WorksheetsFinal Exam
Total questions: 15
Worksheet time: 16mins
Imagine you are asked to use machine learning to build a model that can predict whether or not a customer will upgrade to a premium subscription plan for a streaming service. You are given data describing variables including customer ID, monthly subscription cost, number of hours watched per month, and payment history (e.g., whether the customer has made late payments). You have this data for many previous customers along with a variable, plan upgrade, recording whether or not the customer upgraded their plan.
Of the variables mentioned above, select all which you might use as inputs to a machine learning model.
customer ID
monthly subscription cost
number of hours watched per month
payment history
plan upgrade
In the data preparation stage for a classification task, you have created box plots for several variables. Which of the following plots show variables that are contaminated with outliers?
You are working in the analytics department of a company. One of your colleagues sent you an archive containing several datasets for an upcoming project but forgot to specify which dataset to use. In their message, they mentioned you will need to apply one-hot encoding as part of the data preprocessing.
Given this information, which dataset should you choose to proceed with?
Imagine you are tasked with building a machine learning model to predict the sale price of rice based on features such as previous year yield, number of rainy day in the current year, total area cultivated and fertilizer usage. After training your regression model, you need to evaluate its performance.
From the following list of evaluation metrics, which are appropriate for this regression task?
Mean Absolute Error (MAE)
Accuracy
Silhouette Score
F1 Score
R-squared (R²)
When working on a machine learning project, we need to ensure the following steps: first, (a) which involves defining the project goals and objectives; second, (b) where we explore and analyze the data; third, (c) which focuses on cleaning and preprocessing the data; fourth, (d) , where the machine learning model is trained on the prepared data; and finally, (e) to assess the performance of the trained model.
Which step in the machine learning pipeline involves searching for the optimal combination of hyperparameters, such as learning rate and regularization strength, to maximize validation performance? (a)
Many algorithms use distance metrics to calculate instances that are similar to others. Distance X is using the (a) metric, and Distance Y is using the (b) metric.
The Manhattan distance from point (0,0) to point A is (a) and from point A to point B is (b) .
You are working with a dataset of different types of fruits, with features such as weight, color, shape, and size. Your goal is to group the fruits into distinct categories based on these attributes.
Given that the algorithm you are required to use is K-means, this means we are dealing with a (a) task.
You are hired by a used bookstore to develop a machine learning model that can automatically classify incoming stock as being in "acceptable" or "unacceptable" condition based on various features, such as cover damage, page quality, and spine condition. After training and testing your model, you present the results to the bookstore owner and explain that your model made (a) wrong predictions in total.
You are given a dataset, the first few rows of which are shown in the image. You have been asked to build a multilayer perceptron classifier for the data. Which variable should be recoded into groups to reduce the number of different values it can take, before being represented using one-hot encoding
Customer Income
Customer Age
Customer Job
Customer ID
Within the Orange workflow you were given, what is the Mode value for the TAX variable? Write it as a number with no decimal points!
(a)
Within the Orange workflow you were given, the pair of variables with the strongest positive linear correlation is (a) while the pair of variables with the strongest negative linear correlation is (b)
Within the Orange workflow you were given, what regularization technique is used within the Linear Regression model?
No regularization
Ridge regression
Lasso regression
Elastic net regression
Within the Orange workflow you were given, if we choose mean absolute error as our evaluation metric the best model is (a) . If we choose to evaluate our models based on the mean squared error then the best one would be (b)
