Decision Tree Using Artificial Neural Network: A Proposed Model
A decision tree is one of the data mining techniques; it has a tree structure, which consists of internal node, branches, and leaf nodes which are known as decision nodes also. Decision tree works in a transparent way [1 ]. Unlike artificial neural networ
- PDF / 1,135,339 Bytes
- 14 Pages / 439.37 x 666.142 pts Page_size
- 95 Downloads / 360 Views
Decision Tree Using Artificial Neural Network: A Proposed Model Monika Rathore and Shruti Gupta
1 Introduction Artificial neural networks and symbolic learning algorithms are the two approaches through which we can go for machine learning. Decision tree proves to be a quite efficient approach, but when the large amount of data is encountered, [2] it leads to poor statistical efficiency. Artificial neural network always produces the accurate result, but they explicitly reveal the reasoning behind their decisions [1]. This paper proposes a model in which we combine the explicit explanation of the decision tree along with the predictive accuracy offered by artificial neural network.
1.1 Decision Tree A decision tree is a decision support tool where each node represents a “test” on a particular attribute. Each branch of decision tree represents the result of the analysis, and each leaf node of the tree signifies the decision taken after computing all attributes, that is also known as class label [2]. There are three types of nodes in a decision tree: 1. Decision nodes, be a symbol of squares. 2. Chance nodes are represented by circles. 3. End nodes are symbolized by triangles (Fig. 1). M. Rathore (B) · S. Gupta International School of Informatics and Management, Jaipur, Rajasthan, India e-mail: [email protected] S. Gupta e-mail: [email protected] © Springer Nature Singapore Pte Ltd. 2021 S. D. Purohit et al. (eds.), Proceedings of International Conference on Communication and Computational Technologies, Algorithms for Intelligent Systems, https://doi.org/10.1007/978-981-15-5077-5_18
197
198
M. Rathore and S. Gupta
Fig. 1 Decision tree [1]
Decision Rules Decision rules are the if-then statements that are used to predict the results. A decision rule comprises of single or a combination of multiple rules. On the basis of the conditions, the outcome is generated by the decision tree [3]. i f < condition1 > and < condition2 > and < condition N > then outcome.
1.2 Artificial Neural Network An artificial neural network is a set of associated units called artificial neurons, similarly the neurons in a biological brain. Each association of these neurons can send out a signal from one artificial neuron to another neuron. An artificial neuron first receives a signal (Fig. 2). After processing it, and send out this to other neurons. They have distributed representation in their hidden layers which is hard to understand. There are two types of topologies used in artificial neural network [4]: • Feedforward: In this type of topology, the flow of the information is unidirectional. There is no system of feedback. • Feedback: Feedback loops are allowed in the feedback topology of ANN [5].
18 Decision Tree Using Artificial Neural Network: A Proposed Model
199
Fig. 2 Artificial neural network [4]
2 Construction of Decision Tree 2.1 Experiment Setup RStudio 3.4.4 version is used for the decision tree construction. This is done with the help of R programming language.
2.2 Dataset Used Cardiotocography dataset of
Data Loading...
