How does machine learning differ from traditional programming?

• A. Machine learning learns patterns from data to make predictions; traditional programming encodes explicit rules and logic.
• B. Machine learning encodes explicit rules and logic; traditional programming learns patterns from data.
• C. Machine learning ignores data; traditional programming requires data.
• D. Machine learning is always deterministic; traditional programming is probabilistic.

Answer: (A)

Machine learning learns patterns from data to make predictions, while traditional programming encodes explicit rules and logic. In ML, the behavior of the system is discovered from examples—data that shows what outcomes look like for various inputs. The model learns relationships, patterns, and nuances from that data, and we then use the trained model to predict or classify new inputs. In traditional programming, a developer writes precise rules that the program follows for every situation it’s designed to handle.

A good everyday contrast is a spam filter. With ML, you feed the system a large set of emails labeled as spam or not spam, and it learns which features typically signal spam. When a new email arrives, the model estimates the probability it’s spam based on what it learned. With traditional programming, you’d hand-craft rules like “if the subject contains X and the sender is Y, mark as spam,” and the program follows those rules exactly.

Because ML relies on data, it can adapt to new patterns seen in the data and often outputs probabilities rather than fixed yes/no results. Traditional programs execute explicit logic from code and are not driven by patterns found in data in the same way. The statement that best describes the difference is the one that emphasizes learning from data to predict versus encoding explicit rules.