Natural language recognizing apparatus and natural language recognizing method

• US 10,635,859 B2
• Filed: 01/11/2018
• Issued: 04/28/2020
• Est. Priority Date: 10/13/2017
• Status: Active Grant

• Current Assignee: VIA Technologies Incorporated (VIA Technologies)
• Original Assignee: VIA Technologies Incorporated (VIA Technologies)
• Inventors: Zhang, Guo-Feng, Guo, Jing-Jing
• Primary Examiner(s): Le, Thuykhanh
• Application Number:
  US15/867,747
• Publication Number:
  US 20190114317A1
• Time in Patent Office: 838 Days
• Field of Search: None
• CPC Class Codes: G06F 2207/025 String search, i.e. pattern matching, e.g. find identical word or best match in a string
• CPC Class Codes: G06F 40/253 Grammatical analysis; Style critique
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

A natural language recognizing apparatus including an input device, a processing device and a storage device is provided. The input device is configured to provide a natural language data. The storage device is configured to store a plurality of program modules. The program modules include a grammar analysis module. The processing device executes the grammar analysis module to analyze the natural language data through a formal grammar model, and generate a plurality of string data. When at least one of the string data conforms to a preset grammar condition, the processing device judges the at least one of the string data is an intention data, and the processing device outputs a corresponding response signal according to the intention data. In addition, a natural language recognizing method is also provided.

Claims

1. A natural language recognizing apparatus, comprising:

an input device, configured to provide a natural language data;

a storage device, configured to store a plurality of program modules, wherein the plurality of program modules comprise a grammar analysis module; and

a processing device, coupled to the input device and the storage device, the processing device receives the natural language data, and executes the plurality of program modules, wherein the processing device executes the grammar analysis module to analyze the natural language data through a formal grammar model and generates a plurality of string data,

wherein when the processing device determines that at least one of the plurality of string data conforms to a preset valid grammar condition, the processing device determines that the at least one of the string data is an intention data, and the processing device outputs a corresponding response signal according to the intention data,

wherein the formal grammar model comprises a variable symbol, a terminal symbol, a grammar rule, a start symbol and a modifier symbol, and the processing device generates the plurality of string data according to the variable symbol, the terminal symbol, the grammar rule, the start symbol and the modifier symbol by executing the grammar analysis module.

2. The natural language recognizing apparatus according to claim 1, wherein the plurality of program modules further comprise a semantic analysis module, and the processing device executes the semantic analysis module to analyze the plurality of string data, wherein the semantic analysis module determines that the at least one of the plurality of string data conforming to a description of the modifier symbol is the intention data.

3. The natural language recognizing apparatus according to claim 1, wherein the modifier symbol is a global modifier, and when the processing device determines that the at least one of the plurality of string data conforms to a description of the modifier symbol, the processing device determines that the at least one of the string data is the intention data.

4. The natural language recognizing apparatus according to claim 1, wherein the modifier symbol is a slot modifier, and when the processing device determines that the at least one of a portion of the plurality of string data generated according to the variable symbol conforms to a description of the modifier symbol, the processing device determines that the at least one of the portion of the plurality of string data is the intention data.

5. The natural language recognizing apparatus according to claim 4, wherein the intention data comprises an additive intention data corresponding to the at least one of the portion of the plurality of string data generated according to the variable symbol.

6. The natural language recognizing apparatus according to claim 1, wherein the formal grammar model further comprises a multiplex rule, and the multiplex rule is applied to the grammar rule to replace a variable in the grammar rule.

7. A natural language recognizing method, adapted for a natural language recognizing apparatus, wherein the natural language recognizing apparatus comprises an input device and a storage device, and the storage device stores a plurality of program modules, wherein the natural language recognizing method comprises:

receiving a natural language data by the input device, and executing the plurality of modules;

analyzing the natural language data through a formal grammar model by executing a grammar analysis module, and generating a plurality of string data; and

when at least one of the plurality of string data conforms to a preset valid grammar condition, determining that the at least one of the string data is an intention data, and outputting a corresponding response signal according to the intention data,

wherein the formal grammar model comprises a variable symbol, a terminal symbol, a grammar rule, a start symbol and a modifier symbol, wherein the step of analyzing the natural language data through the formal grammar model by executing the grammar analysis module and generating the plurality of string data comprises;

generating the plurality of string data according to the variable symbol, the terminal symbol, the grammar rule, the start symbol and the modifier symbol by executing the grammar analysis module.

8. The natural language recognizing method according to claim 7, wherein when the at least one of the plurality of string data conforms to the preset valid grammar condition, the step of determining the at least one of the string data is the intention data comprises:

analyzing the plurality of string data by executing a semantic analysis module, wherein the semantic analysis module determines that the at least one of the plurality of string data conforming to a description of the modifier symbol is the intention data.

9. The natural language recognizing method according to claim 7, wherein the modifier symbol is a global modifier, and when the at least one of the plurality of string data conforms to the preset valid grammar condition, the step of determining that the at least one of the string data is the intention data comprises:

when the at least one of the plurality of string data conforms to a description of the modifier symbol, determining that the at least one of the string data is the intention data.

10. The natural language recognizing method according to claim 7, wherein the modifier symbol is a slot modifier, and when the at least one of the plurality of string data conforms to the preset valid grammar condition, the step of determining the at least one of the string data is the intention data comprises:

when it is determined that the at least one of a portion of the plurality of string data generated according to the variable symbol conforms to a description of the modifier symbol, determining that the at least one of the portion of the plurality of string data is the intention data.

11. The natural language recognizing method according to claim 10, wherein the intention data comprises an additive intention data corresponding to the at least one of the portion of the plurality of string data generated according to the variable symbol.

12. The natural language recognizing method according to claim 7, wherein the formal grammar model further comprises a multiplex rule, and the step of analyzing the natural language data through the formal grammar model by executing the grammar analysis module and generating the plurality of string data comprises:

applying the multiplex rule to the grammar rule to replace a variable in the grammar rule.