In this article, we will introduce SQL transformations in action. We will also see how to handle missing values in our dataset.
Consider a scenario of a movie rating dataset containing records of different movies along with the average user ratings associated with each movie. The ratings are in numeric form ranging from 1 to 10 with 1 as the lowest rating and, respectively, 10 as the highest rating (though no movie in the history has achieved 10 rating:). Suppose that we want to convert the numeric ratings into categorical ratings. For instance, we want to replace ratings of 1-3 with the categorical value “poor”, of 4-6 with “average” while ratings of 7-10 will have the value “good”. We can accomplish it with SQL transformation in Azure ML Studio. Read More