]

Zhang L., Lei L., Yan D., Comparison of two regression models for predicting crop yield, Geoscience and Remote Sensing, pp. 1521-1524, (2010)

[18]

Zhu L., Xu S., Prediction algorithm based on web mining for multimedia objects in next-generation digital earth, Int. J. of Embedded Systems, 7, 1, pp. 79-87, (2015)

← 1 2 →