JapanFinance and insurance2018 – ongoingJapan Exchange Group
Machine learning for initial trading-surveillance investigations
Japan Exchange Group used AI for initial trading investigations while staff retained final judgment
Problem
Growth in high-speed trading increased the volume of orders that market surveillance had to examine. Investigators needed to complete initial reviews more quickly while preserving detailed scrutiny.
Approach
Japan Exchange Regulation began using AI in cash-equity trading surveillance in March 2018. The models learned investigators' assessments of unusual trading patterns. Staff continued to make final judgments and checked model outputs as the system was refined.
Results
No published figures
Takeaway
AI supported the initial review stage, allowing investigators to concentrate on detailed investigations. Japan Exchange Regulation did not quantify the time saved.
Categories
- Industry
- Finance and insurance
- Business function
- Finance, risk and legal
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Vendor solution
- Evidence
- Company's own claim
Sources
Source summaries
Japan Exchange Regulation described operational AI use in initial trading investigations from March 2018. It reported faster initial reviews and continued human responsibility for final judgments, without a numerical productivity result.
Full details are in the original sources.
