Mila wishes to build a machine learning model to classify PET images as either containing cancerous tumours, or not. Mila chooses to use decision trees as her base model, and performs the following steps: · First, she splits her data into training/test sets. · Next, Mila creates 100 subsets of her training data by resampling images from the training data, with replacement, until each dataset contains 200 images. · Then, for each subset, she chooses to only look at a random sample of 40 pixels (same pixels for images in the same subset) · Then, she trains 100 models (one for each subset) · Finally, Mila runs each of the 100 models through the test set, and uses as her final prediction for each test image the majority vote of the 100 models. Mila’s experiment is an example of the ensemble method known as:

Question

Mila wishes to build a machine learning model to classify PET images as either containing cancerous tumours, or not. Mila chooses to use decision trees as her base model, and performs the following steps: ·  First, she splits her data into training/test sets. ·  Next, Mila creates 100 subsets of her training data by resampling images from the training data, with replacement, until each dataset contains 200 images. ·  Then, for each subset, she chooses to only look at a random sample of 40 pixels (same pixels for images in the same subset) ·  Then, she trains 100 models (one for each subset) ·  Finally, Mila runs each of the 100 models through the test set, and uses as her final prediction for each test image the majority vote of the 100 models. Mila’s experiment is an example of the ensemble method known as:

BlackTom AI · Accepted Answer

To analyze the ensemble method described, I will evaluate each option in turn and map it to the steps Mila performed.
Option a: Bagging. Bagging ( bootstrap aggregating ) involves training multiple base models on bootstrapped samples and aggregating their predictions, typically by voting or averaging. While Mila does use bootstrapped samples and majority voting, bagging on its own does not include the additional randomness introduced by selecting a random subset of features for each base model. Therefore, while similar in spirit, this option is incomplete as the overall approach because it lacks the explicit feature subsampling step.
Option b: Boosting. Boosting creates models sequentially, where each new model attempts to correct the errors of the previous ones, and the final prediction is a weighted combination of all models. Mila’s process does not fit boosting: there is no sequential weighting toward errors, nor a focus on correcting mistakes from prior models. Each of the 100 models is trained independently on a bootstrap sample, not in a boosting cascade.
Option c: Bootstrapstrapping. The term bootstrapstrapping refers broadly to the resampling-with-replacement technique used to create bootstrap samples. While Mila uses bootstrap samples, bootstrapstrapping alone does not define an ensemble method or explain the full procedure of training multiple models with random feature subsets and combining their predictions. It describes only part of the data generation step, not the ensemble mechanism itself.
Option d: Random Forest. This option precisely matches the described workflow: an ensemble of decision trees trained on bootstrap samples (each subset created by resampling with replacement) and, crucially, with random feature subspace selection (she looks at a random sample of 40 pixels for each subset). The final prediction is obtained by majority voting across all trees. This combination of bootstrapping, random feature selection, and aggregation via voting is the hallmark of a Random Forest, distinguishing it from plain bagging or boosting.
In summary, the key characteristics that align with Mila’s procedure are bootstrapped training sets, a random subset of features per model, independent training of many trees, and majority-vote aggregation of their predictions; these collectively identify the ensemble method as Random Forest.

View Explanation

Log in for full answers

Similar Questions

Which of the following classification models provides out-of-bag (OOB) errors for model evaluation?

In a consumer society, many adults channel creativity into buying things

Economic stress and unpredictable times have resulted in a booming industry for self-help products

People born without creativity never can develop it

More Practical Tools for Students Powered by AI Study Helper

Homework AI Solver

Stylized AI Paper Writer

Plagiarism Checker Assistant

Citation AI Academic Writing Tool

In-Class Translation Assistant

AI Note Generator

AI Quiz Answers

Past Exam Questions from University Test Bank

Smart Practice Assistant

Adaptive Practice

Making Your Study Simpler