Fraud detection system and method
AssigneeVail Systems, Inc.
InventorsSrinivasa Rao, Todd Whiteley
Life of this patent
Estimating the term…
Abstract
A system and method for fraud detection for a telephony platform based on an analysis of call detail records (CDRs) that are generated by the telephony platform. The analysis is based on collecting, organizing, transforming, analyzing, and quantifying the CDR data into a plurality of data analytics and data correlations and then applying fuzzy logic to the data analytics to generate a fraud risk rating for each incoming call into the platform.
Claims20
A fraud detection system for a telephony platform where call data records (CDRs) are generated for calls being serviced by the telephony platform, the fraud detection system comprising:
- a call analyzer that (i) queries a history database for one or more history records associated with metadata in an incoming call signal from the caller device, (ii) retrieves from the history database the history records associated with the metadata in the incoming call signal, and (iii) creates a caller analytics record (CAR) representing the retrieved history records;
- and
- a statistical analyzer that receives the caller analytics record (CAR) and one or more other caller analytics records (CARs) and computes a sum and at least one of a mean, a variance, and a standard deviation of an analytical metric across one or more analytical facets for the received caller analytics record (CAR) and the one or more other caller analytics records (CARs),
- wherein the call analyzer comprises one or more knowledge-based heuristic rules that determine an inconsistency in originating line information (OLI) values across the history records associated with the metadata in the incoming call signal, and, based on the determined inconsistency in originating line information (OLI) values, determines a fraud risk score associated with an analytics metric, and
- wherein the fraud detection system outputs an overall fraud risk score associated with the call signal.
A fraud detection system for a telephony platform where call data records (CDRs) are generated for calls being serviced by the telephony platform, the fraud detection system comprising:
- a call analyzer that:
- queries a history database for one or more history records associated with metadata in an incoming call signal from a caller device;
- retrieves from the history database the history records associated with the metadata in the incoming call signal;
- and
- creates a caller analytics record (CAR) representing the retrieved history records;
- and
- a statistical analyzer that receives the caller analytics record (CAR) and computes a sum and at least one of a mean, a variance, and a standard deviation of at least one analytics metric across at least one analytical facet for the received caller analytics record,
- wherein
- the call analyzer comprises one or more knowledge-based heuristic rules that determine an inconsistency in originating line information (OLI) values across the history records associated with the metadata in the incoming call signal, and, based on the determined inconsistency in originating line information (OLI) values, determines a fraud risk score associated with an analytics metric, or
- the statistical analyzer segments the caller analytics record (CAR) and one or more other caller analytics records (CARs) into different facets based on a facet ID included in each of the caller analytics record (CAR) and the one or more other caller analytics records (CARs), and
- wherein the fraud detection system outputs an overall fraud risk score associated with the call signal.
A fraud detection system for a telephony platform where call data records (CDRs) are generated for calls being serviced by the telephony platform, the fraud detection system comprising:
- a call analyzer that (i) queries a history database for one or more history records associated with metadata in an incoming call signal from a caller device, (ii) retrieves from the history database one or more history records associated with the metadata in the incoming call signal, and (iii) creates a caller analytics record (CAR) representing the retrieved one or more history records;
- and
- a statistical analyzer that receives the caller analytics record (CAR) and one or more other caller analytics records (CARs) and computes a sum and at least one of a mean, a variance, and a standard deviation of at least one analytics metric across an analytical facet for the received caller analytic record (CAR) and the one or more other caller analytics records,
- wherein the statistical analyzer segments the caller analytics record (CAR) and one or more other caller analytics records (CARs) into different facets based on a facet ID included in each of the caller analytics record (CAR) and the one or more other caller analytics records (CARs), and
- wherein the fraud detection system outputs an overall fraud risk score associated with the call signal.
Dependencies are inferred from the claim text.