Machine Learning for Absolute Beginners: A Plain English Introduction (First Edition)

your company may wish to examine a subset of customers that purchase at the same time of year and discern what factors influence their purchasing behavior.

A major advantage of unsupervised learning is that it enables you to discover patterns in the data that you weren’t aware existed—such as the presence of two genders.

Logistic regression is typically used for binary classification to predict two discrete classes, i.e. pregnant or not pregnant.

One variable is the result you wish to predict, known as the dependent variable (y).

Unsupervised learning algorithms include k-means clustering, association analysis, social network analysis, and descending dimension algorithms.

Given its strength in binary classification, logistic regression is commonly used in fraud detection, disease diagnosis, emergency detection, loan default

algorithms, such as regression analysis, you can attempt to create a new model to integrate these two independent variables.

As a supervised learning technique, clustering can be utilized to classify new data points into existing clusters through k-nearest neighbors (k-NN).

The model is a rule to predict y based on selected independent variables (X) from the dataset.