Supervised machine learning definition
Supervised machine learning is a machine learning type. A machine learns from labeled training data, and makes predictions based on that data. Essentially, the ‘trainer’ explicitly provides the algorithm with the ‘right answers’ (labels) for a set of examples. The algorithm’s job is to learn the patterns or relationships so that it can make correct predictions on new, unseen data later.
Elements of supervised machine learning
Labeled Data: This is the critical characteristic of supervised learning. Labeled data means every example in the dataset is paired with the correct output. So in a dataset for recognizing cats, each image would be labeled either ‘cat’ or ‘not cat’.
Training: The machine learning model is ‘trained’ using this labeled dataset. It tries to learn the relationship between the input (e.g., features of an image) and the output (e.g., ‘cat’ or ‘not cat’).
Prediction: Once the model is trained, it can then predict the output for new, unseen data. You could show the trained model a new image it has never seen before, and it would determine whether the image is of a cat.
Evaluation: A part of the labeled dataset (which the model hasn’t seen during training) is set aside for evaluation. By comparing the model’s predictions to the actual labels, we can check its accuracy and make improvements.