Multivariate Time Series Data Transformation for Convolutional Neural Network
碩士 === 國立臺灣科技大學 === 工業管理系 === 106 === This thesis proposes a novel framework to encode time series data into two-dimensional images, and aggregate the images into one image for each batch of data. After transformation and aggregation, the images passed through a convolutional neural netw...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | en_US |
Published: |
2018
|
Online Access: | http://ndltd.ncl.edu.tw/handle/sd989v |
id |
ndltd-TW-106NTUS5041077 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-106NTUS50410772019-05-16T00:59:40Z http://ndltd.ncl.edu.tw/handle/sd989v Multivariate Time Series Data Transformation for Convolutional Neural Network 多維時間序列轉換應用於卷積神經網路 Chen-Yi YANG 楊晨宜 碩士 國立臺灣科技大學 工業管理系 106 This thesis proposes a novel framework to encode time series data into two-dimensional images, and aggregate the images into one image for each batch of data. After transformation and aggregation, the images passed through a convolutional neural network, which is outstanding in dealing with computer vision problems. The convolutional neural network learned the image features and predict the status. This study applied four methods to encode time series into images: Gramian Angular Summation Field (GASF), Gramian Angular Difference Field (GADF), Markov Transition Field (MTF), and Recurrence Plot (RP). The overlaying and appending methods to aggregate the images were validated and evaluated using open datasets. The results of the experiments find that encoding time series data into images and aggregating the images by the appending method are helpful in increasing prediction accuracy. Chao-Lung Yang 楊朝龍 2018 學位論文 ; thesis 47 en_US |
collection |
NDLTD |
language |
en_US |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立臺灣科技大學 === 工業管理系 === 106 === This thesis proposes a novel framework to encode time series data into two-dimensional images, and aggregate the images into one image for each batch of data. After transformation and aggregation, the images passed through a convolutional neural network, which is outstanding in dealing with computer vision problems. The convolutional neural network learned the image features and predict the status. This study applied four methods to encode time series into images: Gramian Angular Summation Field (GASF), Gramian Angular Difference Field (GADF), Markov Transition Field (MTF), and Recurrence Plot (RP). The overlaying and appending methods to aggregate the images were validated and evaluated using open datasets. The results of the experiments find that encoding time series data into images and aggregating the images by the appending method are helpful in increasing prediction accuracy.
|
author2 |
Chao-Lung Yang |
author_facet |
Chao-Lung Yang Chen-Yi YANG 楊晨宜 |
author |
Chen-Yi YANG 楊晨宜 |
spellingShingle |
Chen-Yi YANG 楊晨宜 Multivariate Time Series Data Transformation for Convolutional Neural Network |
author_sort |
Chen-Yi YANG |
title |
Multivariate Time Series Data Transformation for Convolutional Neural Network |
title_short |
Multivariate Time Series Data Transformation for Convolutional Neural Network |
title_full |
Multivariate Time Series Data Transformation for Convolutional Neural Network |
title_fullStr |
Multivariate Time Series Data Transformation for Convolutional Neural Network |
title_full_unstemmed |
Multivariate Time Series Data Transformation for Convolutional Neural Network |
title_sort |
multivariate time series data transformation for convolutional neural network |
publishDate |
2018 |
url |
http://ndltd.ncl.edu.tw/handle/sd989v |
work_keys_str_mv |
AT chenyiyang multivariatetimeseriesdatatransformationforconvolutionalneuralnetwork AT yángchényí multivariatetimeseriesdatatransformationforconvolutionalneuralnetwork AT chenyiyang duōwéishíjiānxùlièzhuǎnhuànyīngyòngyújuǎnjīshénjīngwǎnglù AT yángchényí duōwéishíjiānxùlièzhuǎnhuànyīngyòngyújuǎnjīshénjīngwǎnglù |
_version_ |
1719172236772900864 |