Computer-aided decision systems

• US 10,185,917 B2
• Filed: 07/09/2014
• Issued: 01/22/2019
• Est. Priority Date: 01/31/2013
• Status: Active Grant

• Current Assignee: LF Technology Development Corporation Limited
• Original Assignee: LF Technology Development Corporation Limited
• Inventors: Greystoke, Alex, Blick, Shy
• Primary Examiner(s): Airapetian, Mila
• Application Number:
  US14/327,543
• Publication Number:
  US 20150012467A1
• Time in Patent Office: 1,658 Days
• Field of Search: 705 1454
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/045 Explanation of inference; Explainable artificial intelligence [XAI]; Interpretable artificial intelligence
• CPC Class Codes: G06Q 30/0619 Neutral agent

Abstract

Decision systems and processes implementing digital personas are presented. A digital persona is a digital representation of an entity in accordance with a specific set of rules, preferences, or priorities with respect to a defined situation or opportunity. A digital persona may interact with a universe, which can be a set of conditions and information that an artificial intelligence engine implementing the digital personas can perceive. The digital personas can learn, via the artificial intelligence engine, from actions of a user, events in the universe, other personas, or a multitude of other factors. In some examples discussed, the artificial intelligence engine may include a persona artificial intelligence engine and an evolutionary artificial intelligence engine.

Claims

1. A method comprising:

normalizing search results corresponding to a query at a normalizer of a digital persona decision system by extracting data from the search results, transforming the extracted data into a format suitable for a table, loading the data into the table, and processing the data in the table to produce a set of multiple possible solutions;

configuring an artificial intelligence engine using one or more personas of a plurality of personas, the one or more personas corresponding to digital representations of different decision-making characteristics of an entity that initiated the query, the entity including at least one of a user communicating with the digital persona system through a network via a computing device and a computing device operating under control of a computer program;

processing the set to weigh and score each of the multiple possible solutions using the artificial intelligence engine configured according to the one or more personas of a plurality of personas;

determining a selected solution from the multiple possible solutions using the artificial intelligence engine configured according to the one or more personas and based on the scores; and

sending the selected solution within a graphical interface to one of the computing device associated with the user and the computing device operating under the control of the computer program.

2. The method of claim 1, wherein processing the multiple possible solutions comprises:

selecting a persona of the one or more personas;

configuring the artificial intelligence engine with the selected persona;

determining a score for each parameter of a plurality of parameters for each of the multiple possible solutions based on the selected persona; and

repeating the selecting, the configuring, and the determining for each of the one or more personas.

3. The method of claim 1, wherein determining the selected solution comprises:

determining a score for each parameter of each solution of the multiple possible solutions based on each of the one or more personas;

determining an average of all of the scores for each parameter;

summing the scores for each solution of the multiple possible solutions; and

determining the selected solution based on the sum.

4. The method of claim 3, further comprising:

determining a veto reference value associated with a selected digital persona of the one or more digital personas, the veto reference value corresponding to a parameter of the multiple possible solutions; and

when a veto reference value is present for the parameter, selecting a solution of the multiple possible solutions for the selected digital persona based on the score for the parameter.

5. The method of claim 3, wherein determining the score comprises:

determining a relative power differentiation of each of the one or more digital personas and for each parameter; and

selectively amending the score for each parameter for each of the one or more digital personas based on the relative power differentiation.

6. A computing system comprising:

an interface configured to communicate with one or more devices through a communications network;

a processor coupled to the interface; and

a memory accessible to the processor and configured to store instructions that, when executed, cause the processor to;

select one or more personas of a plurality of personas, the one or more selected personas representing at least one of a user of a first device or a program executing on a second device that sent a query to the interface, each of the one or more personas corresponding to a digital representations of a plurality of decision-making characteristics that is different in at least one of the plurality of decision-making characteristics from others of the selected one or more personas;

configure a persona artificial intelligence (AI) engine using the one or more selected personas of a plurality of personas, the persona AI engine configured by the one or more selected personas to receive multiple possible solutions, weigh and score the multiple possible solutions based on each of the one or more selected personas to provide multiple scored solutions;

select a subset of the multiple scored solutions using at least one of the one or more selected personas; and

provide a graphical interface including at least one result from the subset to one of the first device and the second device.

7. The computing system of claim 6, wherein the memory further includes instructions that, when executed, cause the processor to:

process information pertaining to one or more personas to produce a persona adjustment; and

selectively modify at least one persona based on the persona adjustment;

wherein the selectively modified at least one persona can subsequently be selected and applied to the AI engine to cause decision-making by the AI engine to evolve over time based on the selectively modified at least one personal.

8. The computing system of claim 7, wherein the information comprises a plurality of snapshots associated with one or more personas.

9. The computing system of claim 7, wherein the evolutionary AI engine selectively modifies the at least one persona in response to a triggering event.

10. A system comprising:

an interface configured to communicate with one or more data sources through a network;

a processor coupled to the interface;

a memory coupled to the processor and configured to store a plurality of digital personas and to store instructions that, when executed, cause the processor to;

select one or more digital personas from the plurality of digital personas in the memory to represent a user of a computing device that sent a query to the interface, each of the one or more personas corresponding to a digital representation of a plurality of decision-making characteristics that is different in at least one of the plurality of decision-making characteristics from others of the selected one or more personas;

configure an artificial intelligence (AI) engine with the selected one or more digital personas to interact with the one or more data sources to determine a list of potential solutions to a problem, the AI engine configured to weigh and score each of the potential solutions based on each of the selected one or more personas to provide multiple scored solutions; and

execute a selector module to select one of the multiple scored solutions as an output based on the digital persona; and

provide a graphical interface including the output to the computing device through the network.

11. The system of claim 10, further comprising a query normalizer module executable by the processor to:

normalize a query into data formatted for a particular search;

cause the query to be performed;

receive results based on the query; and

provide the results to the artificial intelligence engine.

12. The system of claim 10, wherein the artificial intelligence engine further comprises:

a persona artificial intelligence engine configured to apply the persona to search results; and

an evolutionary artificial intelligence engine configured to initiate changes in the persona.

13. The system of claim 12, wherein the evolutionary artificial intelligence engine is further configured to initiate changes in the persona based on user interactions with the search results.

14. The system of claim 12, wherein the evolutionary artificial intelligence engine is further configured to initiate changes in the persona based on information derived from an other persona.

15. The system of claim 12, wherein the evolutionary artificial intelligence engine is further configured to initiate changes in the persona based on information derived from a universe in which the persona interacts.

16. The system of claim 10, further comprising a persona manager module executable by the processor to select the persona from multiple available personas.

17. The system of claim 10, further comprising:

an input/output (I/O) normalizer module executable by the processor to translates received data into a format suitable for processing by the artificial intelligence engine; and

an application programming interface (API) configured to;

coordinate interactions between the system and external components; and

receive data from a network communicatively coupled to the API; and

provide the data to the I/O normalizer.

18. The system of claim 17, further comprising:

a persona manager module executable by the processor to;

receive data from the I/O normalizer; and

selectively execute one of a persona artificial intelligence engine, which applies selected personas to data to analyze the potential solutions, and an evolutionary artificial intelligence engine, which may initiate changes in selected personas.

19. The system of claim 10, wherein at least one of the selected one or more personas includes an expert persona representing decision-making characteristics of an individual previously determined to be an expert.

20. The system of claim 10, wherein at least one of the selected one or more personas includes a trusted persona representing decision-making characteristics of an individual previously determined by the user to be a trusted individual.