Abstract
The technological advancement can help the entire application field to predict the damage and to forecast the future target of the object. The wealth of the world is in the health of the people. So the technology must support the technologists in predicting the disease in advance. The machine learning is the emerging field which is used to forecast the existence of the heart disease through the values of the clinical parameters. With this view, we focus on predicting the customer churn for the banking application. This paper uses the customer churn bank modeling data set extracted from UCI Machine Learning Repository. The anaconda Navigator IDE along with Spyder is used for implementing the Python code. Our contribution is folded is folded in three ways. First, the data is processed to find the relationship between the elements of the dataset. Second, the data set is applied for Ada Boost regressors and the important elements are identified. Third, the dataset is applied to feature scaling and then fitted to kernel support vector machine, logistic regression classifier, Naive bayes classifier, random forest classifier, decision tree classifier and KNN classifier. Fourth, the dataset is dimensionality reduced with principal component analysis with five components and then applied to the previously mentioned classifiers. Fifth, the performance of the classifiers is analyzed with the indication metrics like precision, accuracy, recall and Fscore. The implementation is carried out with python code using Anaconda Navigator. Experimental results show that, the Naïve bayes classifier is more effective with the precision of 0.90 for dataset with random boost, feature scaled and PCA. Experimental results show that, the Naïve bayes classifier is more effective with the recall of 0.91 for dataset with random boost, feature scaled and PCA. Experimental results show that, the Naïve bayes classifier is more effective with the Fscore of 0.92 for dataset with random boost, feature scaled and PCA. Experimental results show, the Naïve bayes classifier is more effective with the accuracy of 91% without random boost, 93% with random boosting and 92% with principal component analysis.
Publisher
Blue Eyes Intelligence Engineering and Sciences Engineering and Sciences Publication - BEIESP
Subject
Management of Technology and Innovation,General Engineering
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
1. Improved Machine Learning Algorithm for Heart Disease Prediction Based on Hyperparameter Tuning;2023 IEEE International Conference on Artificial Intelligence in Engineering and Technology (IICAIET);2023-09-12