Efficient alert investigations
Traditional alert investigation/transaction monitoring involves an initial review to discount false positive alerts. As needed, this is followed by a detailed review of the customer, with the case referred for additional review if a Suspicious Activity Report is filed with the regulator.
Once again, automation can help, in these ways:
- Straight-through data processing for case creation: This approach can simplify data capture for case creation. The data gathered can be categorized and presented in a visual format (for example, by identifying outlier transactions) and in an automated manner to aid the investigation process.
- Media screening: NLP techniques such as named entity recognition, coreference resolution, relationship extraction and semantic role labeling can identify alerted entity and tag suspects. This approach along with intelligent tools and systems can identify patterns that are new and fed this information into the models, helping investigators make better decisions.
- ML for decision making: While final decisions involve human judgment, ML algorithms can review triggered activity to automate aspects of the decision-making process, by building statistical models that incorporate gathered data and calculate the likelihood of disposition of the alerted transactions, either for closure or escalation.
- Natural Language Generation (NLG) for narrative generation: For non-reported transactions (most transactions) NLG can generate a case narrative by translating data into case summary. This is one of the most challenging aspect in the entire investigation process; it is critical for financial institutions to focus on segments of alerts that can be effectively processed through NLG.
Getting it "right"
Several challenges might arise when implementing an intelligently automated financial crime compliance. These include addressing regulatory expectations, dealing with legacy systems, and designing an effective policy and governance approach.
We can help address those concerns. Accenture has relationships with numerous vendors who offer expertise and knowledge across the entire AML landscape, including lifecycle management, workflow management, data enrichment, threat identification, entity resolution and visualization of reports using dashboards.
We integrate these capabilities with our in-house tools and services to offer a solution aligned to each financial services provider’s needs. Learn how Accenture can help you automate your financial crime compliance for significant efficiency gains.
1 “Uncover the True Cost of Anti-Money Laundering & KYC Compliance,” LexisNexis, 2016. Access at: https://www.lexisnexis.com/risk/intl/en/resources/research/true-cost-of-aml-compliance-apac-survey-report.pdf. “The True Cost of Anti-Money Laundering Compliance – European Edition,” LexisNexis, September 2017, Access at: https://risk.lexisnexis.com/global/-/media/files/corporations%20and%20non%20profits/research/true%20cost%20of%20aml%
20compliance%20europe%20survey%20report%20.pdf.pdf. “Anti-money laundering compliance costs U.S. financial services firms $25.3 billion per year, according to LexisNexis Risk Solutions,” LexisNexis, October 10, 2018. Access at: https://risk.lexisnexis.com/about-us/press-room/press-release/20181010-true-cost-aml.
2 “Standard Chartered teams up with Instabase to automate and optimize client onboarding,” FinanceFeeds. November 7, 2018. Access at: https://financefeeds.com/standard-chartered-teams-instabase-automate-optimise-client-onboarding/. “Anti-Money Laundering and AI at HSBC,” Ayasdi, June 1, 2017. Access at: https://www.ayasdi.com/blog/financial-services/anti-moneylaunderinghsbc/.