인류의 복지와 편익을 위한 인프라 건설을 주도하는토목공학과
제목
(2학년) 빅데이터분석기초_기말시험_주요coding (박영훈 교수)
작성일
2022.06.10
작성자
부천대학교 토목과
#1 머신러닝_단순회귀분석 cars=read.csv("cars.csv") summary(lm(dist~speed+0, data=cars)) #2 머신러닝_다중회귀분석 data=read.csv("salary.csv") head(data, n=10) model=lm(salary~experience+score,data) summary(model) predict(model,data.frame("experience"=c(9,8),"score"=c(85,90)), interval="confidence") #3 공분산 분석 data=read.csv("anorexia.csv") head(data) str(data) data$Treat=as.factor(data$Treat) str(data) summary(data) data$Treat=relevel(data$Treat,ref="Cont") summary(data) out=lm(postwt-Prewt~Prewt+Treat,data) anova(out) install.packages("multcomp") library(multcomp) dunnett=glht(out, linfct=mcp(Treat="Dunnett")) summary(dunnett) summary(out) #4 머신러닝_판별분석 turkey=read.csv("turkey.csv") head(turkey, n=20) str(turkey) turkey=na.omit(turkey) install.packages("MASS") library(MASS) model1=lda(TYPE~HUM+ULN, data=turkey) model1 predict(model1,data.frame("HUM"=c(135,140), "ULN"=c(140,145))) model2=qda(TYPE~HUM+ULN, data=turkey) model2 predict(model2,data.frame("HUM"=c(135,140), "ULN"=c(140,145))) #5 머신러닝_신경망분석 install.packages("nnet") library(nnet) cb=read.delim("Hshopping.txt", stringsAsFactors = FALSE) head(cb,n=20) str(cb) cb$성별=as.factor(cb$성별) cb$출연자=as.factor(cb$출연자) cb$반품여부=as.factor(cb$반품여부) str(cb) install.packages("caret", dependencies = TRUE) library(caret) set.seed(1) inTrain=createDataPartition(y=cb$반품여부, p=0.6, list=FALSE) cb.train=cb[inTrain,] cb.test=cb[-inTrain,] set.seed(1234567) nn_model=nnet(반품여부~성별+구매금액+출연자, data=cb.train, size=7, maxit=1000) install.packages("NeuralNetTools") library(NeuralNetTools) garson(nn_model) predicted=as.factor(predict(nn_model, newdata=cb.test, type="class")) confusionMatrix(predicted,cb.test$반품여부)