We can only succeed in capturing insights from data if we first know how to measure the effectiveness of models. In this course, you will learn appropriate measures that are used to evaluate predictive models in a variety of contexts. You will also learn about procedures that are used to ensure that models do not cheat through, for example, overfitting or predicting incorrect distributions. A predictive exercise is not finished when a model is built. It is important to construct a vast number of models, not only to find the best one, but to ensure that they point in the same direction. This course will equip you with essential skills and knowledge for understanding performance evaluation metrics, to determine whether a model is performing adequately or not. You will also discover the ways that different model evaluation criteria illustrate how one model excels over another, and how to identify when to use certain criteria over others. This is the foundation to performing successful predictive analysis.It allows business owners to create robust models that are capable of only using the most relevant data, for example: customer purchasing behaviour, past sales, and so on. Many of the techniques you'll learn about in the course are used within powerful applications such as recommender systems.The concepts will be brought together in a comprehensive case study that deals with a churn case. You will be tasked with selecting suitable variables to predict whether a customer will leave a telecommunications provider or not, by looking into their calling behaviour, creating various models and benchmarking them by using the appropriate evaluation criteria.
Add a review