BIG DATA ANALYTICS BROKERAGE

• US 20150310015A1
• Filed: 04/28/2014
• Published: 10/29/2015
• Est. Priority Date: 04/28/2014
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Alexander, Tabari H., Bivens, John A., Buzzetti, Michael C., Chan, Yuk L., Washington, Derrick L.
• Granted Patent: US 9,495,405 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2336 Pessimistic concurrency control approaches, e.g. locking or multiple versions without time stamps

Abstract

In one embodiment, a computer-implemented method includes receiving a query. A target confidence level is established for the query, the target confidence level representing a requested level of accuracy for a result of the query. At least one individual confidence level is assigned to each of a plurality of analytics engines. One or more analytics engines are queried based on the query. A group of the analytics engines are selected, by a computer processor, where the analytics engines in the selected group have query responses to the query that are deemed to be similar to one another, and where the selection of the selected group is at least partially based on the target confidence level. The query responses from the selected group of analytics engines are summarized into a final result, where the final result is an answer to the query.

Claims

1. A computer-implemented method, comprising:

receiving a query;

establishing a target confidence level for the query, the target confidence level representing a requested level of accuracy for a result of the query;

assigning at least one individual confidence level to each of a plurality of analytics engines;

querying one or more of the analytics engines based on the query;

selecting, by a computer processor, a group of the analytics engines whose query responses to the query are deemed to be similar to one another, wherein the selection of the group is at least partially based on the target confidence level; and

summarizing into a final result the query responses from the selected group of analytics engines, the final result being an answer to the query.

2. The method of claim 1, wherein selecting the group of analytics engines comprises:

grouping the analytics engines into two or more groups, each group comprising analytics engines having query results deemed similar to one another; and

selecting one of the two or more groups as the selected group;

wherein the summarized result excludes the query responses from the analytics engines outside the selected group.

3. The method of claim 1, further comprising adjusting the individual confidence levels of the analytics engines in the selected group based on contributions of those analytics engines to the final result.

4. The method of claim 1, further comprising calculating a group confidence level of the selected group of analytics engines, based at least in part on the individual confidence levels and query responses of the analytics engines in the selected group.

5. The method of claim 1, further comprising, repeatedly, modifying which analytics engines are in the selected group and recalculating the group confidence level, until a termination condition is met, wherein the termination condition is met when the group confidence level at least meets the target confidence level.

6. The method of claim 1, wherein selection of the selected group of analytics engines is further based at least partially on a target monetary cost of the query.

7. The method of claim 1, wherein assigning at least one individual confidence level to each of the plurality of analytics engines comprises assigning a first confidence level to a first analytics engines for potential queries in a first category and assigning a second confidence level to the first analytics engine for potential queries in a second category.

8. A system comprising:

an initialization unit configured to receive a query and establish a target confidence level for the query, the target confidence level representing a requested level of accuracy for a result of the query;

a confidence unit configured to assign at least one individual confidence level to each of a plurality of analytics engines; and

a query unit configured to;

query one or more analytics engines based on the query;

select a group of the analytics engines whose query responses to the query are deemed to be similar to one another, wherein the selection of the group is at least partially based on the target confidence level; and

summarize into a final result the query responses from the selected group of analytics engines, the final result being an answer to the query.

9. The system of claim 8, the query unit being further configured to:

group the analytics engines into two or more groups, each group comprising analytics engines having query results deemed similar to one another; and

select one of the two or more groups as the selected group;

wherein the summarized result excludes the query responses from the analytics engines outside the selected group.

10. The system of claim 8, the confidence unit being further configured to adjust the individual confidence levels of the analytics engines in the selected group based on contributions of those analytics engines to the final result.

11. The system of claim 8, the confidence unit being further configured to calculate a group confidence level of the selected group of analytics engines, based at least in part on the individual confidence levels and query responses of the analytics engines in the selected group.

12. The system of claim 8, the query unit being further configured to, repeatedly, modify which analytics engines are in the selected group and recalculate the group confidence level, until a termination condition is met, wherein the termination condition is met when the group confidence level at least meets the target confidence level.

13. The system of claim 8, wherein selection of the selected group of analytics engines is further based at least partially on a target monetary cost of the query.

14. The system of claim 8, the confidence unit being further configured, in assigning at least one individual confidence level to each of the plurality of analytics engines, to assign a first confidence level to a first analytics engines for potential queries in a first category and assigning a second confidence level to the first analytics engine for potential queries in a second category.

15. A computer program product comprising a computer readable storage medium having computer readable program code embodied thereon, the computer readable program code executable by a processor to perform a method comprising:

receiving a query;

establishing a target confidence level for the query, the target confidence level representing a requested level of accuracy for a result of the query;

assigning at least one individual confidence level to each of a plurality of analytics engines;

querying one or more of the analytics engines based on the query;

selecting a group of the analytics engines whose query responses to the query are deemed to be similar to one another, wherein the selection of the group is at least partially based on the target confidence level; and

summarizing into a final result the query responses from the selected group of analytics engines, the final result being an answer to the query.

16. The computer program product of claim 15, wherein selecting the group of analytics engines comprises:

grouping the analytics engines into two or more groups, each group comprising analytics engines having query results deemed similar to one another; and

selecting one of the two or more groups as the selected group;

wherein the summarized result excludes the query responses from the analytics engines outside the selected group.

17. The computer program product of claim 15, the method further comprising calculating a group confidence level of the selected group of analytics engines, based at least in part on the individual confidence levels and query responses of the analytics engines in the selected group.

18. The computer program product of claim 15, the method further comprising, repeatedly, modifying which analytics engines are in the selected group and recalculating the group confidence level, until a termination condition is met, wherein the termination condition is met when the group confidence level at least meets the target confidence level.

19. The computer program product of claim 15, wherein selection of the selected group of analytics engines is further based at least partially on a target monetary cost of the query.

20. The computer program product of claim 15, wherein assigning at least one individual confidence level to each of the plurality of analytics engines comprises assigning a first confidence level to a first analytics engines for potential queries in a first category and assigning a second confidence level to the first analytics engine for potential queries in a second category.