Automated entity identification for efficient profiling in an event probability prediction system
A computer-implemented method and system for automated entity identification for efficient profiling in an event probability prediction system. A first subset of entities belonging to one or more entity classes is defined. At least one historical profile is constructed for each entity in the subset of entities based on a set of possible outcomes of transaction behavior of each entity in the first subset of entities. Based on the historical profiles, a second subset of entities having transaction behavior associated with a transaction is selected, the transaction behavior being predictive of at least one targeted outcome from the set of possible outcomes. The first subset of entities is redefined with the second subset of entities.
1. A computer-implemented method comprising:
generating, by the one or more data processors, a historical profile for each entity in a first subset of entities, the historical profile being based on a set of possible outcomes of transaction behavior of each entity, the first subset of entities belonging to one or more entity classes;
selecting, by the one or more data processors, a second subset of entities based on the historical profiles, each entity in the second subset of entities having transaction behavior predictive of at least one targeted outcome from the set of possible outcomes of transaction behavior;
generating, by the one or more data processors, a historical profile for each entity in the second subset of entities;
combining, by the one or more data processors, the second subset of entities with the first subset of entities to generate a third subset of entities; and
monitoring, by the one or more data processors, the historical profiles for the third set of entities for transaction behavior indicative of a fraud process.
2. The computer-implemented method of claim 1 , wherein the targeted outcome is selected from the set of possible outcomes that consists of a desirable outcome and an undesirable outcome.
3. The computer-implemented method of claim 1 , wherein the targeted outcome includes the fraud process in connection with the transaction behavior.
4. The computer-implemented method of claim 1 , wherein at least one entity class includes automatic teller machines.
5. The computer-implemented method of claim 1 , further comprising:
generating, by the one or more data processors, at least one score related to the transaction based on one or more mathematical models; and
combining the at least one score with information of the transaction behavior to generate a probability of the targeted outcome.
6. The computer-implemented method of claim 1 , further comprising:
generating, by the one or more data processors, at least one score related to the transaction based on one or more mathematical models; and
combining the at least one score with information from the historical profiles for the entities involved in the transaction to produce an accurate score.
7. The computer-implemented method of claim 6 , wherein the historical profiles are used only if the accurate score is within a predetermined range of values that indicate a high probability of the targeted outcome.
8. The computer-implemented method of claim 1 , further comprising generating an estimate of the probability of the targeted outcome based only on the historical profiles.
9. The computer-implemented method of claim 1 , wherein the values in the historical profile for a selected entity are computed using information from transaction records that involved the selected entity.
10. The computer-implemented method of claim 9 , wherein the values in the historical profile for an entity are computed using input variables and output scores computed for the transaction records by one or more mathematical models.
11. A system comprising:
at least one programmable processor; and
a machine-readable medium storing instructions that, when executed by the at least one processor, cause the at least one programmable processor to perform operations comprising:
generate a historical profile for each entity in a first subset of entities, the historical profile being based on a set of possible outcomes of transaction behavior of each entity, the first subset of entities belonging to one or more entity classes;
select a second subset of entities based on the historical profiles, each entity in the second subset of entities having transaction behavior predictive of at least one targeted outcome from the set of possible outcomes of transaction behavior;
generate a historical profile for each entity in the second subset of entities;
combine the second subset of entities with the first subset of entities to generate a third subset of entities; and
monitor by the one or more data processors, the historical profiles for the third set of entities for transaction behavior indicative of a fraud process.
12. The system of claim 11 , wherein the targeted outcome is selected from the set of possible outcomes that consists of a desirable outcome and an undesirable outcome.
13. The system of claim 11 , wherein the targeted outcome includes the fraud process in connection with the transaction behavior.
14. The system of claim 11 , wherein at least one entity class includes automatic teller machines.
15. The system of claim 11 , further comprising generating an estimate of the probability of the targeted outcome based only on the historical profiles.
16. The system of claim 11 , in which the values in the historical profile for a selected entity are computed using information from transaction records that involved the selected entity.
17. The system of claim 16 , in which the values in the historical profile for an entity are computed using input variables and output scores computed for the transaction records by one or more mathematical models.