What is machine learning in simple terms?

It is a method for estimating patterns from examples so software can predict, rank, classify or detect new cases under a defined task.

What is the difference between supervised and unsupervised learning?

Supervised learning uses labels for a target outcome; unsupervised learning explores structure such as clusters or anomalies without that target.

Why do beginners need a baseline?

A baseline shows whether machine learning and added complexity improve meaningfully over a simple rule or historical method.

Overfitting occurs when a model learns training-sample quirks and performs worse on new representative data.

Why can accuracy be misleading?

A model can achieve high accuracy by predicting a common class while failing to find rare, costly events. Metrics must reflect consequences.

What is data leakage?

Leakage is using information during training or evaluation that would not legitimately be available when the real prediction is made.

Which algorithm should a beginner learn first?

Begin with linear or tree-based models and a reproducible evaluation pipeline. Understanding data and errors matters more than starting with a complex neural network.

Traditional Chinese · Rusaka

了解機器學習的基礎：初學者指南

機器學習是人工智慧的一個子集，專注於使電腦能夠從數據中學習並在無需明確編程的情況下做出決策。本初學者指南旨在揭開機器學習基礎知識的神秘面紗，並為對該領域感興趣的人提供堅實的基礎。