US 11,836,201 B2
System and method for automated data screening for background verification
Krishnam Raju Alamuri, Hyderabad (IN); Satish Raghavendran, Hyderabad (IN); and Saurabh Mazumdar, New Delhi (IN)
Assigned to COGNIZANT TECHNOLOGY SOLUTIONS INDIA PVT. LTD., Chennai (IN)
Filed by Cognizant Technology Solutions India Pvt. Ltd., Chennai (IN)
Filed on Sep. 10, 2021, as Appl. No. 17/471,477.
Claims priority of application No. 202041039257 (IN), filed on Sep. 11, 2020.
Prior Publication US 2022/0083615 A1, Mar. 17, 2022
Int. Cl. G06F 16/9535 (2019.01); G06F 16/9538 (2019.01); G06F 16/955 (2019.01)
CPC G06F 16/9535 (2019.01) [G06F 16/9538 (2019.01); G06F 16/9566 (2019.01)] 11 Claims
OG exemplary drawing
 
1. A method for automated data screening for background verification, wherein the method is implemented by a processor executing instructions stored in a memory, the method comprises:
analyzing a first input file and a second input file, wherein the first input file is representative of client data and Connected Parties (CPs) data and the second input file is representative of a hit file which provides hit details extracted from an internal database of an organization or from external data providers of the organization that are subscribed to by evaluating historical data;
performing a data enrichment operation on the first input file and the second input file based on captured client and CPs data from one or more Universal Resource Locators (URLs) which are extracted from one or more open media sources or from the data sources that the organization has subscribed to obtain an enriched first input file and second input file, and wherein one or more priority URLs are segregated from the extracted URLs for carrying out the data enrichment operation;
performing a matching operation between the enriched first input file and the second input file, wherein matched results of the first input file and the second input file are classified as a true match or a false match or a potential match;
determining adverse data associated with the clients and the CPs data determined as the true match and the potential match in the first input file and the second input file, and wherein a sentiment score is computed for each of the client and CPs data based on the extracted adverse data;
cleaning and filtering the extracted adverse data to generate screened data associated with the clients and the CPs data; and
generating an output folder comprising an output file including the screened client and CPs data and hit data.