Insight and algorithmic clustering for automated synthesis

• US 9,607,023 B1
• Filed: 05/06/2016
• Issued: 03/28/2017
• Est. Priority Date: 07/20/2012
• Status: Active Grant

• Current Assignee: OOL LLC
• Original Assignee: OOL LLC
• Inventors: Swamy, Gitanjali
• Primary Examiner(s): Hoang, Son T
• Application Number:
  US15/148,877
• Time in Patent Office: 326 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 16/217 Database tuning G06F16/2282 takes precedence; database performance monitoring G06F11/3409
• CPC Class Codes: G06F 16/2428 Query predicate definition using graphical user interfaces, including menus and forms G06F16/2423 takes precedence
• CPC Class Codes: G06F 16/2465 Query processing support for facilitating data mining operations in structured databases
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/3326 using relevance feedback from the user, e.g. relevance feedback on documents, documents sets, document terms or passages
• CPC Class Codes: G06F 16/3328 using graphical result space presentation or visualisation
• CPC Class Codes: G06F 16/355 Class or cluster creation or modification
• CPC Class Codes: G06F 18/24317 Piecewise classification, i.e. whereby each classification requires several discriminant rules
• CPC Class Codes: G16B 40/00 ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding

Abstract

A decision support system and method, which receives user inputs comprising: at least one user criterion, and at least one user input tuning parameter representing user tradeoff preferences for producing an output; and selectively produces an output of tagged data from a clustered database in dependence on the at least one user criterion, the at least one user input tuning parameter, and a distance function; receives at least one reference-user input parameter representing the at least one reference-user'"'"'s analysis of the tagged data and the corresponding user inputs, to adapt the distance function in accordance with the reference-user inputs as a feedback signal; and clusters the database in dependence on at least the distance function, wherein the reference-user acts to optimize the distance function based on the user inputs and the output, and on at least one reference-user inference.

Claims

1. A decision support system, comprising:

a user input port configured to receive user inputs, from a user, comprising at least one user criterion and at least one user input tuning parameter comprising a dimensionless quantitative variable representing a tradeoff preference of the user between at least two competing criteria;

an information repository configured to store a set of tagged data comprising a plurality of hidden dimensions;

a reference-user input port configured to receive, from at least one reference-user, at least one reference-user input parameter representing at least one analysis from the reference-user of at least a portion of the set of tagged data with respect to the received user inputs, wherein the at least one reference-user being distinct from the user; and

at least one processor, configured to;

define an adaptively optimized clustering distance function adapted independence on at least the at least one user criterion, the at least one user input tuning parameter, and the at least one reference-user input parameter; and

produce a clustered output of the at least portion of the tagged data in dependence on the at least one user criterion, the at least one user input tuning parameter, and the adaptively optimized clustering distance function, wherein

the clustered output having a number of dimensions less than a number of the plurality of hidden dimensions, and

the at least one user input tuning parameter impacts a cluster assignment of members of the set of tagged data according to the plurality of hidden dimensions by altering an application of the at least two competing criteria with respect to the adaptively optimized clustering distance function.

2. The system according to claim 1, wherein at least the adaptively optimized clustering distance function is applied to cluster the set of tagged data before the at least one reference-user input parameter is received.

3. The system according to claim 1, wherein the adaptively optimized clustering distance function is further adaptive to tagged data received after the at least one reference-user input parameter is received.

4. The system according to claim 1, wherein the set of tagged data comprises a valuation or rating.

5. The system according to claim 1, wherein the reference-user'"'"'s analysis represents at least one of a valuation and a validation.

6. The system according to claim 1, wherein the at least one reference-user is selected from a group of potential reference-users, based on at least a past performance of a respective potential reference-user.

7. The system according to claim 1, wherein the at least one user input tuning parameter balances completeness and correctness of the set of tagged data in the clustered output.

8. A decision support method, comprising:

receiving user inputs, from a user, comprising at least one user criterion, and at least one user input tuning parameter comprising a dimensionless quantitative variable representing tradeoff preferences of the user between at least two competing criteria;

receiving at least one a reference-user input parameter, from at least one reference-user, representing at least one analysis from the reference-user of at least a portion of a set of tagged data comprising a plurality of hidden dimensions, with respect to the received user inputs, wherein the at least one reference-user being distinct from the user;

defining an adaptively optimized clustering distance function adapted in dependence on at least the at least one user criterion, the at least one user input tuning parameter, and the at least one reference-user input parameter; and

producing, with at least one automated processor, a clustered output of the at least portion of the set of tagged data in dependence on the at least one user criterion, the at least one user input tuning parameter, and the adaptively optimized clustering distance function, wherein

the clustered output having a number of dimensions less than a number of the plurality of hidden dimensions of the set of tagged data, and

the at least one user input tuning parameter impacts a cluster assignment of members of the set of tagged data according to the plurality of hidden dimensions by altering an application of the at least two competing criteria with respect to the adaptively optimized clustering distance function.

9. The method according to claim 8, further comprising employing the at least one reference-user input parameter representing the at least one reference-user'"'"'s analysis of the at least portion of the set of tagged data with respect to the receive user inputs as feedback to adapt the optimized clustering distance function.

10. The method according to claim 8, wherein at least the adaptively optimized clustering distance function is applied to cluster the set of tagged data before receiving the at least one reference-user input parameter.

11. The method according to claim 8, further comprising adapting the adaptively optimized clustering distance function based on at least tagged data received after the at least one reference-user input parameter is received.

12. The method according to claim 8, wherein the set of tagged data comprises a valuation or rating;

the reference-user'"'"'s analysis represents at least one of a user valuation and a user validation; and

the at least one user input tuning parameter balances completeness and correctness of the tagged data in the clustered output.

13. The method according to claim 8, wherein the at least one reference-user is selected from a group of potential reference-users, based on at least a past performance of a respective potential reference-user.

14. A decision support method, comprising:

receiving a set of inputs from at least one user, the set comprising at least one user criterion, and at least one user input tuning parameter comprising a dimensionless quantitative variable representing tradeoff preferences of the at least one user between at least two independent criteria;

selectively producing, by at least one automated processor, a clustered output of a set of tagged data comprising a plurality of hidden dimensions, in dependence on the set of inputs from the at least one user, and an adaptive clustering distance function adapted to define clusters of the set of tagged data having a number of dimensions less than a number of the plurality of hidden dimensions;

receiving at least one reference-user input tuning parameter, from at least one reference-user, representing at least one analysis of the reference-user of the set of tagged data and the set of inputs from the at least one user; and

adapting the adaptive clustering distance function in accordance with the reference-user input tuning parameter as a feedback signal, to optimize the adaptive clustering distance function based on the set of user inputs and the clustered output,

wherein the at least one user input tuning parameter impacts a cluster assignment of members of the set of tagged data according to the plurality of hidden dimensions by altering an application of the at least two independent criteria with respect to the adaptive clustering distance function.

15. The method according to claim 14, wherein the at least one reference-user is automatically selected from a pool of users based on at least an accuracy of selection according to an objective criterion.

16. The method according to claim 14, further comprising clustering the set of tagged data using at least the adaptive clustering distance function before receiving the at least one reference-user input tuning parameter;

and

adapting the adaptive clustering distance function based on at least tagged data received after the at least one reference-user input parameter is received.

17. The method according to claim 14, wherein the set of tagged data comprises a valuation or rating;

the reference-user'"'"'s analysis represents at least one of a valuation and a validation; and

the at least one user input tuning parameter balances completeness and correctness of the tagged data in the clustered output.

18. The method according to claim 14, wherein the at least one reference-user is selected from a group of potential reference-users, based on at least a past performance of a respective potential reference-user.

19. The method according to claim 14, wherein the receiving the set of inputs comprises:

receiving a semantic user input comprising an indication of interest in information;

determining a context of the at least one user distinctly from the semantic user input comprising the indication of interest in information;

monitoring a user interaction with the clustered output; and

modifying the optimized clustering distance function in dependence on at least the monitored user interaction.

20. The method according to claim 19, further comprising selecting at least one commercial advertisement extrinsic to the set of tagged data from the clustered database for presentation to the at least one user, in dependence on at least:

at least one of the semantic user input and the output of clustered data, and the determined context.