Classification of Imbalanced Dementia Historical Data via Machine Learning Technique

Classification of Imbalanced Dementia Historical Data via Machine Learning Technique

碩士 === 東海大學 === 應用數學系 === 107 === The purpose of thesis is to establish CDR score data and screening models for physician diagnosis through different machine learning methods, including classifiers such as Naïve Bayes, Multilayer perceptron, Bootstrap aggregating with decision tree, and support vect...

Full description

Bibliographic Details
Main Authors: DU, SIN-KAI, 杜歆楷
Other Authors: HUANG, HUANG-NAN
Format: Others
Language:zh-TW
Published: 2019
Online Access:http://ndltd.ncl.edu.tw/handle/5cmzx6
id ndltd-TW-107THU00507008
record_format oai_dc
spelling ndltd-TW-107THU005070082019-10-23T05:45:35Z http://ndltd.ncl.edu.tw/handle/5cmzx6 Classification of Imbalanced Dementia Historical Data via Machine Learning Technique 藉由機器學習探究失智症分群的非平衡數據問題 DU, SIN-KAI 杜歆楷 碩士 東海大學 應用數學系 107 The purpose of thesis is to establish CDR score data and screening models for physician diagnosis through different machine learning methods, including classifiers such as Naïve Bayes, Multilayer perceptron, Bootstrap aggregating with decision tree, and support vector machine. The CDR score data are divided into six, five and three stages according to the severity of dementia. Bootstrap aggregating with decision tree classifier is the best among the others. Also, the population sizes for three stages case are obviously imbalanced, SMOTE method is used to adjust a small amount for the normal statege and the study result shows the improvement is rather limited. Finally, principal component analysis (PCA) is carried to to simplify CDR questionares, and certain questions are selected and verified through the reliability analysis. Our study shows that newly added language-assesment questionnaire plays an important role in our analysis. HUANG, HUANG-NAN 黃皇男 2019 學位論文 ; thesis 74 zh-TW
collection NDLTD
language zh-TW
format Others
sources NDLTD
description 碩士 === 東海大學 === 應用數學系 === 107 === The purpose of thesis is to establish CDR score data and screening models for physician diagnosis through different machine learning methods, including classifiers such as Naïve Bayes, Multilayer perceptron, Bootstrap aggregating with decision tree, and support vector machine. The CDR score data are divided into six, five and three stages according to the severity of dementia. Bootstrap aggregating with decision tree classifier is the best among the others. Also, the population sizes for three stages case are obviously imbalanced, SMOTE method is used to adjust a small amount for the normal statege and the study result shows the improvement is rather limited. Finally, principal component analysis (PCA) is carried to to simplify CDR questionares, and certain questions are selected and verified through the reliability analysis. Our study shows that newly added language-assesment questionnaire plays an important role in our analysis.
author2 HUANG, HUANG-NAN
author_facet HUANG, HUANG-NAN
DU, SIN-KAI
杜歆楷
author DU, SIN-KAI
杜歆楷
spellingShingle DU, SIN-KAI
杜歆楷
Classification of Imbalanced Dementia Historical Data via Machine Learning Technique
author_sort DU, SIN-KAI
title Classification of Imbalanced Dementia Historical Data via Machine Learning Technique
title_short Classification of Imbalanced Dementia Historical Data via Machine Learning Technique
title_full Classification of Imbalanced Dementia Historical Data via Machine Learning Technique
title_fullStr Classification of Imbalanced Dementia Historical Data via Machine Learning Technique
title_full_unstemmed Classification of Imbalanced Dementia Historical Data via Machine Learning Technique
title_sort classification of imbalanced dementia historical data via machine learning technique
publishDate 2019
url http://ndltd.ncl.edu.tw/handle/5cmzx6
work_keys_str_mv AT dusinkai classificationofimbalanceddementiahistoricaldataviamachinelearningtechnique
AT dùxīnkǎi classificationofimbalanceddementiahistoricaldataviamachinelearningtechnique
AT dusinkai jíyóujīqìxuéxítànjiūshīzhìzhèngfēnqúndefēipínghéngshùjùwèntí
AT dùxīnkǎi jíyóujīqìxuéxítànjiūshīzhìzhèngfēnqúndefēipínghéngshùjùwèntí
_version_ 1719275554519121920

Similar Items