Tutorials

My previous post introduced a machine-learning model that used logistic regression to predict whether text input to it expresses positive or negative sentiment. We used the probability that the text expresses positive sentiment as a sentiment score, and saw that expressions such as “The long lines and poor customer service really turned me off” score…

One of the more novel yet practical uses for binary classification is sentiment analysis, which examines a piece of text such as a product review, a tweet, or a comment left on a Web site and scores it on a scale of 0.0 to 1.0, where 0.0 represents very negative sentiment and 1.0 represents very…

The machine-learning model featured in my previous post was a regression model that predicted taxi fares based on distance traveled, the day of the week, and the time of day. Now it’s time to tackle classification models, which predict categorical outcomes such as what type of flower a set of measurements represent or whether a…

When you build a machine-learning model, the first and most important decision you make is what learning algorithm to use to fit the model to the training data. In my previous post, I introduced some of the most widely used learning algorithms for regression models: linear regression, decision trees, random forests, gradient-boosting machines (GBMs), and…

Most machine-learning models fall into one of two categories. Supervised-learning models make predictions. For example, they predict whether a credit-card transaction is fraudulent or a flight will arrive on time. Unsupervised-learning models don’t make predictions; they provide insights into existing data. The previous post in this series introduced unsupervised learning and used a popular algorithm…