Divergent trend detection and mitigation computing system
A computer system may be configured to analyze a set of data and detect divergent trends of one or more subsets of the set of data. The set of data may include reactions of a plurality of customers in response to an action of an organization. For example, the action may relate to goods or services offered by the organization, and the reactions may include responses of the customers to the goods or services. The customer reactions may include divergent trends that indicate a source of risk to the organization. Once detected, the computer system may send an alert to the organization related to the divergent trends, such that the organization may address a source of risk as indicated by the divergent trends. In some examples, the computer system may further offer or autonomously execute a mitigating action to address the source of risk.
1. A method comprising:
identifying, by a computing device, an action related to a robo-advisor and performed by an institution that is directed to a set of customers, wherein the action includes advice provided by the robo-advisor;
identifying, by the computing device, a probabilistic distribution of reactions to the action of the set of customers in response to identifying the action;
gathering, by the computing device, a set of reactions of the set of customers to the action, wherein the set of reactions includes at least one reaction for each customer of the set of customers;
determining, by the computing device, that the set of reactions is within a threshold percentage of the probabilistic distribution;
identifying, by the computing device, a plurality of identifying characteristics of customers of the set of customers, wherein at least some of the plurality of identifying characteristics are mutually exclusive with others of the plurality of identifying characteristics;
identifying, by the computing device, a subset of customers in which each customer of the subset relates to a common identifying characteristic of the plurality of identifying characteristics, wherein identifying the subset includes identifying a corresponding subset of reactions of customers of the subset of customers;
determining, by the computing device, that the subset of reactions of the subset of customers is more than the threshold percentage outside of the probabilistic distribution; and
performing, by the computing device, one or more mitigating actions to alter the advice provided by the robo-advisor such that the subset of reactions are less than or equal to the threshold percentage outside of the probabilistic distribution.
2. The method of claim 1 , further comprising gathering customer data of the set of customers in response to gathering the set of reactions, wherein the customer data includes the identifying characteristics of the set of customers.
3. The method of claim 1 , wherein the action relates to a sale or offer for sale of the one or more products.
4. The method of claim 1 , further comprising:
determining, by the computing device, a refined probabilistic distribution of reactions to the action of the subset of customers in response to identifying the subset of customers; and
wherein the determining that the probabilistic distribution of the subset of reactions of the subset of customers is more than the threshold percentage outside of the probabilistic distribution includes comparing the subset of reactions of the subset of customers to the refined probabilistic distribution.
5. The method of claim 1 , wherein performing the one or more mitigating actions to alter the advice further comprises:
autonomously altering, by the computing device, an aspect of the advice provided by the robo-advisor in response to determining that the distribution of the subset of reactions of the subset of customers is more than the threshold percentage outside of the probabilistic distribution.
6. The method of claim 5 , wherein the automatically altering an aspect of the advice provided by the robo-advisor is further in response to determining that a current time is outside of standard business hours.
7. The method of claim 1 , wherein the determining the probabilistic distribution of reactions to the action of the set of customers further comprises:
gathering historical reaction data that is in response to historical actions related to one or more historical products; and
extrapolating the probabilistic distribution of reactions from the historical reaction data.
8. The method of claim 5 , wherein performing the one or more mitigating actions to alter the advice further comprises:
identifying, by the computing device and prior to autonomously altering the advice provided by the robo-advisor, the aspect of the advice provided by the robo-advisor as having a potential causal relationship with the subset of reactions that is more than the threshold percentage outside of the probabilistic distribution.
9. The method of claim 1 , wherein performing the one or more mitigating actions to alter the advice includes updating, by the computing device, an algorithm of the robo-advisor.
10. The method of claim 1 , wherein the gathering the set of reactions of the set of customers further comprises:
monitoring activity of the set of customers subsequent to the action; and
identifying some activity of the set of customers as reactions to the action.
11. A computing device comprising:
at least one processor; and
a memory coupled to the processor, the memory storing instructions that, when executed, cause the at least one processor to:
identify an action related to a robo-advisor and performed by an institution that is directed to a set of customers, wherein the action includes advice provided by the robo-advisor;
identify a probabilistic distribution of reactions to the action of the set of customers in response to identifying the action;
gather a set of reactions of the set of customers to the action, wherein the set of reactions includes at least one reaction for each customer of the set of customers;
determine that the set of reactions is within a threshold percentage of the probabilistic distribution;
identify a plurality of identifying characteristics of customers of the set of customers, wherein at least some of the plurality of identifying characteristics are mutually exclusive with others of the plurality of identifying characteristics;
identify a subset of customers in which each customer of the subset relates to a common identifying characteristic of the plurality of identifying characteristics, wherein identifying the subset includes identifying a corresponding subset of reactions of customers of the set of customers;
determine that the subset of reactions of the subset of customers is more than the threshold percentage outside of the probabilistic distribution; and
perform one or more mitigating actions to alter the advice provided by the robo-advisor such that the subset of reactions are less than or equal to the threshold percentage outside of the probabilistic distribution.
12. The computing device of claim 11 , the memory further storing instructions that, when executed, cause the at least one processor to gather customer data of the set of customers in response to gathering the set of reactions, wherein the customer data includes the identifying characteristics of the set of customers.
13. The computing device of claim 11 , the memory further storing instructions that, when executed, cause the at least one processor to:
determine a refined probabilistic distribution of reactions to the action of the subset of customers in response to identifying the subset of customers; and
wherein the determining that the probabilistic distribution of the subset of reactions of the subset of customers is more than the threshold percentage outside of the probabilistic distribution includes comparing the subset of reactions of the subset of customers to the refined probabilistic distribution.
14. The computing device of claim 11 , wherein at least one reaction of the set of reactions includes inaction.
15. The computing device of claim 11 , the memory further storing instructions related to the determining the probabilistic distribution of reactions to the action of the set of customers that, when executed, cause the at least one processor to:
gather historical reaction data that is in response to historical actions related to one or more historical products; and
extrapolate the probabilistic distribution of reactions from the historical reaction data.
16. A non-transitory computer-readable medium comprising instructions, that when executed, cause one or more processors of a computing device to:
identify an action related to a robo-advisor and performed by an institution that is directed to a set of customers, wherein the action includes advice provided by the robo-advisor;
identify a probabilistic distribution of reactions to the action of the set of customers in response to identifying the action;
gather a set of reactions of the set of customers to the action, wherein the set of reactions includes at least one reaction for each customer of the set of customers;
determine that the set of reactions is within a threshold percentage of the probabilistic distribution;
identify a plurality of identifying characteristics of customers of the set of customers, wherein at least some of the plurality of identifying characteristics are mutually exclusive with others of the plurality of identifying characteristics;
identify a subset of customers in which each customer of the subset relates to a common identifying characteristic of the plurality of identifying characteristics, wherein identifying the subset includes identifying a corresponding subset of reactions of customers of the set of customers;
determine that the subset of reactions of the subset of customers is more than the threshold percentage outside of the probabilistic distribution; and
perform one or more mitigating actions to alter the advice provided by the robo-advisor such that the subset of reactions are less than or equal to the threshold percentage outside of the probabilistic distribution.
17. The non-transitory computer-readable storage medium of claim 16 , further comprising instruction that, when executed, cause the one or more processors to gather customer data of the set of customers in response to gathering the set of reactions, wherein the customer data includes the identifying characteristics of the set of customers.
18. The non-transitory computer-readable storage medium of claim 16 , further comprising instruction that, when executed, cause the one or more processors to:
determine a refined probabilistic distribution of reactions to the action of the subset of customers in response to identifying the subset of customers; and
wherein the determining that the probabilistic distribution of the subset of reactions of the subset of customers is more than the threshold percentage outside of the probabilistic distribution includes comparing the subset of reactions of the subset of customers to the refined probabilistic distribution.
19. The non-transitory computer-readable storage medium of claim 16 , wherein at least one reaction of the set of reactions includes inaction.
20. The non-transitory computer-readable storage medium of claim 16 , further comprising instruction related to the determining the probabilistic distribution of reactions to the action of the set of customers that, when executed, cause the one or more processors to:
gather historical reaction data that is in response to historical actions related to one or more historical products; and
extrapolate the probabilistic distribution of reactions from the historical reaction data.