Multivariate Statistical Analysis LDA,QDA, decision tree
- 최초 등록일
- 2012.05.08
- 최종 저작일
- 2012.04
- 13페이지/
압축파일
- 가격 1,000원
![할인쿠폰받기](/images/v4/document/ico_det_coupon.gif)
소개글
주어진 데이터를 가지고 R프로그램에서
LDA, QDA, decision tree 만들고,
이의 sensitivity, specificity, 10-fold cross validation, information gain,
sum of squared error 을 계산한 레포트 입니다.
목차
1. Consider the following two-class data.
2. Consider the dataset (Cleveland Heart Disease Dataset.xls) and its description (Cleveland Heart Disease.doc) that I upload in the portal system. Split the dataset into two parts as follows:
3. Consider the dataset (ToyotaCorolla.xls) that I upload in the portal system.
본문내용
1. Consider the following two-class data.
library(MASS) #이것을 해야 linear discriminant boundary와 quadratic discriminant boundary
(1) Find the linear discriminant boundary based on the normal distribution and draw a sketch of the decision boundary.
m<- lda(Y~ ., data = data, prior = c(1, 1)/2, CV=TRUE) # = m<-lda(Y~ ., data = data)
ct<-table(data$Y, m$class)
sum(diag(prop.table(ct)))
plot(m, dimen=1, type="both") #(1)
library(klaR)
levels(data$Y)<- c(`1`, `2`)
partimat(Y~X1+X2,data=data,method="lda") #(2)
pairs(data[c("X1","X2")], main="My Title ", pch=22,
bg=c("red", "yellow")[unclass(data$Y)]) #(3)
참고 자료
없음
압축파일 내 파일목록
Cleveland Heart Disease Dataset.xlsx
Cleveland Heart Disease Description.doc
Multivariate Statistical Analysis.hwp
ToyotaCorolla.csv