IPA IP 2021 Question 20
A question IPA Japan published in 令和3年度 ITパスポート試験 公開問題 問20, translated into English. ITPEC is based on the same exams.
Which of the following is an appropriate example of machine learning in an image recognition system?
Explanation
A computer that learns to extract the features of fruit from many photographs — machine learning is the computer finding the rules from data itself.
In machine learning a computer is given a large amount of data and works out the patterns in it by itself, instead of following rules a person wrote. Fed many photographs of fruit, an image recognition system learns on its own which features (shape, colour, texture) mark each fruit, and can then recognize fruit in new images. Learning features automatically from many images is how deep learning works.
Answer (b)
Why not others:
- •(a) — describes using iris authentication and its effect on security; nothing is said about the system learning from data
- •(c) — describes fingerprint unlocking as a use of image recognition, again with no learning involved
- •(d) — a help desk answering typed questions in conversation is a chatbot, natural language processing, not image recognition
Key rule: An example of machine learning shows the system learning patterns or features from large amounts of data; merely using a recognition technology (biometrics, chatbots) is not one.
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.