SERIF is a family of fraud detection, analysis and investigation solutions for insurance companies, which, by using Big Data and Data Science technologies, enables the analysis of large volumes of data optimizing the results.
Analysis of digital files (for example, documents and images). This solution detects inconsistencies or alterations in the structure, the content and the format of the files, generating alerts associated with fraud indications (tampering of documents or images, inconsistencies in dates, locations or authorship of the document, etc).
SERIF Image Screening applies multiple image processing techniques and forensic analysis algorithms for digital files that enable insurance companies to control potential fraud channels.
Analysis of relations between the data stored in the insurance company's systems. This tool models all the data existing in the company, allowing for the visualization of different types of relationships. An interactive interface facilitates the investigation of insured accidents, people or assets, as well as the discovery of potential fraud networks.
SERIF Link Analysis offers different types of data visualization (network, map and statistical graphics among others) and different functions that speed up the exploitation of the information, such as searching for the path existing between two entities (people, suppliers, vehicles, etc). Additionally, it has a module that allows for the generation and validation of fraud rules in a simple and user-friendly way.
Analysis of the public information available on the Internet connected to the accidents of an insurance company. This solution enables the enrichment of the company's internal data, often of low quality and incomplete, with external data that is considered to be relevant to the study of the accident.
SERIF Social Media automatically detects fraud indications based on the information and inconsistencies drawn from external sources.
the time spent in
investigating an accident.
to those accidents with a higher probability of fraud.
decision‐making by means of short response time.
large volumes of data through the use of Big Data technologies.
Automatic data cleaning and correction of errors.
internal and external data from various sources (the internet, social networking sites, etc).
sources that generate streaming data and historic data sets of the company.
structured and unstructured data in a unified way.
fraud indications based on various sources of information.