Skip to content

Latest commit

 

History

History
30 lines (22 loc) · 1.89 KB

File metadata and controls

30 lines (22 loc) · 1.89 KB

Machine Learning
Customer segmentation project

Overview

Final project of the course "Machine Learning" in which we tried to answer the following question: "is it possible to divide the customers of an e-commerce site on the basis of their purchases in order to improve marketing communications?"
To answer this question, the Recency-Frequency-Monetary (RFM) model was applied which made possible to cluster customers by identifying 3 distinct groups but with homogeneous characteristics that would allow a more incisive marketing communication. Various evaluation techniques were then implemented for the clusters obtained through the use of internal indices such as Rand and Silhouette and the validity paradigm was applied to verify the hypotesis that there is no structure on the dataset.
The dataset we used was found on kaggle and can be downloaded from the following links:

Software

The project was carried out with the use of Knime software and with the help of R where necessary for operational purposes.

File

  • Report: It describes all the steps and choices made in italian languages
  • Project: Export of the developed Knime project
  • Image: contains the image of the complete knime workflow

About us

Riccardo Confalonieri - Data Science Student @ University of Milano-Bicocca

Rebecca picarelli - Data Science Student @ University of Milano-Bicocca

Silvia Ranieri - Data Science Student @ University of Milano-Bicocca