Scribd
Upload
9 views

Activity Sheet

Uploaded by

arfahmedina123
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
9 views

Activity Sheet

Uploaded by

arfahmedina123
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

1.

Objective: Build SVM classification model to predict if the customer is likely to accept the
personal loan offered by the bank.
2. Another library kernlab for kernel SVM
3. Grid search
Dataset Details Attribute Description
ID Customer ID
Age Customer's age in completed years
Experience #years of professional experience
Income Annual income of the customer ($000)
ZIPCode Home Address ZIP code.
Family Family size of the customer
CCAvg Avg. spending on credit cards per month
($000)
Education Education Level. 1: Undergrad; 2: Graduate; 3:
Advanced/Professional
Mortgage Value of house mortgage if any. ($000)
Personal Loan Did this customer accept the personal loan
offered in the last campaign? (Target
attribute)
Securities Account Does the customer have a securities account
with the bank?
CD Account Does the customer have a certificate of
deposit (CD) account with the bank?
Online Does the customer use internet banking
facilities?
CreditCard Does the customer use a credit card issued by
UniversalBank?

############### Classification using e1071 #############


1. Load Data into R
2. Data preparation
a. Remove the columns ID & ZIP
b. Split the data into train and test datasets
c. Variable “Education” has 3 categories, so create dummy variables
d. Standardization of data
3. Model Building

#install.packages("e1071")
library(e1071)
# Store the independent variables and target variable separately
#Build the model on train data
model = svm(x.train,y.train, method = "C-classification", kernel = "linear", cost = 10, gamma = 0.1)
summary(model)
#The "cost" parameter balances the trade-off between having a large margin and
#classifying all points correctly. It is important to choose it well to have
#good generalization.
4. Predict on train & test data
5. Build the confusion matrix
6. Compute the error metrics

Note: Build SVM model by changing the kernel function to “radial” and check if the accuracies are
better.
####### Classification using KSVM #############
#install.packages("kernlab")
library(kernlab)
#Build model using ksvm with "rbfdot" kernel
kern_rbf <- ksvm(as.matrix(train_bankdata[,-7]),train_bankdata[,7],
type='C-svc',kernel="rbfdot",kpar=list(sigma=(0:1)),
C=10, cross=5)
kern_rbf
kern_rbf <- ksvm(as.matrix(train_bankdata[,-7]),train_bankdata[,7],
type='C-svc',kernel="rbfdot",kpar="automatic",
C=10, cross=5)
#Build model using ksvm with "vanilladot" kernel
kern_vanilla <- ksvm(as.matrix(train_bankdata[,-7]),train_bankdata[,7],
type='C-svc',kernel="vanilladot", C = 10)
kern_vanilla
7. Predict on train & test data
8. Build the confusion matrix
9. Compute the error metrics

#Perform a grid search


tuneResult <- tune(svm, train.x = x, train.y = y, ranges = list(gamma = 10^(-6:-1), cost =
2^(2:3)))
print(tuneResult)
10. Predict on train & test data
11. Build the confusion matrix
12. Compute the error metrics

You might also like