Analyzis a dataset provided by an e-commerce marketplace called Olist, a leading e-commerce service that connects merchants to main marketplaces in Brazil. Dataset provide a wide range of offers from inventory management, dealing with reviews and customer contacts to logistic services.
The dataset consists of 9 cvs files, 100k orders from 2016 and 2018 that were made on Olist store.
Exploration of 2 specific business cases:
-
The impact of removing underperforming sellers?
(remove_underperforming_sellers.ipynb)
-
The impact of restricting customer states?
(restricting_customer_states.ipynb)
The excecutables :
data.py :
- Generate the Data, and Link the 9 tables between them according to the common keys.
Order.py :
- Features engineering, creation of Business insight related to orders.
Product.py :
- Features engineering, creation of product-related business insight
Seller.py :
- Features engineering, creation of business insight related to vendors
Utils.py :
- Function creation used in our 3 previous Python files