论文总字数:32788字
题 目_基于复杂网络特征描述的脑电信号分类算法研究_
_______数学_____院(系)____统计学___专业
学 号________ 07315103_______________
学生姓名________ 章舒江________________
指导教师_________刘庆山________________
起止日期___2019.01.01—2019.05.30________
设计地点______东南大学数学学院_________
\def\CTeXPreproc{Created by ctex v0.2.12, don't edit!}
\documentclass[oneside,openany,12pt]{cctbook}
\zihao{3}\ziju{0.15}
\pagestyle{myheadings} \textwidth 16.0cm \textheight 22 truecm
\special{papersize=21.0cm,29.7cm} \headheight=1.5\ccht
\headsep=20pt \footskip=12pt \topmargin=0pt \oddsidemargin=0pt
\setcounter{section}{0}
\frontmatter
\def\nn{\nonumber}
\newcommand{\lbl}[1]{\label{#1}}
\newcommand{\bib}[1]{\bibitem{#1} \qquad\framebox{\scriptsize #1}}
\renewcommand{\baselinestretch}{1.5}
\newtheorem{theorem}{定理}
\newtheorem{proposition}{命题}
\newtheorem{lemma}{引理}
\newtheorem{remark}{Remark}
\newtheorem{corollary}{推论}
\newtheorem{defi}{定义}
\newcommand{\be}{\begin{equation}}
\newcommand{\ee}{\end{equation}}
%\def\c{\theta}
\newcounter{local}
\newcounter{locallocal}
\newcommand{\scl}{\stepcounter{local}}
\setcounter{local}{0}
%\renewcommand{\theequation}{\arabic{chapter}.\arabic{section}.\arabic{local}}
\renewcommand{\theequation}{\arabic{chapter}.\arabic{equation}}
%\renewcommand{\theequation}{\arabic{local}.\arabic{local}}
\def\s#1{\setcounter{local}{#1}}
%\usepackage[nooneline,center]{caption2}
%\usepackage[dvips]{graphics,color}
\usepackage{Picinpar}
\usepackage{amsmath,amssymb}
\usepackage{graphicx}
\usepackage{subfigure}
\usepackage{flafter}
\usepackage{fancyhdr}
\usepackage{mathrsfs}
\usepackage{makecell}
\usepackage{multirow}
%%%%%%%%%%%%%%%%%%%%%%%%%%%设置页眉双下划线%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\newcommand{\makeheadrule}{%
\makebox[0pt][l]{\rule[0.55\baselineskip]{\headwidth}{0.4pt}}%
\rule[0.7\baselineskip]{\headwidth}{0.4pt}}
\renewcommand{\headrule}{%
{\if@fancyplain\let\headrulewidth\plainheadrulewidth\fi
\makeheadrule}}
\makeatother
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\pagestyle{fancy}
\renewcommand{\chaptermark}[1]{\markboth{\kaishu{\chaptername}~~~#1~}{}}
\fancyhead[l]{\kaishu{~~~东~南~大~学~本~科~ 毕~业~论~文}}
\fancyhead[c]{}
\fancyhead[r]{\leftmark}
\fancyfoot[l]{}
\fancyfoot[c]{\thepage}
\fancyfoot[r]{}
\begin{document}
\begin{titlepage}
\end{titlepage}
\frontmatter
\setlength{\footskip}{20pt}
\begin{center}{\kaishu \zihao{2}{基于复杂网络特征描述的
脑电信号分类算法研究}}\end{center}
\vskip 0.5cm
\begin{center}{\kaishu\zihao{4} 摘\ \ \ \ 要}
\end{center}
\addcontentsline{toc}{chapter}{摘\ \ \ \ 要} {\kaishu \ \
脑电信号分类一直是一个热门研究问题,使用传统的机器学习和深度学习的方法进行数据分析已经不能满足当代人类的需要。复杂网络由于其稳定性强、运行强健等优势已经在数据分析领域崭露头角,因此本文将其应用于脑电信号分类的研究。
本文通过提取工作记忆实验脑电数据复杂的网络特征,完成对脑电信号的分类识别,主要的创新工作如下:
首先,提出了一种基于神经网络分类器的单阈值模型。构建初始模型,对原始数据的波动情况进行了研究,选择了时域特征进行了提取。在确定了单阈值的邻接矩阵模型之后对三种不同的特征提取方法进行了试验,结果显示合理的最佳方案是选择复杂网络度的分布作为样本数据的特征。
其次,在上一种单阈值模型的基础上提出了多阈值进行分类。模型构建过程基本一致,但是欧式距离矩阵的密度分布带来了新的思考与启发,选择单阈值可能会遗漏很多的结构特性。本文尝试了均匀设置多个阈值点和根据密度图的特性不均匀设置阈值两种方法,以更好地表达结构特性。
最后,在应用不均匀多阈值联合复杂网络模型的情况下,本文得到了最优分类结果,证明了提出的方法比深度学习方法更有效,并将分类误差降低到5.75\%。虽然模型泛化能力不足,但采用复杂网络方法提取EEG数据特征的方法给模型分类算法提供了一种新的尝试。
}
\vskip 1cm \noindent{\kaishu 关键词: \认知负荷,\脑电信号,\网络特征,\多阈值,\复杂网络 }
\newpage
\thispagestyle{plain}
\begin{center}{\rm EEG Signal Classification Algorithm Based on Complex Network Feature Description}\end{center}
\vskip 0.5cm
\begin{center}{\rm\zihao{4} Abstract}
\end{center}
\addcontentsline{toc}{chapter}{Abstract}
\par
The classification of EEG signals has always been a very important research topic in the field of brain science. The use of traditional machine learning and deep learning methods for data analysis can no longer meet the needs of contemporary humans.
Complex networks have emerged in the field of data analysis because of their strong stability and robust operation. Therefore, this paper applies it to the research of EEG signal classification.
This paper completes the classification of EEG signals by extracting the complex network characteristics of EEG data from working memory experiments. The main innovations are as follows:
Firstly, a single threshold model based on neural network classifier is proposed. The time domain features were selected and the three different feature extraction methods were tested and compared after the single threshold threshold model was determined. The distribution of complex network degrees was selected as the characteristics of the sample data.
Secondly, based on the single threshold classification model, a multi-threshold method is proposed for classification. The model construction process is basically the same, but the density distribution of the distance matrix brings new enlightenment. In this paper, two methods of uniformly setting multiple threshold points and setting the threshold according to the characteristics of the density map are tried to better express the structural characteristics.
Finally, in the case of applying a heterogeneous multi-threshold joint complex network model, the classification error is reduced to 5\%. Although the model's generalization ability is insufficient, the method of extracting EEG data features by complex network method provides a new attempt for the model classification algorithm.
\vskip 0.8cm \noindent{\rm Key Words:\ cognitive load, \ EEG signal, \
network characteristics, \ multi-threshold, \ complex network}
\tableofcontents
%\mainmatter
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\newpage
\mainmatter
%\markboth{\protect
%\scriptsize{\underline{\underline{\hspace{1cm}{\bf
%东~南~大~学~毕~业~设~计~报~告}\hspace{0.5cm}{\bf 第一章 \qquad 引言}
% \hspace{1cm}}}}}
% {\protect \scriptsize{
% \underline{\underline{\hspace{1cm}{\bf 东~ 南~大~学~毕~业~设~计~报~告}\hspace{7cm}{\songti
% 第一章 \qquad 引言 }\hspace{1.5cm}}}}}
\chapter{绪论}
\s0 \vskip 3mm
%微型计算机中压缩数据的存储、空间探索中的微型机械、显微外科工具以及现代通讯都是植根于微系统技术的广大应用中的一部分.
近几年来,随着科技的发展和社会压力的猛增,人们对脑科学领域的探索欲望也随之增强。一方面,随着数据挖掘分析技术的飞速发展,人们有了更强的工具和算法对脑电数据进行探索。另一方面,人们日常工作、生活的压力越来越大,人们急需了解脑领域的状况以便对其作出措施,减少认知负荷,更加高效地生活。在这种火热而严峻的形势下,对脑电数据的探索势在必行。而以往存在的机器学习和深度学习方向的脑电分析模型已经显得不够那么优秀。更加准确和适合的模型是学者们的探索方向。而复杂网络由于其稳定而竞争性强的特点,已经在数据分析领域有所建树。本文将其应用在脑电信号分类的方面,试图使得分类结果更加准确,令人信服。
\section{文献综述}
针对人类各类活动的分析近年来开展得如火如荼,各类应用问题不断涌现。其中,对大脑的研究更是其中的重点探索方向,大脑的复杂性使得这一领域充满挑战性。大脑的综合活动通常通过磁共振和脑机接口(BCI)的方法来分析。
脑电信号(EEG)由在大脑皮层的脑神经细胞电活动反映,包含了大量认知信息、病理信号和运动感应等等,在医学方面有重大研究价值。过去的研究中,深度学习已经在许多分类分析任务上有所建树,如计算机视图,它的信号比脑电图数据更强大。卷积神经网络(CNN)和循环神经网络(RNN)已经显示出它们在处理和挖掘脑电图信号特征中的潜在应用。但是直接对 EEG 时间序列使用深度学习的方法并不能比传统机器学习方法得到更好的结果。一方面,大部分情况下脑电数据集对于深度学习模型训练而言不够大。另一方面,脑电数据集一般都有很多的通道,但只有一部分通道都与特定的认知任务有关。因此,学习脑电图的强有力特征表达十分重要。人们提出了许多种将脑电信号时间序列转换为复杂网络的方法。
剩余内容已隐藏,请支付后下载全文,论文总字数:32788字
该课题毕业论文、开题报告、外文翻译、程序设计、图纸设计等资料可联系客服协助查找;