Last week I described how Informatica Identity Resolution (IIR) can be used to match data from different lists or databases even when the data includes typos, translation mistakes, transcription errors, invalid abbreviations, and other errors. IIR has a wide range of use cases. Here are a few.
Financial services: Every year, banks and insurance companies are paying millions of dollars in fines and penalties for not complying with know-your-customer, Anti-money-laundering and unclaimed-property regulations. Not being able to accurately screen customer records against compliance lists (OFAC, Death Master File, Politically exposed people, etc…) is at the core if these compliance issues. Identity data in watch lists are often incomplete, spelled incorrectly and usually from various languages, countries and cultures. Romanization and transliteration of foreign names normally result in even more variations and inaccuracies.
Solution: With Standard Populations specifically designed for OFAC and Anti Money Laundering initiatives, IIR allows compliance officers to adjust the tolerances in the matching process to their comfort level, ensuring no matches are missed while minimizing the amount of false positives.
Data Aggregators: Organizations that make data their business are almost always hitting technology limitations in dealing with an increasing amount of data coming from more diverse sources. Home built matching solutions are costly to maintain and lack the flexibility to allow new types of data to be quickly matched. With data volumes quickly breaking the billion records barrier, many data aggregators are having difficulty maintaining their SLAs
Solution: IIR has a proven record in enabling extremely fast fuzzy search requests (>1000 searches / second) against very large data sets (>1 billion records). Used in core operations at the top credit bureaus and major marketing and industry information providers, IIR provides extreme flexibility allowing users to quickly support the constantly changing market demands.
Border control & Homeland security: Immigration and Visa issuing authorities are unable to accurately screen visa-seekers against lists of undesirable or dangerous individuals, resulting in non-matches or false matches. Similar problems exist for Border Control officers who are trying to prevent entry of undesirable visitors. Customs authorities lack a fast and accurate method to screen manifests against lists of potentially dangerous goods. Identity data in border control and visa watch lists are often incomplete, spelled incorrectly and usually from various languages, countries and cultures. Manifest data often contain free-form consignment descriptions, and frequently use acronyms and alternate words to describe contents Intelligence.
Solution: IIR provides a more accurate match of a potential visitor’s identity with watch list data, minimizing false positives and missed matches. Border Control agencies around the world use Informatica technology in a real-time environment, maintaining the same accuracy but improving throughput at airports and border crossings. Customs use Identity Resolution technology for fast, reliable screening of manifests for dangerous goods.
Law enforcement: Full identification and threat level of suspects cannot be accurately assessed in the field, putting law enforcement at risk. Similar problems exist when trying to match individuals with other alert lists (wanted persons, missing persons, known sex offenders, etc.).Full criminal history is not available for criminal justice system for purposes of making critical determinations such as bail or sentencing. Persons with criminal histories may use multiple aliases/identities in their “careers”. Current integrated criminal history systems do not have the ability to find & link all related identities. As a result, multiple identities are recorded for a single individual.
Solution: With IIR, Law enforcement agencies and courts are able to
- Tie together related criminal identities resulting in more complete criminal histories
- Run fast, accurate real-time searches against hot-files, and criminal histories
- Support the investigator in creating identity links across diverse sources of data
Welfare / Improper payments: Eligibility systems frequently add welfare recipients multiple times due to reapplications over time. Agencies that deliver cash, food and medical assistance to the poor, aged and disabled suffer losses from accidental or fraudulent duplicate payments. Case workers are not able to access complete information about beneficiaries and recipients across a variety of gov’t divisions and applications (Healthcare, Child Welfare, Family Support, Medicaid, etc…). Error & variations in multiple entries of identity data are compound by the fact that some states and jurisdictions do not require SSN for certain types of welfare or healthcare benefits. Unintentional duplicates are commonly created due to multiple guardians/custodians of minors
Solution: IIR provides a fast and accurate search for welfare eligibility verification and duplicate prevention enabling real-time search capability for case workers to see a consolidated view and history of welfare recipients. Grouping and linking welfare records with Human Services various divisions provides complete views and history of relationships.
Once again, it is indeed possible for a rule-based system, using time-tested algorithms, to turn data errors (garbage) into valuable business information.