What question did this study set out to answer?

To develop statistical relational learning models that help regulatory agencies prioritize oversight by predicting which securities brokers are most likely to commit serious regulatory violations.

August 21, 2005

Using relational knowledge discovery to prevent securities fraud

View Full Paper

Ask AI

Bookmark

To develop statistical relational learning models that help regulatory agencies prioritize oversight by predicting which securities brokers are most likely to commit serious regulatory violations.
Built predictive ranking models using statistical relational learning algorithms that incorporated organizational relationship data, such as past coworker connections.
Evaluated model performance using more than 18 months of unseen regulatory records alongside over two person-weeks of expert NASD staff assessments.
Model-generated risk predictions correlated highly with the subjective evaluations of experienced regulatory examiners.
Across all performance measures, the relational models performed as well as or better than the handcrafted rules used by the regulatory agency.

Discussion

No takes yet. Share an insight, caveat, or question.

Implication

Algorithmic evaluation study demonstrates superior prediction of regulatory violations among securities brokers, indicating the utility of relational learning in financial oversight.

Cite This Study

Neville et al. (2005) studied this question.

synapsesocial.com/papers/6a153f5372316eef384e21fc https://doi.org/10.1145/1081870.1081922

View Full Paper

Ask AI

Bookmark

Implication

Algorithmic evaluation study demonstrates superior prediction of regulatory violations among securities brokers, indicating the utility of relational learning in financial oversight.

Using relational knowledge discovery to prevent securities fraud

Key Points

Discussion

Implication

Cite This Study

Using relational knowledge discovery to prevent securities fraud

Key Points

Discussion

Implication

Cite This Study

Keep reading

Keep reading