IPA IP 2024 Question 65
A question IPA Japan published in 令和6年度 ITパスポート試験 公開問題 問65, translated into English. ITPEC is based on the same exams.
Which of the following is the appropriate combination of terms to be inserted into blanks A through C in the description below concerning learning methods of machine learning in AI?
Supervised learning is a learning method in which, by inputting training data with correct answers attached, it becomes possible to divide unknown data into multiple classes by a technique called A, and to find relationships in data by a technique called B. Unsupervised learning is a learning method in which, by inputting training data without correct answers attached, it gradually becomes possible to correctly divide data into groups by a technique such as the one called C.
Explanation
A: Classification, B: Regression, C: Clustering — the two tasks of supervised learning and the typical task of unsupervised learning.
- •Classification (supervised) — learns from labelled examples and assigns unknown data to one of several classes, such as spam or not spam: blank A
- •Regression (supervised) — learns the relationship between input variables and a numeric value, so it can predict values such as sales or prices: blank B
- •Clustering (unsupervised) — no correct answers are given; the algorithm groups data by similarity on its own, such as customer segments: blank C
Clustering cannot be a supervised technique here, because it needs no labelled data, and regression and classification cannot be the unsupervised one, because both learn from correct answers.
Answer (c)
Why not others:
- •(a) — swaps A and B: dividing data into classes is classification, finding relationships is regression
- •(b) — puts clustering under supervised learning and regression under unsupervised learning
- •(d) — puts clustering under supervised learning and regression under unsupervised learning, with classification in the right place
Key rule: Supervised learning (labelled data): classification predicts a class, regression predicts a value. Unsupervised learning (no labels): clustering finds the groups itself.
Basic Theory · AI-written, may contain errors
This question comes from the past questions IPA Japan publishes, translated into English by ITPEC Practice. ITPEC Practice is an independent study tool and is not affiliated with IPA or ITPEC. See the original paper (PDF) or Report an issue.