Decision tree in data mining is open to comprehend, however exceptional for multifaceted datasets. This marks them an extremely useful means. Let’s discuss in brief:-

  • Decision trees consist of three key portions decision nodes (representing decision), chance nodes (representing likelihood), and end nodes (representing results).
  • Decision trees can be used to compact with multifaceted datasets and can be clipped if essential to evade overfitting.
  • Notwithstanding having many advantages, decision trees are not suitable for all types of information, e.g. continuous variables or unbalanced datasets.
  1. What is the decision tree in data mining?
  2. Decision Tree Algorithm in Data Mining
  3. Important Terms of Decision Tree in Data Mining
  4. Root nodes
  5. Application of Decision Tree in Data Mining
  6. Advantages of Decision Tree
  7. Disadvantages of Decision Tree

1) What is the decision tree in data mining?

A decision tree is a plan that includes a root node, branches, and leaf nodes. Every internal node characterizes an examination on an attribute, each division characterizes the consequence of an examination, and each leaf node grasps a class tag. The primary node in the tree is the root node.

The subsequent decision tree is for the thought buy a computer that shows whether a purchaser at an enterprise is expected to buy a computer or not. Each internal node characterizes an inspection on an attribute. Each leaf node signifies a class.

2) Decision Tree Algorithm in Data Mining

Decision Tree algorithm relates to the persons of directed intelligence techniques. Unlike other-directed education procedures, the decision tree algorithm can be used to answer deterioration and arrangement difficulties.

The objective of using a Decision Tree is to craft a preparation ideal that can use to foresee the class or value of the mark variable by learning easy judgement procedures incidental from previous information (training data).

In Decision Trees, for estimating a class tag for best ever we start with the root of the tree. We make relations with the root attribute to the record’s attribute. We make division agreeing to that value and jump to the subsequent node on the base of choice.

3) Important Terms of Decision Tree in Data Mining

Decision trees can handle complicated data, which is a portion of what results in them valuable. Though, this doesn’t mean that they are difficult to know. At their centre, all decision trees finally include three vital portions or nodes.

  • Decision nodes: Represents a decision and is normally displayed with a square.
  • Chance nodes: Represents chance or confusion and is normally displayed with a circle.
  • End nodes: Represents a result and is normally displayed with a triangle.

By connecting these different nodes, we get divisions. We can use nodes and divisions an unlimited number of times to form trees of different difficulties. Let’s see how these portions appear before we include any information.

Fortunately, many decision tree vocabulary keep an eye on the tree equivalence, which marks it full calmer to recollect! Let’s explore these terminologies now:-

4) Root nodes

The blue decision is called the ‘root node’. This is at all the times the primary node in the path. It is the knot from which all other choices, forecasts and end knots finally divide.

  • Leaf nodes

In the figure above, the lavender end nodes are called the ‘leaf nodes.’ These display the conclusion of a decision route (or outcome). Every time you recognize a leaf node because it doesn’t fragment, or subdivide any more like a real leaf.

  • Internal nodes

In between the origin knots and the leaf knots, we can have any number of internal ties. These can comprise decisions and chance nodes (for ease, this image only uses chance nodes). It is really easy to identify an internal node as each internal nodes have branches of its own while also joining to the earlier node.

  • Splitting

Dividing or ‘splitting’ is said when any node divides two or more substitute nodes. These substitute nodes can also be another internal node, or they can tip to result (a leaf/ end node)

  • Pruning

Rarely decision trees can become attractively miscellaneous. In these circumstances, they can close up giving too much load to immaterial information. To sidestep this difficulty, we can eliminate definite nodes using a procedure well known as ‘pruning’. Pruning is precisely what it echoes like if the tree develops branches we don’t require, we basically cut them off.

5) Application of Decision Tree in Data Mining

Notwithstanding their disadvantages, decision trees are static an influential and prevalent means. They are usually used by information experts to bring out an analytical investigation (e.g., improve procedures policies in trades). They are to a prevalent means for machine learning and artificial intelligence, where they are used as preparation procedures for administered wisdom (i.e. classifying information based on various tests, such as ‘sure’ or ‘nope’ classifiers.)

Mostly, decision trees are used in an extensive variety of businesses, to crack numerous categories of difficulties. Because of their elasticity, they are used in areas from know-how and fitness to the fiscal formation. Illustrations comprise:

  • A know-how corporate assessing extension occasions based on examination of earlier revenue information.
  • A puppet business determining where to objective its partial marketing financial strategy, based on what demographic information guides consumers is likely to purchase.
  • Banks and loan providers using past information to forecast how likely it is that a debtor will default on their payments.

6) Advantages of Decision Tree

  1. In comparison to other procedures, decision trees need not as much energy for information training during pre-processing.
  2. A decision tree does not involve stabilization of information.
  3. A decision tree does not need scaling of information as well.
  4. Omitted values in the information also do not disturb the procedure of constructing a decision tree to any substantial degree.
  5. A Decision tree model is identical instinctive and stress-free to describe to practical squads as well as investors.

7) Disadvantages of Decision Tree

  1. A minor variation in the information can cause a huge variation in the configuration of the decision tree triggering unpredictability.
  2. For a Decision tree occasionally calculation can go far extra multifaceted in comparison to other procedures.
  3. Decision tree repeatedly takes greater time to train the model.
  4. Decision tree preparation is comparatively lavish as the difficulty and period have taken are extra.
  5. The Decision Tree procedure is insufficient for relating deterioration and forecasting uninterrupted values.


Decision Trees helps to forecast upcoming events and are easy to understand. They work more efficiently with discrete attributes. They may suffer from error propagation.

If you are interested in making a career in the Data Science domain, our 11-month in-person Postgraduate Certificate Diploma in Data Science course can help you immensely in becoming a successful Data Science professional. 



Are you ready to build your own career?