AI0-001 Machine Learning and Deep Learning Practice Question
An organization needs to classify customer emails into categories. They have labeled data for some categories but not all. Which approach should they use?
⚠ Common exam trap
CompTIA often tests the distinction between semi-supervised and unsupervised learning, trapping candidates who assume that any use of unlabeled data automatically means unsupervised learning, ignoring the critical role of the existing labeled data.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Semi-supervised learning
Semi-supervised learning (D) is the correct approach because the organization has labeled data for some categories but not all. This technique leverages a small amount of labeled data to guide the clustering or classification of a larger pool of unlabeled data, effectively combining supervised and unsupervised methods to handle partially labeled datasets.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Unsupervised clustering then labeling
Why it's wrong here
Clustering groups unlabelled emails by similarity, but the categories are predefined and some already have labels, so clusters will not align to those categories. Semi-supervised learning exploits the labelled subset to classify the rest. Clustering suits fully unlabelled discovery, not partial-label classification.
- ✗
Supervised learning for all categories
Why it's wrong here
Supervised learning requires labelled examples for every target category, yet the stem states some categories lack labels entirely. Training would therefore fail to predict those classes. Supervised learning is correct when a complete labelled dataset exists for all categories, which this scenario explicitly denies.
- ✗
Reinforcement learning
Why it's wrong here
Reinforcement learning optimises sequential decisions via reward signals, whereas email categorisation is a single-pass mapping from text to a fixed label set. It cannot exploit the existing labelled examples. Reinforcement learning would be correct for tasks such as dialogue policy or adaptive control, not static classification.
- ✓
Semi-supervised learning
Why this is correct
Semi-supervised learning combines a small labelled set with a larger unlabelled set, using the labelled examples to seed category boundaries and propagating labels through structure in the unlabelled emails. This suits partial category coverage where full supervision is unavailable.
About these practice questions
One of 962 original AI0-001 practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This AI0-001 practice question is part of Courseiva's free CompTIA certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AI0-001 exam.