Classification is a type of supervised machine learning that predicts categorical output variables. It involves using labeled examples to train a model to classify new examples. Some key points:
- Classification predicts categories like good/bad, cat/dog rather than continuous values
- It requires labeled training data that maps features to categories
- Features can be any data type like numbers, text, categories
- Models are trained to learn patterns in features to classify new unlabeled examples
- Common applications include image recognition, sentiment analysis, medical diagnosis