Machine Learning for Chief Future Officers

Data Science for Finance Meetup

Transcript

1. Machine Learning for Chief Financial Future Officers Data Science in Finance Meetup - March 10, 2016 Christophe Bourguignat - @chris_bour
2. Disclaimer And Background @chris_bour
3. Data breaks down domain frontiers “ The winning team included a mathematician and an engineer, but no doctor ”
4. CONTEXT AND ISSUES
5. The CFO as an organization control tower Data - Invoices - Orders - Payments - HR - Contracts - ...
6. Uses case : suppliers invoices processing
7. Reporting...
8. @chris_bour … lot of reporting
9. If people could see in high dimensions machine learning would not be necessary “ ” “A Few Useful Things to Know about Machine Learning”, Pedro Domingos
10. a bit of THEORY
11. Raw Data
12. Framing The Problem Id Supplier Id Budget Code Buyer Id Order Date Amount Invoice Date Payment Date Ageing 34 SH24 FR657 B26 12/03/14 7600 15/06/14 25/06/14 10 35 SH87 FR425 B22 02/03/14 12390 10/07/14 30/07/14 20 98 SH65 FR034 B33 02/10/14 72980 01/11/14 25/11/14 24 With these (heterogeneous) features .... … predict that target
13. Features engineering (1/2) Original features Additional engineered features @chris_bour
14. Features engineering (2/2) Order Date -> Day of the week, month, holiday indicator Invoice Date -> Day of the week, month, holiday indicator, number of unpaid invoices to date Order Date, Invoice Date -> number of days between Order and Invoice Supplier ID -> average order amount, average number of order per year, size of the supplier, revenue of the supplier Budget Code -> average number of order per year, % of budget Buyer Id -> average number of order per year
15. Where we are now 2 3 0 5 3 7800 23 12 57 3 4 0 4 9 12300 2 54 10 1 10 1 4 1 5440 4 2 78 5 3 0 3 2 1500 0 67 23 4 12 1 5 1 54988 2 34 76 20 000 rows 30 columns 36 72 14 38 22 target
16. Training a Machine Learning Algorithm Machine Learning Algorithm Predictive model Prediction New invoice Features Target
17. TOOLS
18. Data Science Tools : choose your weapon
19. Selection criteria - Small / medium Data (< 1 GB) - Batch - Flat Files Data Sources - Open Source - Machine Learning - (French )
20. Selection criteria - Small / medium Data (< 1 GB) - Batch - Flat Files Data Sources - Open Source - Machine Learning - (French )
21. CODE
22. Python Code Feature Engineering Modeling and cross validation (VS a mean payment delay of 50 days)
23. Features importance
24. Train / Test split advanced strategies (1/2) Problem with random train / test split : bulk supplier payments makes the model performance over optimistic Supplier ID Invoice Date Payment Date S921 20/02/2013 31/03/2013 S921 20/02/2013 31/03/2013 S921 20/02/2013 31/03/2013 Supplier ID Invoice Date Payment Date S921 20/02/2013 31/03/2013 S921 20/02/2013 31/03/2013 Supplier ID Invoice Date Payment Date S921 20/02/2013 31/03/2013 Train Test Test set contains observations already seen in train set. This is cheating ...
25. Train / Test split advanced strategies (2/2) Better split by date, coherently with the fiscal period E.g : train = fiscal year N. Test = fiscal year N+1
26. Accuracy improvement More Data (High Data) More Feature Engineering (Wide Data) Advanced models Advanced tuning
27. THE BEGINNING, NOT THE END
28. The Loneliness of The Data Scientist who made a ML model Will anybody ever use it ?
29. Issue 1 : Change Management Process Payment Delay Prediction Process A Process B High Standard Before After
30. Issue 2 : Friction-Less Data Apps
31. No UI is the new UI
32. FURTHER IDEAS
33. Other Use Cases Late payement risk prediction Cash forecasting Audit / anomaly detection Contracts assessements Automated validation Smart Procurement ... @chris_bour
34. Thank You. Questions ? @chris_bour P.S. : “Le Meilleur Data Scientist de France” - 10/05/2016