At a company study session, the difference between machine learning and traditional programming came up as a topic. Compared with traditional procedural programs in which people write out all the processing, which description BEST represents machine learning (ML)?
A question about choosing the definition of machine learning.
A method that learns patterns from data to predict and decide
Correct. Machine learning is a method that learns patterns from data so it can predict and decide without people writing explicit rules. This is the difference from traditional rule-based programs.
A method in which people write in all the decision rules by hand
Writing all the rules by hand is a traditional rule-based program, which differs from machine learning.
The description is wrong, so this is incorrect.
A method that runs only on predefined knowledge without using data
The essence of machine learning is learning patterns from data, and learning cannot happen without data.
The description is wrong, so this is incorrect.
A technology that merely executes predefined steps at high speed
High-speed execution of predefined steps is a description of traditional automation and programming.
It is not a description of machine learning, which learns from data to improve its decisions, so this is incorrect.
Machine learning (ML) is a method that learns patterns from data and performs prediction and decision-making without writing explicit rules. It differs from traditional rule-based programs in which people write out all rules. "Write all rules by hand", "not using data", and "a technology that increases communication speed" are all wrong as descriptions of machine learning.