This project was part of Udacitys Nanodegree for data scientist program. The task was to identify customer segments from an actual dataset provided from a German mailorder company. The data itself could not be disclosed.
I used PCA and Kmeans clustering to identify 5 unique customer segments from tha data. The most important features for identifying new customers for the mail order company was people with relativly low income and low social status