Below are some basics on machine learning algorithms. This is for beginners who are trying to get a better understanding about machine learning and getting a simple idea of what machine learning algorithms are being used.
Supervised learning uses a training dataset to make predictions. The training dataset includes input data and response values. The goal of the supervised learning algorithm is to build predictions of the response values for a new dataset. Test dataset are used to validate your model. the larger the training datasets the better you are able to predict for new datasets.
Supervised learning includes two categories of algorithms:
Classification – classification is used for categorical values, such as when each value of your data can be put into a specific class. Some examples of classification machine learning algorithms are: Support vector machines (SVM), Neural networks, Naïve Bayes classifier, Decision trees, Discriminant analysis and Nearest neighbors (kNN).
Regression – regression is used for continuous-response values. Some examples of regression machine learning algorithms are: Linear regression, Nonlinear regression, Generalized linear models, Decision trees and Neural networks.
Unsupervised Learning Algorithms do not have a known result and are used to help visualize and help deduce structures present in the data.
Examples are clustering, dimensionality reduction and association rule learning and machine learning algorithms include: the Apriori algorithm and k-Means.
Semi-supervised Learning Algorithms are a mixture of the above examples.
Semi-supervised is perfect for problems when you must learn the structure of your data in order to make predictions.
Example problems are classification and regression and machine learning algorithms are extensions to other flexible methods to make assumptions about how to model the unlabeled data.
Ultimately when trying to make decisions you will likely use supervised and unsupervised learning and the associated algorithms.