Exercise Before Final: Intro to ML
Part 1 Introduction to Machine Learning
I) Matching questions
Q1: Matching the following terms with their characteristic
Characteristic
A. Subset of AI that gives computers the ability to learn without being explicitly programmed
B. Learns from labeled data to predict discrete class labels or continuous numeric values
C. Models the underlying or hidden structure in the data
D. Any technique which enables computers to mimic human behavior
E. Involves both labeled and unlabeled data
F. Learns through reward feedback to improve performance
Term
Machine Learning <-
Supervised Learning <-
Unsupervised Learning <-
Reinforcement Learning <-
I) Multiple answer questions
Q2: Which of the following are categories of Supervised Learning? (Choose all that apply)
Classification <-
Regression <-
Clustering <-
Association <-
Dimensionality Reduction <-
Anomaly Detection <-
Q3: Which of the following are true about Regression in Supervised Learning? (Choose all that apply)
It predicts a discrete class label output <-
Its goal is to assign input examples to one of a finite set of categories or classes <-
It predicts a continuous numeric value output <-
Its goal is to map input variables to a continuous output variable <-
It is used for clustering data points <-
It is used for discovering association rules between products <-
Q4: Which of the following problems can be solved using Unsupervised Learning? (Choose all that apply)
Predicting a customer’s likelihood to churn<-
Grouping customers based on purchasing behavior <-
Estimating a house’s price based on its features<-
Discovering association rules between products <-
Classifying emails as spam or not spam <-
Reducing the dimensionality of a dataset <-
III) Multiple choice questions
Q5: You work for a wildlife conservation organization and have collected a vast dataset of images captured by camera traps in various forests. Your goal is to develop a system that can classify these images into two categories: “Contains a Tiger” or “Doesn’t Contain a Tiger.” Which type of machine learning task is this?
Answer:
Q6: You are developing a self-driving car, and you want to train it to navigate through complex urban environments while obeying traffic rules, avoiding obstacles, and reaching a destination efficiently. Which machine learning paradigm is most suitable for teaching the car to make decisions in real-time?
Answer:
Q8: You work for a market research firm, and you’re tasked with analyzing customer reviews from various e-commerce websites. Your objective is to group similar reviews together based on the topics or themes mentioned by customers. What type of machine learning task corresponds to this analysis?
Answer:
Q8: You’re employed by an e-commerce giant, and you’ve been given access to a massive dataset containing the browsing and purchasing history of millions of users. Your task is to uncover patterns like “Users who view product X also tend to buy product Y.” What machine learning approach should you apply here?
Answer:
Q9: You’re tasked with processing and analyzing a large corpus of text documents for a research project. To make the analysis more manageable, you need to reduce the dimensionality of the text data while preserving the essential information. Which machine learning technique would you employ for this task?
Answer:
Q10: In a factory setting, you want to optimize the operations of robotic arms used in assembly lines. The robots need to learn how to grasp and manipulate objects efficiently while minimizing errors. Which machine learning paradigm should you apply to enable the robots to learn and improve their actions over time?
Answer:
Q11: You’re working on creating a chatbot for customer support, and you want the chatbot to learn how to respond to customer inquiries and improve its responses based on user feedback. What type of machine learning is appropriate for training the chatbot to provide better customer assistance over time?
Answer:
Q12: As an economist, you’ve gathered historical data on housing prices in a city over the past 30 years, including factors like square footage, number of bedrooms, and proximity to the city center. Your objective is to create a model that predicts the future price of houses in this city. What type of machine learning task is this?
Answer:
Q13: You are employed by a large online retailer, and you want to analyze the shopping habits of customers. You aim to discover patterns such as “Customers who buy product A are also likely to buy product B.” Which type of machine learning task is appropriate for this scenario?
Answer:
Q14: You are a part of a customer retention team at a telecom company, and you want to predict which customers are likely to cancel their subscriptions. What type of machine learning task is appropriate for identifying potential churners among your customer base?
Answer: