Time series prediction using real-time recurrent networks

Show full item record

Title: Time series prediction using real-time recurrent networks
Author: Li, Ruina
Abstract: The purpose of this work is to investigate the possibility of using time series prediction of the Electrocardiogram (ECG ) data by the Real -Time Recurrent Networks (RTRN ) . The RTRN models have been constructed using the Real -Time Recurrent Learning (RTRL ) algorithm with teacher forcing . Both single -point prediction and multi -point prediction were used to forecast the ECG behaviors . The ECG data come from the ECG recordings gathered from a group of patients by the Massachusetts Institute of Technology Division of Health Sciences and Technology . The RTRNs were trained with normal ECG data and were used to predict both normal and abnormal ECG behaviors of the same patient . We found that the single -point prediction of most RTRNs achieved successful results in the forecasting of both normal and abnormal ECG behaviors . However , the multi -point prediction fails to produce the desired results .
URI: http : / /hdl .handle .net /2346 /19988
Date: 1997-05

Citation

Time series prediction using real-time recurrent networks. Master's thesis, Texas Tech University. Available electronically from http : / /hdl .handle .net /2346 /19988 .

Files in this item

Files Size Format View
31295011735460.pdf 16.62Mb application/pdf View/Open

This item appears in the following Collection(s)

Show full item record

Search DSpace

Advanced Search

Browse