Capstone project for data scientist nano degree
This is for the capstone project of Udacity Data Scientist Term 2.
-
pandas
-
numpy
-
os
-
matplotlib
-
seaborn
-
sklearn
-
xgboost
-
operator
The major motivation of this project is to obtain the customer demographic characteristics of a company and
use the obtained information to build a model to predict customers that can be targeted in a marketing campaign.
- Readme.md
- Scipt for the project execution
- Predicted results on the test data
There are four data files associated with this project:
- Udacity_AZDIAS_052018.csv: Demographics data for the general population of Germany; 891 211 persons (rows) x 366 features (columns).
- Udacity_CUSTOMERS_052018.csv: Demographics data for customers of a mail-order company; 191 652 persons (rows) x 369 features (columns).
- Udacity_MAILOUT_052018_TRAIN.csv: Demographics data for individuals who were targets of a marketing campaign; 42 982 persons (rows) x 367 (columns).
- Udacity_MAILOUT_052018_TEST.csv: Demographics data for individuals who were targets of a marketing campaign; 42 833 persons (rows) x 366 (columns).
There are also three spreadsheets to decribe features in the data:
- DIAS Information Levels - Attributes 2017.xlsx: Information for the feature levels
- DIAS Attributes - Values 2017.xlsx: Information for description and value meaning of each feature
- feature_info_reorganixed.csv: Reorgnized feature infomation datasheet to make it consistent with the population dataset
K-means clustering found money-savers and people owning high-end cars are most likely to be customers, and people with low income,living in rural area and possibly much old in age are unlikely to be customers. Supervised learning model were trained and tested based on labeled data; it was used to predict the responders in the Kaggle Competition.
I would like to thank Arvato Bertelsman providing the dataset for our capstone project, and also the help I received from all Udacity mentors and students.