Scribd
Upload
17 views

CT2 Assignment

Uploaded by

Punitha Prathi
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
17 views

CT2 Assignment

Uploaded by

Punitha Prathi
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/ 3

Register No

SRM INSTITUTE OF SCIENCE AND TECHNOLOGY


21CSC555J – MACHINE LEARNING ALGORITHMS
Assignment
Theory
Program Offered: M. Tech Year / Semester:I/ I
Max. Marks: 40 Date of Submission: 22/12/2024

PART A (2 x 20 =40 Marks, Answer any Two)


(1)You are managing an online retail business with three delivery
methods: Standard Courier (low cost, slower), Expedited Courier
(higher cost, faster), and Local Delivery (moderate cost, fast). The
decision criteria include order size (small/large), urgency
(standard/expedited), and customer location (local/national). Using a
decision tree, describe how you would select the most suitable
delivery method based on these factors. For a large, expedited order
located outside the city, trace the decision path and justify the choice.
Finally, discuss the advantages and limitations of using a decision tree
in this scenario and suggest any alternative decision-making
techniques if applicable.

(2)Construct a decision tree to help us decide whether we should play


volleyball based on the weather conditions.
(3)You are tasked with predicting customer churn for a telecom
company using machine learning. The company has a dataset with
various features like usage patterns, payment history, and customer
service interactions.
(a) Briefly explain how Bagging, Random Forest, and Boosting work,
and highlight their key differences. (10 marks)
(b) Given a noisy dataset with a mix of categorical and continuous
features, which technique would you choose for predicting churn,
and why? (3 marks)
(c) How does Random Forest reduce overfitting compared to a single
decision tree? (3 marks)
(d) What are the challenges or limitations of using Bagging, Random
Forest, and Boosting in this scenario? (4 marks)

(4) Use Naïve Bayes Classifier for the following dataset to predict
whether we can pet or not?
(5) You need to classify customers into segments like "Frequent Buyers"
and "Occasional Buyers" based on their purchasing behavior using the K-
Nearest Neighbors (KNN) algorithm.

(a) Briefly explain how KNN works and the role of "K" in the algorithm. (10
marks)

(b) How would you prepare a dataset with both numerical and categorical
features for use with KNN? (3 marks)

(c) If K=3, how would a new customer's classification be determined? (3


marks)

(d) What are the challenges of using KNN for this task, and how could they
be addressed? (4 marks)

You might also like