Microsoft Azure Machine Learning: Algorithm Cheat Sheet ANOMALY DETECTION One-class SVM >100 features, aggressive boundary PCA-based anomaly detection CLUSTERING MULTI-CLASS CLASSIFICATION K-means Fast training, linear model Multiclass logistic regression Accuracy, long training times Multiclass neural network Accuracy, fast training Multiclass decision forest Accuracy, small memory footprint Multiclass decision jungle Fast training Finding unusual data points Discovering structure Three or more REGRESSION Ordinal regression Data in rank ordered categories Poisson regression Predicting event counts Fast forest quantile regression Predicting a distribution Linear regression Fast training, linear model START Predicting categories Depends on the two-class classifier, see notes below Predicting values TWO-CLASS CLASSIFICATION Linear model, small data sets Two-class averaged perceptron Neural network regression Accuracy, long training time Two-class logistic regression Decision forest regression Boosted decision tree regression © 2015 Microsoft Corporation. All rights reserved. Accuracy, fast training Accuracy, fast training, large memory footprint Created by the Azure Machine Learning Team One-v-all multiclass Two Two-class SVM Bayesian linear regression This cheat sheet helps you choose the best Azure Machine Learning Studio algorithm for your predictive analytics solution. Your decision is driven by both the nature of your data and the question you’re trying to answer. Two-class Bayes point machine Email: AzurePoster@microsoft.com >100 features, linear model Fast training, linear model Fast training, linear model Fast training, linear model Accuracy, fast training Two-class decision forest Accuracy, fast training, large memory footprint Two-class boosted decision tree Accuracy, small memory footprint Two-class decision jungle >100 features Two-class locally deep SVM Accuracy, long training times Two-class neural net