國立高雄大學圖資館 |

Language: English

Back

基因演算法應用在睡眠階段辨識之腦波特徵值優化 = Application...

傅文享

基因演算法應用在睡眠階段辨識之腦波特徵值優化 = Application of Genetic Algorithms to EEG-Based Feature Optimization for Sleep Staging

Record Type:	Language materials, printed : monographic
Paralel Title:	Application of Genetic Algorithms to EEG-Based Feature Optimization for Sleep Staging
Author:	傅文享,
Secondary Intellectual Responsibility:	國立高雄大學
Place of Publication:	[高雄市]
Published:	撰者;
Year of Publication:	2014[民103]
Description:	87面圖，表 : 30公分;
Subject:	基因演算法
Subject:	Genetic Algorithm (GA)
Online resource:	http://handle.ncl.edu.tw/11296/ndltd/84254551730132383977
Notes:	參考書目：面74-77
Notes:	103年12月16日公開
Summary:	本論文主要是將基因演算法(GA)應用在以腦波進行睡眠便是所需最適特徵值的選取，藉以改善睡眠階段判讀的辨識率。本文中將以過去的相關研究所使用的腦電波圖(EEG)、眼電圖(EOG)及肌電圖(EMG)所計算出的特徵值為根據並以GA挑選出最佳的特徵值來進行睡眠辨識。在研究中，首先我們產生多組隨機基因，每組特徵值依據這些基因挑選出對應的維度，這些經挑選過的特徵值透過碼簿經由K-means演算法的訓練。接著再以該訓練過的碼簿為基礎，以信號特徵值用來訓練睡眠階段判讀的DHMM模型，訓練過的DHMM模型經過判讀測試統計出辨識率後，以GA選取辨識率較高的組合來做交配及突變。經由上述的這些步驟來讓GA不斷的演化產生準確率更高的基因組合。在本論文中的實驗結果顯示了藉由GA來進行特徵值選取可使整體睡眠狀態判讀率有相當程度的提升。 This thesis used Genetic Algorithm (GA) to select some suitable features calculated EEG, EOG and EMG for sleep staging, such that the recognition rate of sleep staging can be improved. The features developed in previous literatures are adopted in this research. First, a set of chromosomes with binary-valued genes are randomly generated. The features corresponding to the genes with value 1 in a chromosome are selected. The selected features are then used to train a codebook by k-means algorithm. Thereafter, a DHMM is trained by using the codebook for all testing features. Finally, the trained DHMM model is then used for sleep staging. Through the evolution of GA and repeat the above procedure, better chromosomes or more suitable features are obtained. It can be seen from the experiments in this thesis, the proposed method can obtain better recognition rate for sleep staging.

基因演算法應用在睡眠階段辨識之腦波特徵值優化 = Application of Genetic Algorithms to EEG-Based Feature Optimization for Sleep Staging
傅, 文享

基因演算法應用在睡眠階段辨識之腦波特徵值優化 = Application of Genetic Algorithms to EEG-Based Feature Optimization for Sleep Staging / 傅文享撰 - [高雄市] : 撰者, 2014[民103]. - 87面 ; 圖，表 ; 30公分.
參考書目：面74-77103年12月16日公開.
基因演算法Genetic Algorithm (GA)

基因演算法應用在睡眠階段辨識之腦波特徵值優化 = Application of Genetic Algorithms to EEG-Based Feature Optimization for Sleep Staging
LDR:03103nam0a2200289 450 001 430135
005 20170214094803.0
009 430135
010 0 $b 精裝
010 0 $b 平裝
100 $a 20170214d2014 k y0chiy05 e
101 1 $a chi $d chi $d eng
102 $a tw
105 $a ak am 000yy
200 1 $a 基因演算法應用在睡眠階段辨識之腦波特徵值優化 $d Application of Genetic Algorithms to EEG-Based Feature Optimization for Sleep Staging $z eng $f 傅文享撰
210 $a [高雄市] $c 撰者 $d 2014[民103]
215 0 $a 87面 $c 圖，表 $d 30公分
300 $a 參考書目：面74-77
300 $a 103年12月16日公開
314 $a 指導教授：潘欣泰博士
328 $a 碩士論文--國立高雄大學資訊工程學系碩士班
330 $a 本論文主要是將基因演算法(GA)應用在以腦波進行睡眠便是所需最適特徵值的選取，藉以改善睡眠階段判讀的辨識率。本文中將以過去的相關研究所使用的腦電波圖(EEG)、眼電圖(EOG)及肌電圖(EMG)所計算出的特徵值為根據並以GA挑選出最佳的特徵值來進行睡眠辨識。在研究中，首先我們產生多組隨機基因，每組特徵值依據這些基因挑選出對應的維度，這些經挑選過的特徵值透過碼簿經由K-means演算法的訓練。接著再以該訓練過的碼簿為基礎，以信號特徵值用來訓練睡眠階段判讀的DHMM模型，訓練過的DHMM模型經過判讀測試統計出辨識率後，以GA選取辨識率較高的組合來做交配及突變。經由上述的這些步驟來讓GA不斷的演化產生準確率更高的基因組合。在本論文中的實驗結果顯示了藉由GA來進行特徵值選取可使整體睡眠狀態判讀率有相當程度的提升。 This thesis used Genetic Algorithm (GA) to select some suitable features calculated EEG, EOG and EMG for sleep staging, such that the recognition rate of sleep staging can be improved. The features developed in previous literatures are adopted in this research. First, a set of chromosomes with binary-valued genes are randomly generated. The features corresponding to the genes with value 1 in a chromosome are selected. The selected features are then used to train a codebook by k-means algorithm. Thereafter, a DHMM is trained by using the codebook for all testing features. Finally, the trained DHMM model is then used for sleep staging. Through the evolution of GA and repeat the above procedure, better chromosomes or more suitable features are obtained. It can be seen from the experiments in this thesis, the proposed method can obtain better recognition rate for sleep staging.
510 1 $a Application of Genetic Algorithms to EEG-Based Feature Optimization for Sleep Staging $z eng
610 0 $a 基因演算法 $a 腦電波圖 $a 離散隱藏式馬可夫模型 $a 睡眠狀態判讀
610 1 $a Genetic Algorithm (GA) $a EEG $a sleep staging $a Discrete Hidden Markov Model (DHMM)
681 $a 008M/0019 $b 464103 2300 $v 2007年版
700 1 $a 傅 $b 文享 $4 撰 $3 673359
712 0 2 $a 國立高雄大學 $b 資訊工程學系碩士班 $3 353878
801 0 $a tw $b NUK $c 20141021 $g CCR
856 7 $z 電子資源 $2 http $u http://handle.ncl.edu.tw/11296/ndltd/84254551730132383977