March 2019

Other classification methods

Classification methods such as decision trees, random forests, and support vector machines are powerful methods used by industry professionals to solve a wide range of real-world problems. For example, Microsoft's Kinect gaming platform uses decision trees under the hood for real-time human pose recognition.
February 2019

Model selection

Because most data have a large number of features, it is nearly impossible to manually select features that are statistically significant in a prediction problem. "Model selection and regularization" are efficient alternatives to Ordinary Least Squares fitting that automatically select features to find the best-fitting linear model.
February 2019

Resampling methods

In most scenarios, we do not have full or enough data available to test our models for accuracy and error. Resampling methods enable us to estimate these measures so that we can be more confident about our models.
February 2019

Classification

In most situations, we want to predict whether something is or isn't. Male or female, student or employee. Classification methods are responsible for classifying a given data between such groups.
February 2019

Linear regression

Simple and widely used, linear regression is a supervised learning algorithm that serves as a "gateway" method in most analyses. To understand the rest of statistical or machine learning, a good grasp of linear regression is absolutely necessary.