File(s) under permanent embargo

Reason: Restricted by author. A copy can be supplied under Section 51(2) of the Australian Copyright Act 1968 by submitting a document delivery request through your library, or by emailing document.delivery@monash.edu

Forecasting in Big Data with Recurrent Neural Networks

thesis
posted on 29.10.2020, 01:57 by HERATH MUDIYANSELAGE KASUN GAYAN BANDARA
With the advent of Big Data, many application databases have produced large collections of related time series, which may share key time series properties in common. To exploit these similarities, Global Forecasting Models (GFM) that simultaneously learn from a set of time series have been introduced. This research studies aims to develop a series of novel forecasting methodologies, using Recurrent Neural Networks as the principal forecasting architecture to improve the forecasting accuracy of GFMs in different circumstances. Furthermore, this study demonstrates the empirical evidence of the proposed GFM architectures to address real-world forecasting challenges in the retail and health-care industries.

History

Campus location

Australia

Principal supervisor

Christoph Bergmeir

Additional supervisor 1

Wray Buntine

Additional supervisor 2

Dan Lubman

Year of Award

2020

Department, School or Centre

Clayton School of IT

Additional Institution or Organisation

Clayton School of Information Technology

Course

Doctor of Philosophy

Degree Type

DOCTORATE

Exports