College of Software Engineering
Undergraduate Cour Syllabus
Cour ID 311021020 Cour Name Pattern Recognition
Cour
鱼料Attribute
□Compulsory ■Selective Cour Language□English ■Chine Credit Hour 2 Period32
Semester□First Fall □First Spring □Second Fall □Second Spring
□Third Fall ■Third Spring □Fourth Fall □Fourth Spring Instructors He Kun
Description This cour will mainly introduce the following knowledge to the students: （1）Bayes formula Decision Theory。
（2）Probability density function estimation。
（3）linear difference function。
（4）nonlinear difference function。
（5）neighbor method.。
（6）empirical risk minimization and orderly risk minimization method。
（7）Characteristics choo and extraction。
（8）K-L expansion bad Feature Extraction。
小学生小实验（9）unsupervid studying method。
（10）Artificial Neural Network。
（11）Fuzzy Pattern Recognition method。
（12）statistical learning theory.Support vector machine。
Prerequisites
Calculus, Probability Statistics, Linear Algebra, Discrete Mathematics, C Language Programming Textbook
《Pattern Recognition》，Biao Zaoqi，Advanced education Press，2003，Second Edition.
是否原文教材？
Resource 《Pattern Recognition》，Huang Fenggang，Advanced education Press，1992
《Pattern Recognition》，Written by Caiyuanlong，Xi An Telecom engineering Press
《Pattern Recognition and Condition Monitoring》（first edition），Wen Xishen，Chansha：National University of Defence Technology Press，1997 .11
《Fuzzy Information Processing and application》（third edition），Chao Xiedong，Beijing：Science Press，2004.08 《Introduction to 》（fourth edition），Sheng qing，Changsha：National University of Defence Technology Press，1999.04
Grading
assignments, class participation, & term project (40%), final exam (60%)
Topics
Chapter1 Introduction
1. Object and Requirements
a．Know some basic concept of Pattern Recognition
2. Teaching content
(1) Concept of pattern recognition and pattern
(2) pattern recognition system
(3)Problems related to pattern recognition
Chapter2 Bayes Decision Theory
1. Object and Requirements
a．Master some decision rules
b. Master the statistical decision of normal distribution
c. Know the design of quential classification and Classifier
2. Teaching content
(1) Some general decision rules
a．Main content
Minimum Error Ratio bad bayes decision theory，Minimum risk bad Bayes Decision，min-max decision ,design of the classifier
b．Basic concept and knowledge points
Minimum Error Ratio bad bayes decision theory，Minimum risk bad Bayes Decision，min-max decision ,design of the classifier
c．Applications（capability requirements）
Understand the general decision rules
(2)statistical decision of normal distribution
a．Main content
Definition and property of normal distribution function, multivariate normalized probability type minimum error ratio bayes discriminate function and decision interface
b．Basic concept and knowledge points
Definition and property of normal distribution function,
bayes discriminate function and decision interface
c．Applications（capability requirements）
Understand the general decision rules
(3) Problems related to classifier (lected)
a．Main content
fault ratio problems of classifier：Theoretical Calculation of fault ratio in particular condition b．Basic concept and knowledge points
fault ratio problems of classifier
c．Applications（capability requirements）
Understand the general decision rules
Chapter3 Probability density function estimation
1. Object and Requirements
a．Understand parameter estimation, supervid estimation and unsupervid estimation
2. Teaching content
(1) Concept of parameter estimation
a．Main content
Maximum Likelihood Estimation，Bayes estimation and Bayes studying
b．Basic concept and knowledge points
parameter estimation
c．Applications（capability requirements）
Understand the general decision rules
(2)supervid Parameter Estimation of normal distribution
a．Main content
Maximum Likelihood Estimation，Bayes estimation and Bayes studying
b．Basic concept and knowledge points
Maximum Likelihood Estimation
c．Applications（capability requirements）
Understand supervid parameter estimation
(3) un supervid parameter estimation (lected)
a．Main content
Un supervid maximum likelihood estimation and some problems with it
b．Basic concept and knowledge points
Un supervid maximum likelihood estimation
c．Applications（capability requirements）
Understand un supervid parameter estimation of normal distribution
Chapter 4 Linear Discriminate Function
1. Object and Requirements
a．Understand Fisher linear discriminate perceptual rule function
b. Understand minimum fault ratio sample Criterion s, minimum square error Criterion Function；
c. Understand random fault ratio linear discriminate function and multi-class problem
2. Teaching content
(1) Introduction
a．Main content
Concept of linear discriminate, linear discriminate function, the steps of designing linear classifier蚕蛹的营养价值与功效
b．Basic concept and knowledge points
estimation of the fault ratio of classifier
c．Applications（capability requirements）
Understand the estimation of the fault ratio of classifier
(2)Fisher linear discriminant
a．Main content
Fisher linear discriminant
沧海一栗
b．Basic concept and knowledge points
Fisher linear discriminantc．Applications（capability requirements）
Understand Fisher linear discriminant
(3) perceptual rule function
a．Main content
Gradient Descent Algorithm and perceptual rule function
b．Basic concept and knowledge points
Gradient Descent Algorithm; perceptual rule function
开门红祝福语c．Applications（capability requirements）
Understand Gradient Descent Algorithm and perceptual rule function
(4) minimum fault ratio sample Criterion s
二年级古诗二首
a．Main content
conjugate gradient method for sloving Linear Inequalities，arching method for solving linear inequalities.
b．Basic concept and knowledge points
conjugate gradient method
c．Applications（capability requirements）
Master conjugate gradient method
(5) minimum square error Criterion Function
a．Main content
random minimum fault ratio linear criterion function , Gradient Descent Algorithm of MSE criterion function, square error Criterion Function and its Pudo-Invers solution.
b．Basic concept and knowledge points
minimum square error Criterion Function
c．Applications（capability requirements）
Understand minimum square error Criterion Function
(6) class problem
a．Main content
concept of multi-class, introduction of decision tree
b．Basic concept and knowledge points
concept of multi-class, introduction of decision tree
c．Applications（capability requirements）
Understand concept of multi-class, introduction of decision tree
Chapter 5 Un linear Discriminate Function
1. Object and Requirements
a．Know concept of piecewi linear ntential function
b. Understand the union of concave function which reprents the piecewi linear ntential function
c. Master the piecewi linear classifier in overlapping region
d. Know quadratic discrimination function
2. Teaching content
(1)concept of piecewi linear ntential function
a．Main content
concept of piecewi linear ntential function, distance-bad piecewi linear ntential function, considerations when designing piecewi linear classifier
b．Basic concept and knowledge points
piecewi linear ntential function, piecewi linear classifier
c．Applications（capability requirements）
Understand piecewi linear ntential function and piecewi linear classifier
(2)Understand the union of concave function which reprents the piecewi linear ntential function
a．Main content
The meaning of piecewi linear ntential function
b．Basic concept and knowledge points
the union of concave function reprents the piecewi linear ntential function
c．Applications（capability requirements）
Understand the union of concave function which reprents the piecewi linear ntential function
(3) Design piecewi linear classifier in overlapping region
a．Main content
Basic idea of the algorithm is as follows: partial training method, decision rule, quadratic discrimination function
b．Basic concept and knowledge points
小时候的电视剧prototype,overlapping region, partial training method, decision rule, quadratic discrimination function c．Applications（capability requirements）
阴阳八卦Understand the basic idea of the algorithm
3. Practice
4. teaching method
Multimedia Demonstration in class，group discussion
Chapter 6 Neighbor method
