Advancements in Communication and Systems
Comparative Analysis on the Effect of Feature Selection on Classification Performance
Authors: Sunita Beniwal, Neha Kathuria and Ashwani Kumar
Publishing Date: 13-09-2024
ISBN: 978-81-955020-7-3
Abstract
A huge amount of raw data is present in the information industry and this raw da-ta has to be converted into useful information. Feature selection is a process of selecting the useful and relevant features from the data set. Feature selection is important as it helps in reducing the size of the data and complexity of the model and makes it simpler and easily understandable. Feature selection aims to minimize the cost and improve the performance of the model. This research work is performed using two steps, firstly feature selection is performed using Bacterial Foraging optimization algorithm on two datasets i.e., Iris and Diabetic datasets and performance of selected features is evaluated using Naïve Bayes and KNN. The results of the research are compared with accuracy of original datasets with-out feature selection. The feature selection using BFO yielded better results.
Keywords
Bacterial foraging optimization, dimensionality, feature selection, KNN, naïve bayes.
Cite as
Sunita Beniwal, Neha Kathuria and Ashwani Kumar, "Comparative Analysis on the Effect of Feature Selection on Classification Performance", In: Ashish Kumar Tripathi and Vivek Shrivastava (eds), Advancements in Communication and Systems, SCRS, India, 2024, pp. 577-582. https://doi.org/10.56155/978-81-955020-7-3-51