Microsoft Azure Machine Learning

Summary

In this last chapter, you solved another real-world problem. You started with understanding the problem and then, acquired the necessary data. After initial data exploration, you realized that the data has a large number of columns, so you used Python script modules to first split the data into two sets of features, and then used the PCA algorithm to get a reduced set of features. Then, you used the Filter Based Feature Selection module, which can identify most of the important features from the reduced dataset. To select the right model, you tried different algorithms and trained them with optimum parameters using the Sweep Parameters modules. Finally, you selected the model and proceeded to publish it as a web service.