In this talk we are going to get deep into real world scenarios modeling using Azure Machine Learning. We will use business cases as reference, explaining how to create predictive models that collect useful information (insights) to help taking corporate decisions. We will also see how these models can be easily improved to get better results.
The first case will be building from scratch a predictive model that classifies the risk of granting loans in a bank. From this point we will se how to improve the accuracy of the initial model, comparing different training algorithms and adding custom modules.
The second is a social media case: Twitter Sentiment Analysis. What is it, how it works, what challenges do we have to create our own API, how to consume it, and what benefits will have for my company.