# US 2019/0171713 A1 · Semantic parsing method and apparatus

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

- Number: US 2019/0171713 A1
- Document ID: US20190171713A1
- Status: Published application
- Kind: A1
- Application number: 16/303,000
- Filed: 2017-04-20
- Published: 2019-06-06
- Earliest claimed: 2016-05-19
- Assignees: Beijing Jingdong Shangke Information Technology Co., Ltd.; Beijing Jingdong Century Trading Co., Ltd.
- Applicants: Beijing Jingdong Shangke Information Technology Co., Ltd.; Beijing Jingdong Century Trading Co., Ltd.
- Inventors: Xi Zang; Xin Zhou
- Web page: https://patentgrep.com/patent/US20190171713A1

## CPC

- G06F 40/30: Semantic analysis
- G06F 16/00: Information retrieval; Database structures therefor; File system structures therefor
- G06F 16/243: {Natural language query formulation}
- G06F 40/205: Parsing
- G06F 40/211: Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A semantic parsing method and a semantic parsing apparatus. The method includes: receiving semantic information; parsing the semantic information to obtain a primary text feature; calculating a similarity degree between the primary text feature and a linguistic data feature; using a linguistic data to which a linguistic data feature most similar to the primary text feature belongs, as a primary parsing text; extracting a feature of the primary parsing text to obtain a secondary text feature; obtaining a service feature matching the secondary text feature; obtaining a weight value associated with the obtained service feature; using a service field to which a service feature having a greatest weight value belongs, as the service field to which the primary parsing text belongs; and parsing the primary parsing text in the service field to which the primary parsing text belongs, to obtain a secondary parsing text.

## Claims

### Claim 1

A method for parsing semantics, the method comprising:
receiving semantic information inputted by a user;
parsing the semantic information to obtain a primary text feature of an associated service;
calculating a similarity degree between the primary text feature and a linguistic data feature in a corpus;
using a linguistic data to which a linguistic data feature most similar to the primary text feature belongs, as a primary parsing text;
extracting a feature of the primary parsing text to obtain a secondary text feature;
obtaining a service feature matching the secondary text feature in a service feature database and a service field to which the obtained service feature belongs;
obtaining a weight value associated with the obtained service feature in a service feature weight database;
using a service field to which a service feature having a greatest weight value belongs, as a service field to which the primary parsing text belongs; and
parsing the primary parsing text in the service field to which the primary parsing text belongs, to obtain a secondary parsing text.

### Claim 2 (depends on 1)

The method for parsing semantics according to claim 1, the method further comprising:
querying an answer corresponding to the secondary parsing text in an answer set corresponding to the service field to which the primary parsing text belongs; and
presenting the answer corresponding to the secondary parsing text.

### Claim 3 (depends on 1)

The method for parsing semantics according to claim 1, wherein the parsing the semantic information to obtain a primary text feature of an associated service comprises: obtaining text information of the semantic information; performing cleaning and dimensionality reduction on the text information; performing a semantic and syntactic analysis on the cleaned and dimensionality-reduced text information to obtain a text information feature, and extracting a feature of the associated service in the text information feature to obtain the primary text feature; and/or
the extracting a feature of the primary parsing text to obtain a secondary text feature comprises: obtaining the primary parsing text; performing cleaning and dimensionality reduction on the primary parsing text; and performing a semantic and syntactic analysis on the cleaned and dimensionality-reduced primary parsing text to obtain the secondary text feature.

### Claim 4 (depends on 1)

The method for parsing semantics according to claim 1, wherein the obtaining a service feature matching the secondary text feature in a service feature database comprises:
obtaining the service feature matching the secondary text feature in the service feature database;
performing an anaphora resolution on the service feature matching the secondary text feature to obtain an anaphora-resolved service feature; and
determining the anaphora-resolved service feature as the obtained service feature.

### Claim 5 (depends on 4)

The method for parsing semantics according to claim 4, the method further comprising:
obtaining a sample of service information in a marked service field;
performing a word segmentation on the sample based on statistic and machine learning and determining a part of speech of a segmented word; and
determining the service feature in the service feature database and the service field to which the service feature belongs, according to a frequency of occurrence and the part of speech of the segmented word.

### Claim 6 (depends on 5)

The method for parsing semantics according to claim 5, the method further comprising:
setting a weight of the service feature in the service feature weight database, according to a granularity of the service feature and a preset corresponding relationship between the granularity and the weight.

### Claim 7

An apparatus for parsing semantics, the apparatus comprising:
a processor and a memory storing computer readable instructions, wherein the instructions, when executed, cause the processor to perform operations, the operations comprising:
receiving semantic information inputted by a user;
parsing the semantic information to obtain a primary text feature of an associated service;
calculating a similarity degree between the primary text feature and a linguistic data feature in a corpus;
using a linguistic data to which a linguistic data feature most similar to the primary text feature belongs, as a primary parsing text;
extracting a feature of the primary parsing text to obtain a secondary text feature;
obtaining a service feature matching the secondary text feature in a service feature database and a service field to which the obtained service feature belongs;
obtaining a weight value associated with the obtained service feature in a service feature weight database;
using a service field to which a service feature having a greatest weight value belongs, as a service field to which the primary parsing text belongs; and
parsing the primary parsing text in the service field to which the primary parsing text belongs, to obtain a secondary parsing text.

### Claim 8 (depends on 7)

The apparatus for parsing semantics according to claim 7, the operations further comprising:
querying an answer corresponding to the secondary parsing text in an answer set corresponding to the service field to which the primary parsing text belongs; and
presenting the answer corresponding to the secondary parsing text.

### Claim 9 (depends on 7)

The apparatus for parsing semantics according to claim 7, wherein the parsing the semantic information to obtain a primary text feature of an associated service comprises: obtaining text information of the semantic information; perform cleaning and dimensionality reduction on the text information; performing semantic and syntactic analysis on the cleaned and dimensionality-reduced text information to obtain a text information feature, and extracting a feature of the associated service in the text information feature to obtain the primary text feature; or
the extracting a feature of the primary parsing text to obtain a secondary text feature comprises: obtaining the primary parsing text; performing cleaning and dimensionality reduction on the primary parsing text; and performing a semantic and syntactic analysis on the cleaned and dimensionality-reduced primary parsing text to obtain the secondary text feature.

### Claim 10 (depends on 7)

The apparatus for parsing semantics according to claim 7, wherein the obtaining a service feature matching the secondary text feature in a service feature database comprises: obtaining the service feature matching the secondary text feature in the service feature database; performing an anaphora resolution on the service feature matching the secondary text feature to obtain an anaphora-resolved service feature; and determining the anaphora-resolved service feature as the obtained service feature.

### Claim 11 (depends on 10)

The apparatus for parsing semantics according to claim 10, the operations further comprising:
obtaining a sample of service information in a marked service field;
performing a word segmentation on the sample based on statistic and machine learning and determining a part of speech of a segmented word; and
determining the service feature in the service feature database and the service field to which the service feature belongs, according to a frequency of occurrence and the part of speech of the segmented word.

### Claim 12 (depends on 11)

The apparatus for parsing semantics according to claim 11, the operations further comprising:
setting a weight of the service feature in the service feature weight database, according to a granularity of the service feature and a preset corresponding relationship between the granularity and the weight.

### Claim 13

(canceled)

### Claim 14

A non-volatile computer storage medium, storing computer readable instructions executable by a processor, wherein when executed, the computer readable instructions cause the processor to perform operations, the operations comprising:
receiving semantic information inputted by a user;
parsing the semantic information to obtain a primary text feature of an associated service;
calculating a similarity degree between the primary text feature and a linguistic data feature in a corpus;
using a linguistic data to which a linguistic data feature most similar to the primary text feature belongs, as a primary parsing text;
extracting a feature of the primary parsing text to obtain a secondary text feature;
obtaining a service feature matching the secondary text feature in a service feature database and a service field to which the obtained service feature belongs;
obtaining a weight value associated with the obtained service feature in a service feature weight database;
using a service field to which a service feature having a greatest weight value belongs, as a service field to which the primary parsing text belongs; and
parsing the primary parsing text in the service field to which the primary parsing text belongs, to obtain a secondary parsing text.
