Scribd Logo
Upload

Welcome to Scribd!

  • 36 views

    Linear Discriminant Analysis (Lda)

    Uploaded by

    Rohit Ghai
    This document discusses Linear Discriminant Analysis (LDA), a linear classification technique for problems with more than two classes. LDA works by creating a new axis that maximizes the distance between class means while minimizing variation within each class. The document then demonstrates LDA on a wine dataset, building an LDA model, making predictions, and calculating accuracy.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd

    Linear Discriminant Analysis (Lda)

    Uploaded by

    Rohit Ghai
    36 views11 pages
    This document discusses Linear Discriminant Analysis (LDA), a linear classification technique for problems with more than two classes. LDA works by creating a new axis that maximizes the distance between class means while minimizing variation within each class. The document then demonstrates LDA on a wine dataset, building an LDA model, making predictions, and calculating accuracy.

    Original Title

    2.5-Project-5-LDA-wine-in-R

    Copyright

    © © All Rights Reserved

    Available Formats

    PPTX, PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This document discusses Linear Discriminant Analysis (LDA), a linear classification technique for problems with more than two classes. LDA works by creating a new axis that maximizes the distance between class means while minimizing variation within each class. The document then demonstrates LDA on a wine dataset, building an LDA model, making predictions, and calculating accuracy.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    36 views11 pages

    Linear Discriminant Analysis (Lda)

    Uploaded by

    Rohit Ghai
    This document discusses Linear Discriminant Analysis (LDA), a linear classification technique for problems with more than two classes. LDA works by creating a new axis that maximizes the distance between class means while minimizing variation within each class. The document then demonstrates LDA on a wine dataset, building an LDA model, making predictions, and calculating accuracy.

    Copyright:

    © All Rights Reserved
    Download as PPTX, PDF, TXT or read online from Scribd
    Download as pptx, pdf, or txt
    You are on page 1of 11

    Linear Discriminant Analysis

    ( LDA )
    • Logistic regression is a classification
    algorithm traditionally limited to only two-
    class classification problems.
    • If you have more than two classes then
    Linear Discriminant Analysis is the preferred
    linear classification technique.
    Linear Discriminant Analysis
    ( LDA )
    Two criteria are
    used by LDA to
    create a new axis:
    1.Maximize the
    distance
    between means
    of the two
    classes.
    2.Minimize the
    variation within
    each class.
    library(MASS)
    library(car)
    data(wine, package = 'rattle.data’)
    ?wine # to know about data
    df=wine
    str(df)
    summary(df)
    dev.new()
    scatterplotMatrix(df[2:6])
    ggplot(df,aes(df$Alcohol,df$Malic,
    col=df$Type))+
    geom_point()

    ggplot(df,aes(df$Ash,df$Alcalinity,
    col=df$Type))+
    geom_point()
    set.seed(123)
    rno=sample(nrow(df),nrow(df)*0.7)
    trn=df[rno,]
    tst=df[-rno,]
    # LDA model building

    library(caret)
    lda1=train(Type~.,data=trn,method="lda")
    Lda1
    # LDA – Prediction & Accuracy

    lda1.pred=predict(lda1,newdata = tst)

    confusionMatrix(lda1.pred,tst$Type)
    dev.new()
    par(mfrow=c(2,7))
    for(i in 2:14){
    hist(df[,i],
    main=paste('Distribution of',names(df[i]),sep='-'),
    xlab = names(df[i]))
    }
    dev.new()
    par(mfcol=c(2,7))
    for(i in 2:14){
    boxplot(df[,i]~df[,1],
    main=paste('Boxplot-',names(df[i]),sep='-'),
    xlab = names(df[1]),ylab=names(df[i]),
    col=rainbow(3))
    }

    You might also like