DIBA - DATA CHALLENGE - ING-DIBA AG BART BUTER, DATA ENGINEERING / INTERNATIONAL ADVANCED ANALYTICS - FRANKFURT BIG DATA LAB
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
DiBa - Data Challenge ING-DiBa AG Bart Buter, Data Engineering / International Advanced Analytics Goethe Universität Frankfurt • 27. Okt 2016
Inhaltsübersicht 1. ING-DiBa Kurzprofil 2. Data Challenges 3. Organisation & Rules 2
Die ING-DiBa an einem Tag
2.500 Email-
20.000 Anrufe im
Eingänge pro Tag
Kundendialog
im Kundendialog
5,2 Mio. Seiten- 400.000 Logins im
aufrufe im Internet Internetbanking &
Brokerage
4.300 Kontakte in 160.000 Zugriffe
der Immobilien- Mobile Banking App
finanzierung
50.000 Vorgänge 29.000 Abhebungen
im Dokumenten an den 1.200 ING-DiBa
Service Geldautomaten
3
ING in Europa und weltweit
ING in Europa ING weltweit
Mehr als 34 Millionen private,
Firmen- und institutionelle Kunden
Mehr als 40 Länder in Europa,
Nord Amerika, Lateinamerika, Asien
und Australien
Mehr als 54.000 Mitarbeiter
Marktführer Benelux Wachstumsmärkte
Herausforderer Ausschließlich Wholesale Banking
4
Wir wollen „Digital Leader“ in der Finanzbranche sein.
Videolegitimation Fotoüberweisung
und Dokumenten-
Upload
Legitimation von
zu Hause aus
SmartSecure App
Fingerabdruck
Mobiler Kreditcheck
ersetzt TAN
Rechnungen und
Dokumente per
Kamera einscannen „Die Filiale der Zukunft haben viele Kunden
und direkt über- schon heute jederzeit und überall in der
Kreditwunsch weisen oder ein-
reichen Hosentasche dabei: das Smartphone“
mobil prüfen
5
Challenge 1: Income Detection using Public Data Knowing information about someone's income can be desirable in some cases and required in others. For instance, in loan processes this information is required and regulated: it needs to be substantiated by evidence, because proof of income is required to assess if someone is capable of repaying a loan. However, in other cases the requirements are less strict. For instance, when marketing a new investment product, this marketing campaign should be to a target audience that has disposable income to put into investments. For other customers this type of marketing will be irrelevant. Having an estimate of someone's income will therefore lead to fewer customers receiving irrelevant marketing and a higher success rate for the campaign. At the same time we want to simplify the banking experience for our customers and not ask them too many questions. Therefore, it might be more customer-friendly if we could ask for less information from a customer, but find different ways of estimating what the income of such a customer is. Challenge No 1: Can you give us an indication of someone’s income based on publicly-available data. Furthermore, since working with personal data carries considerable responsibility, we would like you to investigate the legal and ethical implications of your solutions. We would like you to define a minimal set of data which you would want from the customer and which you can enrich with public data e.g. statistics bureau data, publicly-available profiles, income comparison sites, etc. We expect you to demo a working prototype together with a legal and ethical assessment of the prototype and a documentation regarding the validation of your results. 6
Challenge 1
Data:
- Vacancy websites
- Public profiles
- Statistisches Bundesambt
- Chamber of Commerce
- Friends and Family
Legal Ethical Assesment:
- Are you allowed to use my Xing profile?
Is it desirable?
- Is there an API? What are the conditions
for using it?
Xing: Bart Buter – Head of Data Engineering Validation:
Gehalt.de: Leiter Data Engineering - Is my income truly 4709-7858 ??
Income: 4.709 – 7.858
7
Challenge 2: Next Generation Banking Currently, many financial apps show your current balance and past transactions, others have started to include complex graphs and visualizations, but is this the information our customers really want to see? Maybe some are only interested in knowing that their salary has been deposited into their account and if the utility bills have been paid; others might want to see how much they have spent on their new home theater setup. Again others might want to see how much money someone in their peer group spends on buying a new car, while certain customers are interested in how much they need to invest to have a comfortable retirement. This shows that not every customer has the same wishes and that it makes sense to think of solutions for a certain type of customer. Challenge No 2: Our challenge to you is to design the next generation data-driven banking app for your generation (Millennials born between 1980- 2000, Generation Z 2000-2020). Base your solution on what data is currently available to you in your banking environments and think about what is missing, or what you would like to see. Furthermore, creating applications for customers is not about what you think is a great idea, but rather about what they think is great to use. Therefore, you should identify features that are missing in the current banking apps, identify multiple solutions to solve this and find out what the best solution is. Base your decisions on the needs and wants of the customers, i.e. by doing customer research. In the end we expect a working prototype of your solution, which can be part of the next generation banking app for your generation. Additionally, we expect to see the results of the customer research that led you to the proposed solution. 8
Challenge 2:
Data:
- Your transactions from your bank.
- Friends and families
- Mobile phone
- Location
- Activity
Pitfall:
- I would like xyz. Therefore, it is a good idea
Validation:
- Test if your idea is shared by others.
- Do they have the problem you are trying
to solve?
- Do they agree that your solution is the
right solution?
9DiBa Data Challenge Team
Bettina Hartlich Christine Hennighausen Georgios Gkekas
Stevie Albert Bart Buter Organisation
Every team gets a mentor
Discuss office hours with
mentor
10Prizes 11
Rules - You are not working for DiBa, your ideas, solutions and actions are yours. - Financial matters and related data are private and sensitive, treat them as such. - When in doubt contact us or your professor. 12
Vielen Dank!
www.ing-diba.de
Bart Buter
Head of Data Engineering – Facebook.com/ingdiba
International Advanced Analytics
@ING_DiBa_Presse
ING-DiBa AG Telefon 069 / 27 22266776
Instagram.com/ingdiba
Theodor-Heuss-Allee 2
60486 Frankfurt am Main bart.buter@ing-diba.de
YouTube.com/ingdibaYou can also read
