Corporate high-precision search algorithms automatically find information about individuals on the Web. The search involves scanning of social networks and other open sources. The acquired information gives an insight into the customer and update Bank's data files with information from independent sources.
The search for additional information may proceed by full name, date of birth, residence address, place of birth, etc.; place of employment; e-mail; phone numbers; details about the associated persons, etc. To promote the correct collation of various data formats, the Company applies intellectual normalization algorithms, followed by the assessment of the relevance of information to any specific individual.
Precision of these algorithms has been proven by hundreds of commercial projects involving the Double Data customers. Results of the hundreds of thousands of search requests have been verified manually for purposes of quality assurance. In all, the customers, commercial or otherwise, have run dozens of millions of successful searches.
Application of Big Data technologies to the collected data helps identify unique patterns, such as:
”a man is known by the company he keeps” – strong impact of environment on the borrower’s behavior;
Impact of the lifestyle and Internet usage on solvency of the individual;
Detection of fake accounts (“bots”).
To make ready solutions on the basis of the acquired data, Double Data applies the advanced approaches and Big Data analysis technologies: MapReduce; Amazon Web Services; Hadoop; Hive; mongoDB; scikit-learn; R; neo4j; etc.
We obtain vital information about Bank's customers by studying their social relationships and overlaying them with banking data. To optimize and to expedite solving of problems, the Company has designed and applied proprietary tools for statistical and visual analysis.
Double Data competencies include creation of high-load server solutions to integrate them in the loan provision process (the so-called ‘loan pipeline’). The corporate software products comply with the Federal Law “On Personal data” and with safety requirements applicable to IT systems within the Banks.
Corporate software products are installed on the server in the inside perimeter of the customer. All personal data processing is done by the software on the customer’s servers, meaning that no personal data would ever leave the inside perimeter.
The installed software acquires and processes data from public sources, which is totally in line with the Federal Law “On Personal Data”.