Identiying each segmentation of the credit card customers and each of its characteristics with unsupervised learning.
Part of the projects done in Hacktiv8 (Aug 2022 - Oct 2022), though there might still be future updates containing new informations/improvements that can be added onto the project.
From the conclusion of the findings, there are three clusters of the credit card users that can be characterized as below:
Group 0 - low cash advance, mid purchase freq, low cash advance frequency, low cash advance trx
Group 1 - high cash advance, low purchase freq, high cash advance frequency, high cash advance trx
Group 2 - mid cash advance, high purchase freq, mid cash advance frequency, mid cash advance trx