# US 11,308,948 B2 · Intelligent interaction processing method and apparatus, device and computer storage medium

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

- Number: US 11,308,948 B2
- Document ID: US11308948B2
- Status: Granted
- Kind: B2
- Application number: 16/174,087
- Filed: 2018-10-29
- Published: 2019-05-16
- Granted: 2022-04-19
- Earliest claimed: 2017-11-16
- Assignees: Baidu Online Network Technology (Beijing) Co., Ltd.; Shanghai Xiaodu Technology Co. Ltd.
- Applicants: Baidu Online Network Technology (Beijing) Co., Ltd.
- Inventors: Mengmeng Zhang; Gang Zhang; Li Wan; Jia Liu; Xiangtao Jiang; Ran Xu
- Web page: https://patentgrep.com/patent/US11308948B2

## CPC

- G10L 15/22: Procedures used during a speech recognition process, e.g. man-machine dialogue
- G06F 3/167: {Audio in a user interface, e.g. using voice commands for navigating, audio feedback}
- G06F 16/90332: {Natural language query formulation or dialogue systems}
- G06F 40/216: using statistical methods
- G06F 40/274: Converting codes to words; Guess-ahead of partial word inputs
- G06F 40/284: Lexical analysis, e.g. tokenisation or collocates
- G06F 40/30: Semantic analysis
- G06F 40/35: Discourse or dialogue representation
- G06N 5/022: Knowledge engineering; Knowledge acquisition
- G06N 20/00: Machine learning
- G10L 15/1815: {Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning}
- G10L 15/1822: {Parsing for meaning understanding}
- G10L 2015/225: {Feedback of the input speech}
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

The present disclosure provides an intelligent interaction processing method and apparatus, a device and a computer storage medium. The method comprises: performing intention recognition for a preceding feedback item already returned to the user; continuing to return a subsequent feedback item to the user based on the intention of the preceding feedback item. According to the present disclosure, it is possible to guess the user's subsequent intention based on the preceding feedback item, and continue to return the desired subsequent feedback item to the user without the user's operations, so that the present disclosure is more intelligentized and richer and simplifies the user's operations.

## Claims

### Claim 1

A computer-implemented intelligent interaction processing method, wherein the method comprises:
returning a preceding feedback item to a user based on a query;
performing intention recognition for the preceding feedback item already returned to the user;
continuing to return a subsequent feedback item to the user based on the intention of the preceding feedback item, the subsequent feedback item being independent of the query based on which the preceding feedback item returned to the user is generated, and the subsequent feedback item being independent of a follow-up or subsequent query, wherein the subsequent feedback item is returned without receiving a subsequent query,
wherein the returning a subsequent feedback item to the user based on the intention of the preceding feedback item comprises:
determining a type of the subsequent feedback item corresponding to the intention;
obtaining an entity of the subsequent feedback item;
using the entity of the subsequent feedback item and the type of the subsequent feedback item to determine the subsequent feedback item returned to the user,
wherein the determining a type of the subsequent feedback item corresponding to the intention comprises:
determining the type of the subsequent feedback item corresponding to the recognized intention, according to the preset correspondence relationship between the intentions and the type of the subsequent feedback item, the subsequent feedback item comprising text, audio, video, image, link, and a control event of an application or user equipment.

### Claim 2 (depends on 1)

The method according to claim 1, wherein the performing intention recognition for a preceding feedback item already returned to the user comprises:
extracting a keyword from the preceding feedback item, and determining an intention of the preceding feedback item according to the keyword; or,
performing semantic analysis for the preceding feedback item, and determining the intention of the preceding feedback item; or
matching the preceding feedback item with a preset template, and determining an intention corresponding to the matched template as the intention of the preceding feedback item; or
using a machine learning model obtained by pre-training to perform intention analysis for the preceding feedback item to obtain an intention of the preceding feedback item.

### Claim 3 (depends on 1)

The method according to claim 1, wherein the correspondence relationship further comprises: a type of the preceding feedback item;
the method further comprises: determining the type of the preceding feedback item already returned to the user;
the determining a type of a subsequent feedback item corresponding to the recognized intention, according to a correspondence relationship between the preset intention and the type of the subsequent feedback item comprises:
determining the type of the corresponding subsequent feedback item according to the correspondence relationship between the type of the preceding feedback item and an intention query of the preceding feedback item.

### Claim 4 (depends on 1)

The method according to claim 1, wherein before determining type of the subsequent feedback item corresponding to the intention, the method further comprises:
judging whether an intention confidence of the preceding feedback item satisfies a preset confidence requirement, and if yes, continuing to execute of the determination of the type of the subsequent feedback item corresponding to the intention.

### Claim 5 (depends on 1)

The method according to claim 1, wherein the type of the subsequent feedback item corresponding to the recognized intention is determined further based on the user's environment information.

### Claim 6 (depends on 5)

The method according to claim 5, wherein the user's environment information comprises at least one of system time, user equipment type and user equipment action.

### Claim 7 (depends on 1)

The method according to claim 1, wherein the obtaining the entity of the subsequent feedback item comprises:
regarding an entity extracted from the preceding feedback item as an entity of the subsequent feedback item; or
obtaining an entity from user data or user environment information as an entity of the subsequent feedback item.

### Claim 8 (depends on 1)

The method according to claim 1, wherein the using the entity of the subsequent feedback item and the type of the subsequent feedback item to determine the subsequent feedback item comprises:
using the entity of the subsequent feedback item and the type of the subsequent feedback item to configure a search item;
obtaining a vertical search result corresponding to the search item as the subsequent feedback item.

### Claim 9 (depends on 8)

The method according to claim 8, wherein the using the entity of the subsequent feedback item and the type of the subsequent feedback item to configure a search item comprises:
determining a template corresponding to the type of the subsequent feedback item;
filling the entity of the subsequent feedback item into the determined template to obtain the search item.

### Claim 10 (depends on 1)

The method according to claim 1, wherein the using the entity of the subsequent feedback item and the type of the subsequent feedback item to determine the subsequent feedback item comprises:
using the entity of the subsequent feedback item and the type of the subsequent feedback item to generate a control instruction;
sending the control instruction to an application or user equipment corresponding to the type of the subsequent feedback item.

### Claim 11 (depends on 1)

The method according to claim 1, wherein the returning the subsequent feedback item to the user comprises:
directly returning the determined subsequent feedback item to the user; or
determining whether to return the determined subsequent feedback item to the user based on the user's feedback.

### Claim 12 (depends on 1)

The method according to claim 1, wherein transition wording is returned between the preceding feedback item already returned to the user and the subsequent feedback item returned to the user;
the transition wording comprises a general-purpose word or sentence, blank, symbol, shadow or audio.

### Claim 13 (depends on 1)

The method according to claim 1, wherein the subsequent feedback item comprises text, audio, video, image, link, and a control event of an application or user equipment.

### Claim 14

A device, wherein the device comprises:
one or more processors;
a storage for storing one or more programs,
when said one or more programs are executed by said one or more processors, said one or more processors are enabled to implement an intelligent interaction processing method, wherein the method comprises:
returning a preceding feedback item to a user based on a query;
performing intention recognition for the preceding feedback item already returned to the user;
continuing to return a subsequent feedback item to the user based on the intention of the preceding feedback item, the subsequent feedback item being independent of the query based on which the preceding feedback item returned to the user is generated, and the subsequent feedback item being independent of a follow-up or subsequent query, wherein the subsequent feedback item is returned without receiving a subsequent query,
wherein the returning a subsequent feedback item to the user based on the intention of the preceding feedback item comprises:
determining a type of the subsequent feedback item corresponding to the intention;
obtaining an entity of the subsequent feedback item;
using the entity of the subsequent feedback item and the type of the subsequent feedback item to determine the subsequent feedback item returned to the user,
wherein the determining a type of the subsequent feedback item corresponding to the intention comprises:
determining the type of the subsequent feedback item corresponding to the recognized intention, according to the preset correspondence relationship between the intentions and the type of the subsequent feedback item, the subsequent feedback item comprising text, audio, video, image, link, and a control event of an application or user equipment.

### Claim 15 (depends on 14)

The device according to claim 14, wherein the performing intention recognition for a preceding feedback item already returned to the user comprises:
extracting a keyword from the preceding feedback item, and determining an intention of the preceding feedback item according to the keyword; or,
performing semantic analysis for the preceding feedback item, and determining the intention of the preceding feedback item; or
matching the preceding feedback item with a preset template, and determining an intention corresponding to the matched template as the intention of the preceding feedback item; or
using a machine learning model obtained by pre-training to perform intention analysis for the preceding feedback item to obtain an intention of the preceding feedback item.

### Claim 16

A non-transitory storage medium containing computer executable instructions, wherein the computer executable instructions, when executed by a computer processor, implement an intelligent interaction processing method, wherein the method comprises:
returning a preceding feedback item to a user based on a query;
performing intention recognition for the preceding feedback item already returned to the user;
continuing to return a subsequent feedback item to the user based on the intention of the preceding feedback item, the subsequent feedback item being independent of the query based on which the preceding feedback item returned to the user is generated, and the subsequent feedback item being independent of a follow-up or subsequent query, wherein the subsequent feedback item is returned without receiving a subsequent query,
wherein the returning a subsequent feedback item to the user based on the intention of the preceding feedback item comprises:
determining a type of the subsequent feedback item corresponding to the intention;
obtaining an entity of the subsequent feedback item;
using the entity of the subsequent feedback item and the type of the subsequent feedback item to determine the subsequent feedback item returned to the user,
wherein the determining a type of the subsequent feedback item corresponding to the intention comprises:
determining the type of the subsequent feedback item corresponding to the recognized intention, according to the preset correspondence relationship between the intentions and the type of the subsequent feedback item, the subsequent feedback item comprising text, audio, video, image, link, and a control event of an application or user equipment.