IP Library Granted Patent US 10,984,028
Granted Patent B2
US 10,984,028 · App. 16/202,547 · Granted Apr 20, 2021

Temporal sensitive cognitive interface

Inventors: Craig M. Trim (Ventura, CA); Gandhi Sivakumar (Bentleigh, AU); Rashida Hodge (Ossining, NY); Victor Povar (Vancouver, CA)
Assignee: International Business Machines Corporation
G06F16/3334G06F16/338G06F16/3344
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,984,028
App. No.
16/202,547
Granted
Apr 20, 2021
Kind
B2
Abstract

Embodiments provide a temporal sensitive cognitive interface that determines whether a temporal data value of a response to a user query is fixed or variable over time; determine a time lapse from a current time to a time of generation of a variable one of the temporal data values; generate a first presentation of the response to the user that includes a (first) text word modifier that indicates a quality of certainty of the temporal data value in response to determining that the time lapse does not exceed a confidence threshold that is specified for an attribute of the response temporal data value; and generate another, different (second) presentation of the response to the user that includes a different (second) text word modifier that indicates a quality of uncertainty of the temporal data value in response to determining that the time lapse exceeds the confidence threshold.

Claims (60)

1. A computer-implemented method, comprising executing on a computer processor:

determining, as a function of time, whether a temporal data value of a response to a user query is fixed or variable;

determining a time lapse from a current time to a time of generation of the temporal data value, in response to determining that the temporal data value is variable;

identifying a key driver element value within the query response that is associated to the response temporal data value;

determining a frequency of occurrence of the identified key driver value in a plurality of different iterations of determining whether the time lapse exceeds the confidence threshold for each of a plurality of different query responses, wherein the plurality of different iterations comprises a plurality of conflicting determinations as to whether the time lapse exceeds the confidence threshold;

decreasing, in proportion to the determined occurrence frequency, a weighting of the identified key driver value;

determining whether the time lapse exceeds a confidence threshold that is specified for the response temporal data value as a function of the decreased weighting of the identified key driver value;

in response to determining that the time lapse does not exceed the confidence threshold, generating a first presentation of the response to the user that comprises a first text word modifier that indicates that the temporal data value is certain; and

in response to determining that the time lapse exceeds the confidence threshold, generating a second presentation of the response to the user that comprises a second text word modifier that indicates that the temporal data value is uncertain and is different from the first word modifier.

2. The method of claim 1 , wherein the first and second text word modifiers are selected from a plurality of different lemmas of a root verb.

3. The method of claim 1 , wherein the first and second text word modifiers are selected from a plurality of different image elements that have different semantic meaning to the user.

4. The method of claim 1 , wherein the query response comprises a response plurality of text words, and wherein the determining whether the time lapse exceeds the confidence threshold that is specified for the response temporal data value comprises:

parsing documents of a data source corpus searched in generating the query response at a word level to identify an entity that is related to the response temporal data value; and

storing the identified entity in a temporal-sensitive data corpus in relation to the response temporal data value for use in satisfying additional queries by the user that are related to the response temporal data value.

5. The method of claim 1 , wherein the determining whether the time lapse exceeds the confidence threshold that is specified for the response temporal data value comprises:

determining an information decay rate for the key driver element; and

determining whether the time lapse exceeds the confidence threshold as a function of applying the information decay rate to the time lapse.

6. The method of claim 1 , further comprising:

integrating computer-readable program code into a computer system comprising a processor, a computer readable memory in circuit communication with the processor, and a computer readable storage medium in circuit communication with the processor; and

wherein the processor executes program code instructions stored on the computer-readable storage medium via the computer readable memory and thereby performs the determining whether the temporal data value of the query response is fixed or variable, the determining the time lapse, the identifying the key driver element value, the determining the frequency of occurrence of the identified key driver value, the decreasing the weighting of the identified key driver value, the determining whether the time lapse exceeds the confidence threshold, the generating a first presentation of the response to the user, and the generating the second presentation of the response to the user.

7. The method of claim 6 , wherein the computer-readable program code is provided as a service in a cloud environment.

8. A system, comprising:

a processor;

a computer readable memory in circuit communication with the processor; and

a computer readable storage medium in circuit communication with the processor;

wherein the processor executes program instructions stored on the computer-readable storage medium via the computer readable memory and thereby:

determines whether a temporal data value of a response to a user query is, as a function of time, fixed or variable;

in response to determining that the temporal data value is variable as a function of time, determines a time lapse from a current time to a time of generation of the temporal data value;

identifies a key driver element value within the query response that is associated to the response temporal data value;

determines a frequency of occurrence of the identified key driver value in a plurality of different iterations of determining whether the time lapse exceeds the confidence threshold for each of a plurality of different query responses, wherein the plurality of different iterations comprises a plurality of conflicting determinations as to whether the time lapse exceeds the confidence threshold;

decreases, in proportion to the determined occurrence frequency, a weighting of the identified key driver value;

determines whether the time lapse exceeds a confidence threshold that is specified for the response temporal data value as a function of the decreased weighting of the identified key driver value;

in response to determining that the time lapse does not exceed the confidence threshold, generates a first presentation of the response to the user that comprises a first text word modifier that indicates that the temporal data value is certain; and

in response to determining that the time lapse exceeds the confidence threshold, generates a second presentation of the response to the user that comprises a second text word modifier that indicates that the temporal data value is uncertain and is different from the first word modifier.

9. The system of claim 8 , wherein the first and second text word modifiers are selected from a plurality of different lemmas of a root verb.

10. The system of claim 8 , wherein the first and second text word modifiers are selected from a plurality of different image elements that have different semantic meaning to the user.

11. The system of claim 8 , wherein the query response comprises a response plurality of text words, and wherein the processor executes the program instructions stored on the computer-readable storage medium via the computer readable memory and thereby determines whether the time lapse exceeds the confidence threshold that is specified for the response temporal data value by:

parsing documents of a data source corpus searched in generating the query response at a word level to identify an entity that is related to the response temporal data value; and

storing the identified entity in a temporal-sensitive data corpus in relation to the response temporal data value for use in satisfying additional queries by the user that are related to the response temporal data value.

12. The system of claim 8 , wherein the processor executes the program instructions stored on the computer-readable storage medium via the computer readable memory and thereby determines whether the time lapse exceeds the confidence threshold that is specified for the response temporal data value by:

determining an information decay rate for the key driver element; and

determining whether the time lapse exceeds the confidence threshold as a function of applying the information decay rate to the time lapse.

13. A computer program product for a temporal sensitive cognitive interface, the computer program product comprising:

a computer readable storage medium having computer readable program code embodied therewith, the computer readable program code comprising instructions for execution by a processor that cause the processor to:

determine whether a temporal data value of a response to a user query is, as a function of time, fixed or variable;

in response to determining that the temporal data value is variable as a function of time, determine a time lapse from a current time to a time of generation of the temporal data value;

identify a key driver element value within the query response that is associated to the response temporal data value;

determine a frequency of occurrence of the identified key driver value in a plurality of different iterations of determining whether the time lapse exceeds the confidence threshold for each of a plurality of different query responses, wherein the plurality of different iterations comprises a plurality of conflicting determinations as to whether the time lapse exceeds the confidence threshold;

decrease, in proportion to the determined occurrence frequency, a weighting of the identified key driver value;

determine whether the time lapse exceeds a confidence threshold that is specified for the response temporal data value as a function of the decreased weighting of the identified key driver value;

in response to determining that the time lapse does not exceed the confidence threshold, generate a first presentation of the response to the user that comprises a first text word modifier that indicates that the temporal data value is certain; and

in response to determining that the time lapse exceeds the confidence threshold, generate a second presentation of the response to the user that comprises a second text word modifier that indicates that the temporal data value is uncertain and is different from the first word modifier.

14. The computer program product of claim 13 , wherein the first and second text word modifiers are selected from a plurality of different lemmas of a root verb.

15. The computer program product of claim 13 , wherein the first and second text word modifiers are selected from a plurality of different image elements that have different semantic meaning to the user.

16. The computer program product of claim 13 , wherein the query response comprises a response plurality of text words, and wherein the computer readable program code instructions for execution by the processor further cause the processor to determine whether the time lapse exceeds the confidence threshold that is specified for the response temporal data value by:

parsing documents of a data source corpus searched in generating the query response at a word level to identify an entity that is related to the response temporal data value; and

storing the identified entity in a temporal-sensitive data corpus in relation to the response temporal data value for use in satisfying additional queries by the user that are related to the response temporal data value.

17. The computer program product of claim 13 , wherein the computer readable program code instructions for execution by the processor further cause the processor to determine whether the time lapse exceeds the confidence threshold that is specified for the response temporal data value by:

determining an information decay rate for the key driver element; and

determining whether the time lapse exceeds the confidence threshold as a function of applying the information decay rate to the time lapse.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 28, 2018
From: TRIM, CRAIG M; SIVAKUMAR, GANDHI; HODGE, RASHIDA; POVAR, VICTOR
To: INTERNATIONAL BUSINESS MACHINES CORPORATION
Reel/Frame 047606/0438 →
Continuity (1)
Related Publication 20200167377A1 · May 28, 2020