Читать книгу Machine Learning for Healthcare Applications - Группа авторов - Страница 24

Supervised method uses labeled data. In this case the models are trained to use these data. The sole objective of the supervised ML method is to train the model in a manner such that it can predict the outcome when it is provided with some new set of data. This method can be used in particular case where both inputs and the corresponding outputs are known. The important feature of this method is that it provides the most accurate results. We can categorize supervised ML into regression problem and classification problem. This method is not considered to be close to true Artificial intelligence because the model is first trained for each available data, and then it predicts the correct outcome. Supervised ML includes various algorithms i.e., Linear Regression, Support Vector Machine, Multi-class Classification, Decision tree, Bayesian Logic, etc.