IP Library › Granted Patent US 11,348,015
Granted Patent B2
US 11,348,015 · App. 15/642,871 · Granted May 31, 2022

Method and apparatus for analyzing intention based on artificial intelligence

Inventors: Zhen Guo (Beijing, CN); Wenquan Wu (Beijing, CN); Zhanyi Liu (Beijing, CN)
Assignee: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY CO., LTD.
G06N5/022G06F16/3331G06F40/211G06F40/284G06F40/30G06N3/004G06N5/025
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 11,348,015
App. No.
15/642,871
Granted
May 31, 2022
Kind
B2
Abstract

The present disclosure provides a method and an apparatus for analyzing an intention based on artificial intelligence. The method includes: receiving a query; acquiring a keyword of an intention of the query according to a preset strategy; acquiring a qualifier of the intention of the query according to the keyword based on a syntax tree; and determining the intention of the query according to the keyword and the qualifier.

Claims (84)

1. A method for analyzing an intention of a user of a terminal comprising a processor and memory based on artificial intelligence, comprising:

receiving a query from the user of the terminal;

acquiring a keyword of an intention of the query according to a preset strategy;

acquiring a qualifier of the intention of the query according to the keyword based on a syntax tree; and

determining the intention of the query according to the keyword and the qualifier,

wherein acquiring the keyword of the intention of the query according to a preset strategy comprises:

acquiring the keyword of the intention of the query according to a template strategy,

wherein, acquiring the keyword of the intention of the query according to the template strategy comprises:

analyzing the query and acquiring an interrogative segment or an interrogative sentence pattern of the query;

matching the interrogative segment or the interrogative sentence pattern to a pre-labeled template set; and

acquiring a label corresponding to a successfully matched template, wherein the label is regarded as the keyword of the intention of the query,

wherein the interrogative segment represents nodes included in a path from an interrogative term to a root node of the syntax tree, and the interrogative sentence pattern represents a sentence pattern generated by performing a type generalization on proper nouns and intention terms in an interrogative sentence according to an intention term list.

2. The method according to claim 1 , wherein acquiring the keyword of the intention of the query according to a preset strategy further comprises:

acquiring the keyword of the intention of the query according to a rule strategy; or

acquiring the keyword of the intention of the query according to a model strategy.

3. The method according to claim 2 , wherein acquiring the keyword of the intention of the query according to a model strategy comprises:

analyzing the query and acquiring a query segment of the query;

generating a plurality of candidate intentions according to the query segment; and

screening the keyword of the intention of the query from the plurality of candidate intentions based on a pre-established intention recognition model.

4. The method according to claim 2 , wherein acquiring the keyword of the intention of the query according to the rule strategy comprises:

acquiring a syntactic structure segment of the query and a knowledge base required by the syntactic structure segment;

determining a judgment rule corresponding to the query according to the syntactic structure segment and the knowledge base, wherein the judgment rule corresponding to the query is configured to acquire the key word of the intention of the query; and

acquiring the keyword of the intention of the query based on the judgment rule.

5. The method according to claim 1 , wherein acquiring a qualifier of the intention of the query according to the keyword based on a syntax tree comprises:

acquiring a syntactic structure and a syntactic relation based on the syntax tree; and

determining the qualifier of the intention of the query according to the syntactic structure and the syntactic relation.

6. The method according to claim 1 , further comprising:

searching according to the intention and acquiring a search result corresponding to the intention after the intention is determined.

7. An apparatus for analyzing an intention based on artificial intelligence, comprising:

one or more processors, and

a memory having instructions executable by the processor stored thereon, wherein the one or more processors are configured to perform acts of:

receiving a query from a user of the apparatus;

acquiring a keyword of an intention of the query according to a preset strategy;

acquiring a qualifier of the intention of the query according to the keyword based on a syntax tree; and

determining the intention of the query according to the keyword and the qualifier,

wherein acquiring the keyword of the intention of the query according to a preset strategy comprises:

acquiring the keyword of the intention of the query according to a template strategy,

wherein acquiring the keyword of the intention of the query according to the template strategy comprises:

analyzing the query and acquiring an interrogative segment or an interrogative sentence pattern of the query;

matching the interrogative segment or the interrogative sentence pattern to a pre-labeled template set; and

acquiring a label corresponding to a successfully matched template, wherein the label is regarded as the keyword of the intention of the query,

wherein the interrogative segment represents nodes included in a path from an interrogative term to a root node of the syntax tree, and the interrogative sentence pattern represents a sentence pattern generated by performing a type generalization on proper nouns and intention terms in an interrogative sentence according to an intention term list.

8. The apparatus according to claim 7 , wherein acquiring the keyword of the intention of the query according to a preset strategy comprises

acquiring the keyword of the intention of the query according to a rule strategy; or

acquiring the keyword of the intention of the query according to a model strategy.

9. The apparatus according to claim 8 , wherein acquiring the keyword of the intention of the query according to a model strategy comprises:

analyzing the query and acquire a query segment of the query;

generating a plurality of candidate intentions according to the query segment; and

screening the keyword of the intention of the query from the plurality of candidate intentions based on a pre-established intention recognition model.

10. The apparatus according to claim 8 , wherein acquiring the keyword of the intention of the query according to the rule strategy comprises:

acquiring a syntactic structure segment of the query and a knowledge base required by the syntactic structure segment;

determining a judgment rule corresponding to the query according to the syntactic structure segment and the knowledge base, wherein the judgment rule corresponding to the query is configured to acquire the key word of the intention of the query; and

acquiring the keyword of the intention of the query based on the judgment rule.

11. The apparatus according to claim 7 , wherein acquiring a qualifier of the intention of the query according to the keyword based on a syntax tree comprises:

acquiring a syntactic structure and a syntactic relation based on the syntax tree; and

determining the qualifier of the intention of the query according to the syntactic structure and the syntactic relation.

12. The apparatus according to claim 7 , wherein the one or more processors further configured to perform:

searching according to the intention and to acquire a search result corresponding to the intention after the intention is determined.

13. A non-transitory computer readable storage medium comprising instructions, wherein when the instructions are executed by a processor of a device to perform acts of:

receiving a query from a user of the device;

acquiring a keyword of an intention of the query according to a preset strategy;

acquiring a qualifier of the intention of the query according to the keyword based on a syntax tree; and

determining the intention of the query according to the keyword and the qualifier,

wherein acquiring the keyword of the intention of the query according to a preset strategy comprises:

acquiring the keyword of the intention of the query according to a template strategy,

wherein acquiring the keyword of the intention of the query according to the template strategy comprises:

analyzing the query and acquiring an interrogative segment or an interrogative sentence pattern of the query;

matching the interrogative segment or the interrogative sentence pattern to a pre-labeled template set; and

acquiring a label corresponding to a successfully matched template, wherein the label is regarded as the keyword of the intention of the query,

wherein the interrogative segment represents nodes included in a path from an interrogative term to a root node of the syntax tree, and the interrogative sentence pattern represents a sentence pattern generated by performing a type generalization on proper nouns and intention terms in an interrogative sentence according to an intention term list.

14. The non-transitory computer readable storage medium according to claim 13 , wherein acquiring the keyword of the intention of the query according to a preset strategy comprises:

acquiring the keyword of the intention of the query according to a rule strategy; or

acquiring the keyword of the intention of the query according to a model strategy.

15. The non-transitory computer readable storage medium according to claim 14 , wherein acquiring the keyword of the intention of the query according to a model strategy comprises:

analyzing the query and acquiring a query segment of the query;

generating a plurality of candidate intentions according to the query segment; and

screening the keyword of the intention of the query from the plurality of candidate intentions based on a pre-established intention recognition model.

16. The non-transitory computer readable storage medium according to claim 14 , wherein acquiring the keyword of the intention of the query according to the rule strategy comprises:

acquiring a syntactic structure segment of the query and a knowledge base required by the syntactic structure segment;

determining a judgment rule corresponding to the query according to the syntactic structure segment and the knowledge base, wherein the judgment rule corresponding to the query is configured to acquire the key word of the intention of the query; and

acquiring the keyword of the intention of the query based on the judgment rule.

17. The non-transitory computer readable storage medium according to claim 13 , wherein acquiring a qualifier of the intention of the query according to the keyword based on a syntax tree comprises:

acquiring a syntactic structure and a syntactic relation based on the syntax tree; and

determining the qualifier of the intention of the query according to the syntactic structure and the syntactic relation.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 6, 2017
From: GUO, ZHEN; WU, WENQUAN; LIU, ZHANYI
To: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY CO., LTD.
Reel/Frame 042924/0252 →
Priority Claims (1)
CN 201610972070.2 · Oct 31, 2016 · national
Continuity (1)
Related Publication 20180121810A1 · May 3, 2018