Identifying-Customer-Segments for Avarto
This project uses Unsupervised Machine Learning techniques including Dimensionality Reduction and K-means Clustering to identify potential customers for Avarto. First, Demographic data is used to identify clusters using K-means. These clusters are then compared with existing customers to identify the group of people that could be potential new customers to the company.
Data
As part of the agreement with Avarto, the data could not be provided publicly here.
Code
The IPython Notebook with the code and visualizations can be viewed here.