# US 2021/0241756 A1 · Disambiguation of generic commands for controlling objects

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

- Number: US 2021/0241756 A1
- Document ID: US20210241756A1
- Status: Published application
- Kind: A1
- Application number: 16/779,733
- Filed: 2020-02-03
- Published: 2021-08-05
- Earliest claimed: 2020-02-03
- Assignees: International Business Machines Corporation
- Applicants: International Business Machines Corporation
- Inventors: Lisa Seacat DeLuca; Zachary A. Silverstein; Kelley Anders; Jeremy R. Fox
- Web page: https://patentgrep.com/patent/US20210241756A1

## CPC

- G06F 3/167: {Audio in a user interface, e.g. using voice commands for navigating, audio feedback}
- G10L 15/1815: {Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning}
- G10L 15/22: Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L 2015/223: {Execution procedure of a spoken command}
- G10L 2015/228: {of application context}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

A computer system performs an action on an object identified from a command. A command is analyzed to perform an action on a target object, wherein the command includes a term for the target object that refers to a plurality of different candidate objects. The target object is identified from the plurality of different candidate objects based on historical associations of the term with specific ones of the candidate objects, recent interactions with the different candidate objects, and a state of the candidate objects provided by network devices. The action is performed on the identified target object. Embodiments of the present invention further include a method and program product for performing an action on an object identified from a command in substantially the same manner described above.

## Claims

### Claim 1

A computer-implemented method of performing an action on an object identified from a command, the computer-implemented method comprising:
analyzing, via a processor, a command to perform an action on a target object, wherein the command includes a term for the target object that refers to a plurality of different candidate objects;
identifying, via the processor, the target object from the plurality of different candidate objects based on historical associations of the term with specific ones of the candidate objects, recent interactions with the different candidate objects, and a state of the candidate objects provided by network devices; and
performing, via the processor, the action on the identified target object.

### Claim 2 (depends on 1)

The method of claim 1, wherein the network device includes an Internet of Things device and the target object includes a physical object.

### Claim 3 (depends on 1)

The method of claim 1, wherein the target object incudes a virtual object.

### Claim 4 (depends on 1)

The method of claim 1, wherein the command includes a voice command, and the processor implements a virtual agent to perform the action on the target object.

### Claim 5 (depends on 4)

The method of claim 4, wherein the plurality of different candidate objects referred to by the term are determined from a decision ontology for the virtual agent.

### Claim 6 (depends on 1)

The method of claim 1, further comprising:
determining, via the processor, a score for each of the plurality of different candidate objects based on weights applied to the historical association of the term with that candidate object, recent interactions with that candidate object, and the state of that candidate object provided by the network devices; and
identifying, via the processor, the target object from the plurality of candidate objects based on the score for each of the plurality of different candidate objects.

### Claim 7 (depends on 1)

The method of claim 1, wherein identifying the target object further comprises:
identifying the target object based on crowdsourcing associations between the term and the plurality of candidate objects.

### Claim 8 (depends on 1)

The method claim 1, further comprising:
identifying a virtual agent from among a plurality of different virtual agents to process the command based on the target object and action.

### Claim 9 (depends on 1)

The method of claim 1, further comprising:
storing information for a specific object from a prior command in a decaying storage object, wherein the decaying storage object is used to identify the target object, and wherein a time decay for the decaying storage object is based on one or more from a group of:
a recent interaction with a network device referencing the specific object;
a historical frequency of introduction of other similar named objects; and
a contextual component of network devices derived from a state of the specific object.

### Claim 10 (depends on 1)

The method of claim 1, wherein the historical associations include a candidate object with a greatest quantity of commands, and the state of the candidate objects include presence of a conflict between an operational state of the candidate objects and the action to be performed on the target object.

### Claim 11 (depends on 1)

The method of claim 1, wherein the action to be performed includes parental controls for the target object.

### Claim 12

A computer system for performing an action on an object identified from a command, the computer system comprising:
one or more computer processors;
one or more computer readable storage media;
program instructions stored on the one or more computer readable storage media for execution by at least one of the one or more computer processors, the program instructions comprising instructions to:
analyze a command to perform an action on a target object, wherein the command includes a term for the target object that refers to a plurality of different candidate objects;
identify the target object from the plurality of different candidate objects based on historical associations of the term with specific ones of the candidate objects, recent interactions with the different candidate objects, and a state of the candidate objects provided by network devices; and
perform the action on the identified target object.

### Claim 13 (depends on 12)

The computer system of claim 12, further comprising instructions to:
determine a score for each of the plurality of different candidate objects based on weights applied to the historical association of the term with that candidate object, recent interactions with that candidate object, and the state of that candidate object provided by the network devices; and
identify the target object from the plurality of candidate objects based on the score for each of the plurality of different candidate objects.

### Claim 14 (depends on 12)

The computer system of claim 12, further comprising instructions to:
store information for a specific object from a prior command in a decaying storage object, wherein the decaying storage object is used to identify the target object, and wherein a time decay for the decaying storage object is based on one or more from a group of:
a recent interaction with a network device referencing the specific object;
a historical frequency of introduction of other similar named objects; and
a contextual component of network devices derived from a state of the specific object.

### Claim 15

A computer program product for performing an action on an object identified from a command, the computer program product comprising one or more computer readable storage media collectively having program instructions embodied therewith, the program instructions executable by a computer to cause the computer to:
analyze a command to perform an action on a target object, wherein the command includes a term for the target object that refers to a plurality of different candidate objects;
identify the target object from the plurality of different candidate objects based on historical associations of the term with specific ones of the candidate objects, recent interactions with the different candidate objects, and a state of the candidate objects provided by network devices; and
perform the action on the identified target object.

### Claim 16 (depends on 15)

The computer program product of claim 15, wherein the command includes a voice command, and the processor implements a virtual agent to perform the action on the target object.

### Claim 17 (depends on 16)

The computer program product of claim 16, wherein the plurality of different candidate objects referred to by the term are determined from a decision ontology for the virtual agent.

### Claim 18 (depends on 15)

The computer program product of claim 15, wherein the program instructions further cause the computer to:
determine a score for each of the plurality of different candidate objects based on weights applied to the historical association of the term with that candidate object, recent interactions with that candidate object, and the state of that candidate object provided by the network devices; and
identify the target object from the plurality of candidate objects based on the score for each of the plurality of different candidate objects.

### Claim 19 (depends on 15)

The computer program product of claim 15, wherein the program instructions further cause the computer to:
store information for a specific object from a prior command in a decaying storage object, wherein the decaying storage object is used to identify the target object, and wherein a time decay for the decaying storage object is based on one or more from a group of:
a recent interaction with a network device referencing the specific object;
a historical frequency of introduction of other similar named objects; and
a contextual component of network devices derived from a state of the specific object.

### Claim 20 (depends on 15)

The computer program product of claim 15, wherein the historical associations include a candidate object with a greatest quantity of commands, and the state of the candidate objects include presence of a conflict between an operational state of the candidate objects and the action to be performed on the target object.
