Classification using Artificial Neural Network Optimized with Bat Algorithm

In machine learning, there are two approaches: supervised and unsupervised learning. Classification is a technique which falls under the supervised learning. Out of many classification models, the most popularly used is the Artificial Neural Network. While neural networks work fine in classification...

Full description

Saved in:
Bibliographic Details
Published in:International journal of innovative technology and exploring engineering 2020-01, Vol.9 (3), p.696-700
Main Authors: Kumar, Mr. Narender, Kumar, Dharmender
Format: Article
Language:English
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In machine learning, there are two approaches: supervised and unsupervised learning. Classification is a technique which falls under the supervised learning. Out of many classification models, the most popularly used is the Artificial Neural Network. While neural networks work fine in classification and training a machine, the accuracy of the result might still be under question. To improve the accuracy and speed of result, the optimisation of artificial neural network is done. For this, ANN can be hybridised with a metaheuristic algorithm known as the Bat Algorithm. The benefits of optimising a neural network are mainly the improvement in accuracy of classification, interpretation of the data, reduction in cost and time consumption for getting accurate results etc. In present paper, a comparison between the results of an ANN-Backpropagation model and the proposed ANN-Bat model is done for medical diagnosis. The results were in the favour of the ANN-Bat approach which was significant in reducing the time taken to yield an output as well as the accuracy.
ISSN:2278-3075
2278-3075
DOI:10.35940/ijitee.C8378.019320