2. Money laundering is hard to detect.
Spread throughout government & local
municipalities like cancer.
Lead to Terrorism Financing.
Commercial Solutions are costly.
Off-the-shelf AML products need
customization with local know-hows.
3. Machine Intelligence to detect Money
Laundering and counter Terrorism
Financing with human friendly reporting.
4. Use proven concepts from other regulators
(outside of Thailand).
Applying to domestic scenario of
separatism in the Deep South and local
corruption.
https://fb.com/Watchdog.ACT/
https://fb.com/groups/OpenDataASEAN/
DataKind experts
5. Use training data from fintechsandbox.org and use
DataRobot to create a prototype
Obtain test data from regulators
Store in Hadoop / Apache Spark using Cloudera
Enterprise
Prep data using Alteryx
Feed them to DataRobot to find rankings of more
complex ML models
Visualize findings and generate reports with Tableau
Add narration with Quill and Wordsmith
Use Splunk for ad hoc queries
6. Financial Institutions
Commercial banks with presence in Thailand.
Regulators
BoT: Bank of Thailand
MAS: Monetary Authority of Singapore
FCA: Financial Conduct Authority (UK)
ASIC: Australian Securities & Investments Commissions
FINRA: Financial Industry Regulatory Authority
ESMA: European Securities and Markets Authority
IOSCO: International Organization of Securities Commissions
FINMA: Swiss Financial Market Supervisory Authority
BaFin: Federal Financial Supervisory Authority (Germany)
7. Subscription model for financial institutions
Long-term Enterprise Licensing for
Regulators
8. Tanat Tonguthaisri
https://LinkedIn.com/in/epicure
Tools
Cloud services (AWS, Azure, or Google Cloud)
Cloudera (Big Data via Apache Spark / Hadoop)
Alteryx (Data Preparation)
DataRobot (ML automation)
Tableau (Visualization)
Splunk (ad hoc querying & Hunk for Hadoop)
Narrative Science (Quill for narration)
Automated Insights (Wordsmith for narration)