{"id":85678,"date":"2021-04-28T15:29:16","date_gmt":"2021-04-28T19:29:16","guid":{"rendered":"https:\/\/ibkrcampus.com\/?p=85678"},"modified":"2024-05-20T15:58:17","modified_gmt":"2024-05-20T19:58:17","slug":"machine-learning-classification-algorithms-part-ii","status":"publish","type":"post","link":"https:\/\/www.interactivebrokers.com\/campus\/ibkr-quant-news\/machine-learning-classification-algorithms-part-ii\/","title":{"rendered":"Machine Learning Classification Algorithms \u2013 Part II"},"content":{"rendered":"\n

See Part I<\/a> to get insight on Supervised Learning.<\/em><\/p>\n\n\n\n

Unsupervised Learning<\/h2>\n\n\n\n

In unsupervised learning, as the name suggests, the dataset used for training does not contain the required answers. Instead, the algorithm uses techniques such as clustering<\/strong> and dimensionality reduction<\/strong> to train.<\/p>\n\n\n\n

A major application of unsupervised learning is anomaly detection. This uses a clustering algorithm to find out major outliers in a graph. These are used in credit card fraud detection.<\/p>\n\n\n\n

\n\n\n\n

Types of Supervised Models<\/h2>\n\n\n\n

Supervised models are trained on labeled dataset. It can either be a continuous label or categorical label.<\/p>\n\n\n\n

Regression models<\/h3>\n\n\n\n

Regression is used when one is dealing with continuous values such as the cost of a house when you are given features such as location, the area covered, historic prices etc. Popular regression models are:<\/p>\n\n\n\n