Fraud detection and analysis
Systems and methods generate a risk score for an account event. The systems and methods automatically generate a causal model corresponding to a user, wherein the model estimates components of the causal model using event parameters of a previous event undertaken by the user in an account of the user. The systems and methods predict expected behavior of the user during a next event in the account using the causal model. Predicting the expected behavior of the user includes generating expected event parameters of the next event. The systems and methods use a predictive fraud model to generate fraud event parameters. Generation of the fraud event parameters assumes a fraudster is conducting the next event, wherein the fraudster is any person other than the user. The systems and methods generate a risk score of the next event to indicate the relative likelihood the future event is performed by the user.
1. A method comprising:
generating a first model using account event parameters of at least one account event in an account of a user via an online financial institution system, wherein the at least one account event in the account is at least one of: an online event, an offline event or a multiple-channel event;
using the first model, predicting expected behavior of the user during a future account event in the account via the online financial institution system;
generating fraud event parameters using a second model; and
generating with the first model and the second model a probability the future account event is performed by the user via the online financial institution system;
wherein generating the first model includes generating statistical relationships between the event parameters and derived parameters corresponding to the account, and
wherein the derived parameters include one or more of a geographic area from which a device is initiating the future account event to the online financial institution system, a location of the device, an identification of the device, and an electronic service provider of the device.
2. A method comprising:
automatically generating a first model using account event parameters of a previous account event performed by a user in an account, via an online financial institution system, wherein the first model includes distributions of the account event parameters;
receiving observed account event parameters of a next account event in the online financial institution system wherein the previous account event and the next account event are at least one of: an online event, an offline event or a multiple-channel event;
generating a first probability using the first model, wherein the first probability corresponds to the user conducting the next account event in the online financial institution system;
generating a second probability using a second model, wherein the second probability corresponds to a fraudster conducting the next account event in the online financial institution system, wherein the fraudster is a person other than the user; and
using the first probability and the second probability, generating a likelihood the next account event in the online financial institution system is performed by the user versus the fraudster;
wherein generating the first model includes generating statistical relationships between the account event parameters and derived parameters corresponding to the account, and
wherein the derived parameters include one or more of a geographic area from which a device is initiating the future account event to the online financial institution system, a location of the device, an identification of the device, and an electronic service provider of the device.
3. A system comprising at least one processor configured for:
generating a first model using account event parameters of at least one account event in an account of a user via an online financial institution system, wherein the at least one account event in the account is at least one of: an online event, an offline event or a multiple-channel event;
using the first model, predicting expected behavior of the user during a future account event in the account, in the online financial institution system;
generating fraud event parameters using a second model; and
generating with the first model and the second model a probability the future account event is performed by the user, in the online financial institution system;
wherein generating the first model includes generating statistical relationships between the account event parameters and derived parameters corresponding to the account, and
wherein the derived parameters include one or more of a geographic area from which a device is initiating the future event in the online financial institution system, a location of the device, an identification of the device, and an electronic service provider of the device.
4. A system comprising at least one processor configured for:
automatically generating a first model using account event parameters of a previous account event performed by a user in an account, via an online financial institution system, wherein the first model includes distributions of the event parameters;
receiving observed event parameters of a next account event in the online financial institution system;
wherein the previous account event and the next account event are at least one of: an online event, an offline event or a multiple-channel event;
generating a first probability using the first model, wherein the first probability corresponds to the user conducting the next account event in the online financial institution system;
generating a second probability using a second model, wherein the second probability corresponds to a fraudster conducting the next account event, in the online financial institution system, wherein the fraudster is a person other than the user; and
using the first probability and the second probability, generating a likelihood the next account event in the online financial institution system, is performed by the user versus the fraudster;
wherein generating the first model includes generating statistical relationships between the event parameters and derived parameters corresponding to the account, and wherein the derived parameters include one or more of a geographic area from which a device is initiating the next account event, in the online financial institution system, a location of the device, an identification of the device, and an electronic service provider of the device.
5. A computer program product comprising a non-transitory computer usable medium having computer readable program code embodied therein, the computer readable program code comprising instructions for:
generating a first model using event parameters of at least one account event in an account of a user via an online financial institution system, wherein the at least one account event in the account is at least one of: an online event, an offline event or a multiple-channel event;
using the first model, predicting expected behavior of the user during a future account event in the account in the online financial institution system;
generating fraud event parameters using a second model; and
generating with the first model and the second model a probability the future account event is performed by the user in the online financial institution system;
wherein generating the first model includes generating statistical relationships between the event parameters and derived parameters corresponding to the account, and wherein the derived parameters include one or more of a geographic area from which a device is initiating the future account event, in the online financial institution system a location of the device, an identification of the device, and an electronic service provider of the device.
6. A computer program product comprising a non-transitory computer usable medium having computer readable program code embodied therein, the computer readable program code comprising instructions for:
automatically generating a first model using event parameters of a previous account event performed by a user in an account, via an online financial institution system, wherein the first model includes distributions of the event parameters;
receiving observed event parameters of a next account event in the online financial institution system, wherein the previous account event and the next account event are at least one of: an online event, an offline event or a multiple-channel event;
generating a first probability using the first model, wherein the first probability corresponds to the user conducting the next account event in the online financial institution system;
generating a second probability using a second model, wherein the second probability corresponds to a fraudster conducting the next account event, in the online financial institution system, and
wherein the fraudster is a person other than the user; and
using the first probability and the second probability, generating a likelihood the next account event is performed by the user versus the fraudster in the online financial institution system;
wherein generating the first model includes generating statistical relationships between the event parameters and derived parameters corresponding to the account, and
wherein the derived parameters include one or more of a geographic area from which a device is initiating the next account event, in the online financial institution system, a location of the device, an identification of the device, and an electronic service provider of the device.