Starbucks Capstone Challenge This data set contains simulated data that mimics customer behavior on the Starbucks rewards mobile app. Once every few days, Starbucks sends out an offer to users of the mobile app. An offer can be merely an advertisement for a drink or an actual offer such as a discount or BOGO (buy one get one free). Some users might not receive any offer during certain weeks.
- Installation
- File Descriptions
- Results
- Licensing, Authors, Acknowledgements
Pyhton version3 is used to run the code effectively. No licence is needed.
The related files are attached to the project files. JSON files are provided by Starbucks and Udacity. The data is contained in three files:
- portfolio.json - containing offer ids and meta data about each offer (duration, type, etc.)
- profile.json - demographic data for each customer
- transcript.json - records for transactions, offers received, offers viewed, and offers completed
1-) Insights from visualizations
-
Men customers between the ages of 20 and 45 responded to offers 7 and 8 more favorably than other offers.
-
Men customers between the ages of 45 and 60 responded to alternative offers instead of 7 and 8.
-
Men customers with incomes between 35 and 50K replied to alternative offers instead of 4,6,9.
-
Women tend to devalue offers but are less sensitive to the types of offers. Expand the discounts available to women.
-
Women customers with incomes between $65,000 and $85,000 responded to offers 1 and 10.
-
Pay particular attention to offers 7,8,5 and 1. Higher retention rate.
-
The retention rate is often greater for third- and fourth-year members.
2-) Model results
I have used 2 different models to select the effective one. Decision tree is normally used to prospective customers with using demographics data. Random forest also has similar usages so I try both of them. Decision Tree gives the best results according to F1 score.
You can check detailed report from here