What are false positives?
In principle, a "false positive" occurs when a check or test incorrectly identifies a match of criteria that does not actually exist. In contrast, "false negatives" are test results in which an actual match is not recognized or is incorrectly negated. False negatives not only cause additional work, as is the case with false positives, but could also have significant consequences. This is especially true if a problematic business contact has not been identified.
To illustrate this, let's take the following example: a screening tool is used to check business relationships for risks associated with politically exposed persons (PEPs). A "false positive" could occur in this scenario if the screening process incorrectly identifies a business relationship with a PEP when no real connection exists. This could lead to unnecessary checks and wasted resources, especially if the incorrectly identified person does not actually have any risky activities.
Causes of false positives in risk screening
Depending on the type of risk screening, there is a possibility of false positives. While there is flexibility in adjusting the accuracy of matching engines, in most cases they start with a broad approach to catch potential risks early. An overly restrictive setting could lead to overlooking suspicious business relationships or transactions. Using a comprehensive filter ensures that a wide range of possible risk factors are considered, which simultaneously increases the challenge of false positives.
The more names or data points that need to be matched, the higher the number of results and therefore the higher the probability of false positives. In addition, the data quality has a significant impact on the screening process. The following sources of error can lead to missed warnings or false hits:.
- incorrect spellings, misspellings and fonts of the world
- missing dates of birth
- missing addresses
Verifying these results often requires manual work, which is a time-consuming task and a drain on organizations' resources. Various factors contribute to making the identification of false positives more difficult. Among other things, unclear data sources, lack of standardization of information, and technical challenges in data integration and validation can lead to complications. The extent to which these factors hinder the identification of false positives depends heavily on the individual circumstances.
Consequences and possible solutions
The consequence is that only a small fraction of potential customers, clients or transactions are monitored with the level of detail, frequency and precision required for today's increasingly thorough compliance checks. Nevertheless, this is perfectly feasible, especially with high data quality and optimal list matching.
The quality and trustworthiness of the data running through the screening software (e.g. Partner Screening) is becoming increasingly important. Screening providers such as Pythagoras Solutions rely on high-quality data providers to drastically reduce the proportion of false positives in queries. The data providers connected to the screening software are freely selectable. They offer both unstructured data (news articles from licensed and renowned news sources, see Adverse Media) and structured data (including PEP lists, sanctions lists, adverse media and other data sets).
List matching solutions focus on two main areas to reduce false positives: the optimal configuration of the solutions and the meaningful contextualization of the scanning process.
Efficient compliance checks: Pythagoras Partner Screening
Pythagoras Solutions' Partner Screening offers advanced features to increase accuracy and efficiency. Comprehensive knowledge of different writing styles and fonts enables accurate matching. The integrable Native Character Screening extends this functionality by also recognizing non-Latin characters. This enables a global application that also meets local requirements.
The tool's intelligent search technology enables an exact comparison of your partner data with defined reference data. It delivers precise and consistent results, regardless of data volume, data type and data quality. In addition, the case management tool supports structured alert processing, allowing you to maintain control over all business relationships and document every step in accordance with regulatory and compliance requirements.
This enables companies to carry out their compliance checks more efficiently and effectively without wasting valuable resources on manually checking potential false positives.
