Prediction of Customer Future Transaction using GBTD and Genetic Algorithm

  • Ritu Sharma, Yatin Aggarwal
Keywords: Genetic algorithm, Gradient Boosting Decision Tree (GBDT), Machine learning (ML), Data mining, Artificial Neural Network (ANN)

Abstract

Gradient Boosting Decision Tree (GBDT) is a trendy machine learning method with a number of good implementations, including XG Boost. Despite the use of many technical improvements in these implementations, the efficiency and scalability remain inadequate when the feature dimension is big, and the data amount is huge. Machine learning (ML) techniques are increasingly being used in the financial sector for a variety of applications. In the financial industry, there is constant competition to provide superior and the best consumer experiences than the other competitors. With the most important datasets that can be used is at the customer level of granularity. Various data mining methods are successfully employed in conjunction with developing data manipulation tools. The prediction procedure is carried out with the help of supervised and unsupervised learning techniques. This research implementation intends to improve human intuition using a strategy that can be assist institutions in improving customer satisfaction and detecting key aspects from anonymized datasets. The results of this research comprise of the comparison between the parameters such as accuracy, specificity, precision etc. are compared in a graphical manner.

Published
2021-09-08
How to Cite
Yatin Aggarwal, R. S. (2021). Prediction of Customer Future Transaction using GBTD and Genetic Algorithm. Design Engineering, 6695-6728. Retrieved from http://www.thedesignengineering.com/index.php/DE/article/view/4149
Section
Articles