{"id":10165,"date":"2020-05-13T17:52:36","date_gmt":"2020-05-13T12:22:36","guid":{"rendered":"https:\/\/www.botreetechnologies.com\/blog\/?p=10165"},"modified":"2026-03-27T16:33:24","modified_gmt":"2026-03-27T11:03:24","slug":"decision-tree-algorithm-in-machine-learning","status":"publish","type":"post","link":"https:\/\/www.botreetechnologies.com\/blog\/decision-tree-algorithm-in-machine-learning\/","title":{"rendered":"Decision Tree Algorithm in Machine Learning"},"content":{"rendered":"\n

The application of decision tree algorithm is known by all Machine Learning developers. It is a useful supervised Machine Learning algorithm that focuses on probable outcomes and possibilities from decision nodes. <\/p>\n\n\n\n

It is the easiest and most popular algorithm for classification. As one of the best Machine Learning algorithms, decision trees can help solve both classification and regression problems.<\/p>\n\n\n\n

A decision tree is a flowchart-like tree structure where an internal node represents a feature(or attribute), the branch represents a decision rule, and each leaf node represents the outcome. The topmost decision node in a decision tree is known as the root node.<\/p>\n\n\n\n

Decision trees come under the supervised learning algorithms category. As we mentioned, it is primarily used for regression and classification in machine learning models<\/a>. It provides transparency by offering a single view of all traces and alternatives. Decision tree also assign specific values to problems and decisions, enabling better decision-making.<\/p>\n\n\n\n

Read more : <\/em><\/strong>Machine Learning: Everything you Need to Know<\/strong><\/em><\/a><\/p><\/blockquote>\n\n\n\n

But do you know how a decision tree works? Most people use it because it\u2019s easy and provides a graphical representation of the problem. In this article, we will look at the decision tree algorithm in detail.<\/p>\n\n\n\n

Types of Decision Tree Algorithms<\/h2>\n\n\n\n

There are two different types of decision tree algorithm for machine learning.<\/p>\n\n\n\n

  1. Classification trees:-<\/strong> In this type of decision tree, there is only one outcome from a set of two. The outcome could be either true for a particular dataset or false. The decision variable is categorical. In this decision tree classifier, categorical classifiers are essential.<\/li>
  2. Regression trees:- <\/strong>In this type of decision tree for machine learning algorithms, the outcome is continuous and changes based on the value of variables in the dataset. It is one of the types of decision trees that have continuous target variables. It is also known as a continuous variable decision tree.<\/li><\/ol>\n\n\n\n

    The root of the decision tree algorithm is at the top, and it flows upside down. In this Machine Learning algorithm for classification, the branches split downwards and the internal nodes are there to satisfy a particular condition.<\/p>\n\n\n\n

    \"Types
    source: wikipedia<\/figcaption><\/figure><\/div>\n\n\n\n

    The text in black are the internal nodes, which allows the tree to split into branches based on specific conditions. The end of the branch is the decision or leaf, from which no more branch can pan out. The methodology is also popularized as learning decision tree from data.<\/p>\n\n\n\n

    Read more: Python AI: <\/em><\/strong>Why Python is Better for Machine Learning and AI<\/strong><\/em><\/a><\/p><\/blockquote>\n\n\n\n

    How Does Decision Tree Work ?<\/b><\/h2>\n\n\n\n
    \"How<\/figure><\/div>\n\n\n\n

    Gini Index:<\/a><\/h3>\n\n\n\n

    Another decision tree algorithm CART (Classification and Regression Tree) uses the Gini method to create split points.<\/p>\n\n\n\n

    \"Gini<\/figure><\/div>\n\n\n\n

    Where pi is the probability that a tuple in D belongs to class Ci.<\/p>\n\n\n\n

    The Gini Index considers a binary split for each attribute. You can compute a weighted sum of the impurity of each partition.<\/p>\n\n\n\n

    Entropy:<\/h3>\n\n\n\n

    Entropy is a measure of disorder or uncertainty and the goal of machine learning models and Data Scientists, in general, is to reduce uncertainty.<\/p>\n\n\n\n

    \"entropy\"<\/figure><\/div>\n\n\n\n

    We simply subtract the entropy of Y given X from the entropy of just Y to calculate the reduction of uncertainty about Y given an additional piece of information X about Y. This is called Information Gain. The greater the reduction in this uncertainty, the more information is gained about Y from X.<\/p>\n\n\n\n

    Let\u2019s see an example to train model with diabetes data using the above algorithm<\/p>\n\n\n\n

    Please note that, We are going to use Pandas and Sklearn for training data and using existing dataset of diabetes from Kaggle<\/a>.<\/p>\n\n\n