Regression is another main instance of supervised learning in machine learning. Given a training set of data containing observations and their associated continuous output values, the goal of regression is to explore the relationships between the observations (also called features) and the targets, and to output a continuous value based on the input features of an unknown sample, which is depicted in the following diagram:

The major difference between regression and classification is that the output values in regression are continuous while they are discrete in classification. This leads to different application areas for these two supervised learning methods. Classification is basically used in determining the desired memberships or characteristics as we've seen in previous chapters, such as email being spam or not, newsgroup topics, ad click-through or not. On the other hand, regression mainly involves estimating an outcome or forecasting a response. An example of estimating continuous targets with linear regression is depicted as follows:

Typical machine learning regression problems include the following:

• Predicting house prices based on location, square footage, number of bedrooms, and bathrooms
• Estimating power consumption based on information of a system's processes and memory
• Forecasting retail inventory
• And of course, predicting stock prices