Scribd
Upload
4 views

Week 2 Practical(1)

The document outlines a tutorial for conducting linear regression and logistic regression experiments using housing and vertebrate datasets. It provides step-by-step instructions for downloading datasets, setting roles, splitting data, applying models, and evaluating performance metrics. Additionally, it encourages users to explore model accuracy and interpret results through confusion matrices and R2 square metrics.

Uploaded by

gcgintl.00
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
4 views

Week 2 Practical(1)

The document outlines a tutorial for conducting linear regression and logistic regression experiments using housing and vertebrate datasets. It provides step-by-step instructions for downloading datasets, setting roles, splitting data, applying models, and evaluating performance metrics. Additionally, it encourages users to explore model accuracy and interpret results through confusion matrices and R2 square metrics.

Uploaded by

gcgintl.00
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

Dr Gopal Jamnal

Week 2 - Linear Regression Tutorial, IS&ML

In week 2 practical, we will do experiment on linear regression on housing data. For the
experiment you need to download dataset from VLE and import into your local computer >data
repository. Also make sure you follow the tutorial to connect all process links carefully with
appropriate operators.

Step 1: Download housing dataset and navigate file through “My Computer” and import into your
Local Repository > Data

Step 2: You need to use “set role” operator to define “target role” as label . Check the Parameters
options for attribute name and target role
Step 3: You will be using “Split Data” operator to split dataset into two parts, .7 for training and .3 for
testing. Check parameters option and click on Edit Enumeration tab to Add Entries and click ok.

Step 4: You will be search for linear regression model in Operator search bar. Keep the default
parameters.
Step 5 : To use test dataset, you will be using “apply model” operator and connect unlabelled
dataset and model links.

Apply cross validation as well

Step 6 : for evaluation, you will be using performance operator to check model accuracy using root mean
squared error metric. You need connect link to the output result(res).
Step 7: Examine the differences in performance to interpret the prediction results. Can you improve the
prediction accuracy ?
Hint : Try to adjust model hyperparameter, change split ratio from 7:3 to 8:2.

What do you realise?


Logistic Regression

Step 1: Download veterbrate dataset from week 1 and use “map” operator for value
mapping “old values to “ new values”.”

Step 2: To perform experiment on logistic regression, you need to import dataset, set role and split
data into training and testing set. For this follow the step1,2,3 of previous tutorial(linear
regression).
Step 3: Interpret model predictions, accuracy and confusion matrix.

Does model is good at predicting nonmammals than mammals ? How many wrong
prediction model produced ? Can you tell from confusion matrix ?

Exercise: you have learnt classification and regression model, try with some other dataset and
interpret model accuracy using R2 square and confusion matrix.

You might also like