# US 9,519,686 B2 · Confidence ranking of answers based on temporal semantics

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 9,519,686 B2
- Document ID: US9519686B2
- Status: Granted
- Kind: B2
- Application number: 14/248,762
- Filed: 2014-04-09
- Published: 2015-10-15
- Granted: 2016-12-13
- Earliest claimed: 2014-04-09
- Assignees: International Business Machines Corporation
- Applicants: International Business Machines Corporation
- Inventors: John P. Bufe, III; Donna K. Byron; Alexander Pikovsky; Timothy P. Winkler
- Web page: https://patentgrep.com/patent/US9519686B2

## CPC

- G06F 16/24578: {using ranking}
- G06F 16/243: {Natural language query formulation}
- G06F 16/24575: {using context}
- G06F 16/2477: {Temporal data queries}
- G06F 16/248: Presentation of query results
- G06F 16/9535: Search customisation based on user profiles and personalisation
- G06N 5/02: Knowledge representation; Symbolic representation
- G06N 20/00: Machine learning
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A mechanism is provided, in a data processing system comprising a processor and a memory configured to implement a question and answer system (QA), for providing confidence rankings based on temporal semantics. Responsive to receiving an input question, a set of candidate answers is identified from a knowledge domain based on a correlation between an identified one or more predicates and an identified one or more arguments to the knowledge domain. A confidence score is associated with each of the candidate answers and each confidence score associated with each candidate answer is refined based on a set of temporal characteristics identified in the input question. A set of temporally refined candidate answers is then provided to the user.

## Claims

### Claim 1

A computer program product comprising a computer readable storage medium having a computer readable program stored therein, wherein the computer readable program, when executed on a computing device, causes the computing device to:
responsive to receiving an input question, identify a set of candidate answers from a knowledge domain based on a correlation between an identified one or more predicates and an identified one or more arguments associated with the one or more predicates from the input question;
associate an initial confidence score with each candidate answer of the set of candidate answers;
refine each initial confidence score associated with each candidate answer based on a set of temporal characteristics identified in the input question, wherein each confidence score associated with each candidate answer is refined based on the set of temporal characteristics using a reference time of the input question and a respective reference time associated with the candidate answer thereby forming a temporally refined confidence score associated with the candidate answer, wherein the temporally refined confidence score associated with the candidate answer is identified by:
generating a distance value in terms of years for the respective reference time associated with the candidate answer and the reference time of the input question;
using the distance value, determining a multiplier value with which to weight the confidence score associated with the candidate answer using multiplier function: multiplier value=1/(2*distance value+0.5);
identifying a sentiment value of the candidate answer to weight the determined multiplier value either positively or negatively; and
providing a final weight for the temporally refined confidence score associated with the candidate answer using distance function: final weight=initial confidence score*multiplier value*sentiment value; and
provide a set of candidate answers with the temporally refined confidence scores to the user.

### Claim 2 (depends on 1)

The computer program product of claim 1, wherein the reference time of the input question is identified from the set of temporal characteristics.

### Claim 3 (depends on 1)

The computer program product of claim 1, wherein in the reference time of the input question is a current time if no temporal characteristics are identifiable in the input question.

### Claim 4

An apparatus comprising:
a processor; and
a memory coupled to the processor, wherein the memory comprises instructions which, when executed by the processor, cause the processor to:
responsive to receiving an input question, identify a set of candidate answers from a knowledge domain based on a correlation between an identified one or more predicates and an identified one or more arguments associated with the one or more predicates from the input question;
associate an initial confidence score with each candidate answer of the set of candidate answers;
refine each initial confidence score associated with each candidate answer based on a set of temporal characteristics identified in the input question, wherein each confidence score associated with each candidate answer is refined based on the set of temporal characteristics using a reference time of the input question and a respective reference time associated with the candidate answer thereby forming a temporally refined confidence score associated with the candidate answer, wherein the temporally refined confidence score associated with the candidate answer is identified by:
generating a distance value in terms of years for the respective reference time associated with the candidate answer and the reference time of the input question;
using the distance value, determining a multiplier value with which to weight the confidence score associated with the candidate answer using multiplier function: multiplier value=1/(2*distance value+0.5);
identifying a sentiment value of the candidate answer to weight the determined multiplier value either positively or negatively; and
providing a final weight for the temporally refined confidence score associated with the candidate answer using distance function: final weight=initial confidence score*multiplier value*sentiment value; and
provide a set of candidate answers with the temporally refined confidence scores to the user.

### Claim 5 (depends on 4)

The apparatus of claim 4, wherein the reference time of the input question is identified from the set of temporal characteristics.

### Claim 6 (depends on 4)

The apparatus of claim 4, wherein in the reference time of the input question is a current time if no temporal characteristics are identifiable in the input question.