Knowledge base construction method and device

Through semantic analysis and knowledge graph construction methods, keywords and fields in public opinion information are extracted and knowledge bases are generated, which solves the problem of difficult to quickly understand public opinion information, and achieves streamlined extraction and effective insights of information.

CN112000813BActive Publication Date: 2024-12-03ALIPAY (HANGZHOU) INFORMATION TECH CO LTD
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202010963069.X
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-09-14
Publication Date
2024-12-03
Estimated Expiration
2040-09-14

AI Technical Summary

Technical Problem

In the Internet era, it is difficult for people to quickly and effectively perceive important public opinion information in massive amounts of information, resulting in wasting time and energy.

Method used

By obtaining public opinion information, semantic analysis, determining keywords, building a knowledge graph in the public opinion field, generating a public opinion knowledge base, extracting and organizing effective information.

Benefits of technology

Simplify complex and cumbersome public opinion information, accurately and effectively understand public opinion changes, and save time and energy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112000813B_ABST
    Figure CN112000813B_ABST
Patent Text Reader

Abstract

The present specification provides a knowledge base construction method and device, wherein the knowledge base construction method includes: obtaining public opinion information, and performing semantic analysis on the public opinion information to determine at least one public opinion keyword corresponding to the public opinion information; determining the public opinion field corresponding to each public opinion keyword; constructing a public opinion field knowledge graph corresponding to each public opinion field based on the public opinion information; and generating a public opinion knowledge base based on each public opinion field knowledge graph. The method performs semantic analysis on the public opinion information to determine the public opinion keywords corresponding to the public opinion information, and constructs a public opinion knowledge graph of the public opinion field corresponding to the public opinion keywords based on the public opinion information, thereby generating a public opinion knowledge base. This method can simplify the content of complex and cumbersome public opinion information, extract effective information, and thus accurately and effectively understand the changes in public opinion, saving time and energy.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This specification relates to the field of knowledge graph technology, and in particular to a knowledge base construction method and device. Background Art

[0002] Public opinion refers to the social attitudes that the public, as the subject, generates and holds towards social managers, enterprises, individuals and other organizations, as the object, and their political, social, moral and other orientations, surrounding the occurrence, development and changes of social events within a certain social space.

[0003] With the advent of the Internet age, there will be a huge amount of information to evaluate a certain social event. Among the huge amount of information, there is often a lot of redundant information that is irrelevant to the social event, which brings some unnecessary troubles to the public's understanding of the public opinion information of the social event. The public cannot quickly, effectively and accurately perceive important information from the large amount of information, wasting a lot of time and energy.

[0004] Based on this, an effective method is urgently needed to solve the above problems. Summary of the invention

[0005] In view of this, the embodiments of this specification provide a knowledge base construction method. This specification also relates to a knowledge base construction device, a computing device, and a computer-readable storage medium to solve the technical defects existing in the prior art.

[0006] According to a first aspect of an embodiment of this specification, a knowledge base construction method is provided, including:

[0007] Acquire public opinion information, and perform semantic analysis on the public opinion information to determine at least one public opinion keyword corresponding to the public opinion information;

[0008] Determine the public opinion field corresponding to each public opinion keyword;

[0009] Constructing a public opinion field knowledge graph corresponding to each public opinion field according to the public opinion information;

[0010] Generate a public opinion knowledge base based on the knowledge graph of each public opinion field.

[0011] Optionally, constructing a public opinion field knowledge graph corresponding to each public opinion field according to the public opinion information includes:

[0012] Determine the target public opinion information corresponding to each public opinion field;

[0013] A public opinion field knowledge graph corresponding to the public opinion field is constructed according to the target public opinion information.

[0014] Optionally, constructing a public opinion domain knowledge graph corresponding to the public opinion domain according to the target public opinion information includes:

[0015] Performing knowledge processing on the target public opinion information to obtain candidate data;

[0016] Performing knowledge fusion on the candidate data to obtain structured data;

[0017] Knowledge reasoning is performed on the structured data to obtain a public opinion field knowledge graph in the public opinion field.

[0018] Optionally, performing knowledge processing on the target public opinion information to obtain candidate data includes:

[0019] Entity extraction, relationship extraction and attribute extraction are performed on the target public opinion information to obtain candidate entities, relationships and attribute information of candidate entities in the target public opinion information.

[0020] Optionally, performing knowledge fusion on the candidate data to obtain structured data includes:

[0021] Perform entity disambiguation, entity normalization, and coreference resolution on the candidate entities to determine a target entity;

[0022] The corresponding target relationship and attribute information of the target entity are determined according to the target entity.

[0023] Optionally, performing knowledge reasoning on the structured data to obtain a public opinion field knowledge graph in the public opinion field includes:

[0024] Constructing the ontology of the public opinion field according to the target entity and the target relationship;

[0025] A public opinion domain knowledge graph of the public opinion domain is generated according to the attribute information of the target entity and the ontology of the public opinion domain.

[0026] Optional: Get public opinion information, including:

[0027] Obtain at least one of public opinion news information, authoritative speech information, public regulations information, and industry dynamics information.

[0028] Optionally, also include:

[0029] A public opinion consultation platform is constructed based on the public opinion knowledge base.

[0030] Optionally, also include:

[0031] Receive updated public opinion information, and update the public opinion knowledge base according to the updated public opinion information.

[0032] According to a second aspect of the embodiments of this specification, there is provided a knowledge base construction device, including:

[0033] An acquisition module is configured to acquire public opinion information, and perform semantic analysis on the public opinion information to determine at least one public opinion keyword corresponding to the public opinion information;

[0034] A determination module is configured to determine the public opinion field corresponding to each public opinion keyword;

[0035] A construction module is configured to construct a public opinion field knowledge graph corresponding to each public opinion field according to the public opinion information;

[0036] The generation module is configured to generate a public opinion knowledge base according to the knowledge graph of each public opinion field.

[0037] According to a third aspect of an embodiment of this specification, a computing device is provided, including:

[0038] Memory and processor;

[0039] The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions:

[0040] Acquire public opinion information, and perform semantic analysis on the public opinion information to determine at least one public opinion keyword corresponding to the public opinion information;

[0041] Determine the public opinion field corresponding to each public opinion keyword;

[0042] Constructing a public opinion field knowledge graph corresponding to each public opinion field according to the public opinion information;

[0043] Generate a public opinion knowledge base based on the knowledge graph of each public opinion field.

[0044] According to a fourth aspect of the embodiments of this specification, a computer-readable storage medium is provided, which stores computer-executable instructions, and when the instructions are executed by a processor, the steps of any of the knowledge base construction methods are implemented.

[0045] The knowledge base construction method provided in this specification obtains public opinion information, and performs semantic analysis on the public opinion information to determine at least one public opinion keyword corresponding to the public opinion information; determines the public opinion field corresponding to each public opinion keyword; constructs a public opinion field knowledge graph corresponding to each public opinion field based on the public opinion information; and generates a public opinion knowledge base based on each public opinion field knowledge graph. This method performs semantic analysis on the public opinion information to determine the public opinion keywords corresponding to the public opinion information, and constructs a public opinion knowledge graph of the public opinion field corresponding to the public opinion keywords based on the public opinion information, and then generates a public opinion knowledge base. This method can simplify the content of complex and cumbersome public opinion information, extract effective information, and then accurately and effectively understand the changes in public opinion, saving time and energy. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is a flow chart of a knowledge base construction method provided in an embodiment of this specification;

[0047] Figure 2 It is a flowchart of a method for constructing a knowledge graph in the field of public opinion in a knowledge base construction method provided in an embodiment of this specification;

[0048] Figure 3 is a processing flow chart of a knowledge base construction method provided by another embodiment of this specification;

[0049] Figure 4 It is a structural schematic diagram of a knowledge base construction device provided in an embodiment of this specification;

[0050] Figure 5 It is a structural block diagram of a computing device provided in one embodiment of this specification. DETAILED DESCRIPTION

[0051] Many specific details are described in the following description to facilitate a full understanding of this specification. However, this specification can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of this specification, so this specification is not limited to the specific implementation disclosed below.

[0052] The terms used in one or more embodiments of this specification are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of this specification. The singular forms of "a", "said" and "the" used in one or more embodiments of this specification and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0053] It should be understood that although the terms first, second, etc. may be used to describe various information in one or more embodiments of this specification, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of this specification, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0054] First, the terms involved in one or more embodiments of this specification are explained.

[0055] Public opinion: the abbreviation of public opinion, refers to the social attitudes generated and held by the public as the subject towards the social managers, enterprises, individuals and other organizations as the object and their political, social, moral and other orientations around the occurrence, development and changes of intermediary social events within a certain social space.

[0056] Public opinion information: a description and reflection of public opinion. The so-called public opinion information refers to the information, news, intelligence, instructions, data and signals that objectively reflect the state of public opinion and its movement in the process of collecting, organizing, analyzing, reporting, using and feedback of people's social and political attitudes.

[0057] Knowledge graph: A semantic network that reveals the relationships between entities.

[0058] Knowledge base: It is a knowledge-based and intelligent system that can be composed of multiple knowledge graphs. It is a structured, easy-to-operate, easy-to-use, and comprehensive and organized knowledge cluster.

[0059] In this specification, a knowledge base construction method is provided. This specification also involves a knowledge base construction device, a computing device, and a computer-readable storage medium, which are described in detail one by one in the following embodiments.

[0060] Figure 1 A flowchart of a knowledge base construction method provided according to an embodiment of this specification is shown, which specifically includes the following steps:

[0061] Step 102: Obtain public opinion information, and perform semantic analysis on the public opinion information to determine at least one public opinion keyword corresponding to the public opinion information.

[0062] Public opinion information is a description and reflection of public opinion. The so-called public opinion information refers to the information, news, intelligence, instructions, data and signals that objectively reflect the state of public opinion and its movement in the process of collecting, organizing, analyzing, reporting, using and feedback of people's social and political attitudes. It can be public opinion news information, authoritative speech information from authoritative people, laws and regulations, public regulations, expert opinions, management systems and other information related to public opinion.

[0063] Semantic analysis is a technology that summarizes and analyzes information. By analyzing the text of the information, the core key content of the information can be extracted. Semantic analysis of the public opinion information to determine the public opinion keywords can be achieved through a pre-trained semantic analysis model.

[0064] Public opinion keywords are the core content of the public opinion information. By conducting semantic analysis on the public opinion information and determining the keywords of the public opinion information, we can determine the core focus of the public opinion information. For example, if expert M publishes an article on event A, the keywords can be expert M, event A, and opinion; if partner B releases a new product O, the keywords can be merchant B, product O, and so on.

[0065] Optionally, obtaining public opinion information may include obtaining at least one of public opinion news information, authoritative speech information, public notice information, and industry dynamic information.

[0066] Among them, public opinion news information refers to the information reported or published on authoritative websites, such as a certain public opinion website, a certain news website, etc.

[0067] Authoritative speech information refers to public statements released by senior leaders or authoritative persons, such as a speech by a senior official at a conference or an opinion article published by an authoritative person.

[0068] The public disclosure information refers to some laws, regulations, local policies, etc. that are disclosed externally.

[0069] Industry dynamic information refers to information such as partners or competitors in the industry releasing new products, taking new actions, or making improvements to existing products.

[0070] Step 104: Determine the public opinion field corresponding to each public opinion keyword.

[0071] Based on each public opinion keyword, the public opinion field corresponding to this public opinion keyword can be determined, so as to construct the corresponding public opinion field knowledge graph in the corresponding public opinion field. For example, if the keyword is a certain event, it can be determined as the event field; if the keyword is a certain expert, it can be determined as the expert field; if the keyword is a certain product, it can be determined as the product field; if the keyword is laws and regulations, it can be determined as the laws and regulations field.

[0072] Step 106: Construct a public opinion field knowledge graph corresponding to each public opinion field according to the public opinion information.

[0073] Public opinion information may include multiple keywords, so the same piece of public opinion information can serve as the data basis for public opinion field knowledge graphs corresponding to multiple public opinion fields. Based on the public opinion information, a corresponding public opinion field knowledge graph is constructed for each public opinion field. The public opinion information is used to construct a corresponding data relationship network according to the public opinion field to support upper-level data applications.

[0074] Optionally, constructing a public opinion field knowledge graph corresponding to each public opinion field according to the public opinion information includes:

[0075] Determine the target public opinion information corresponding to each public opinion field;

[0076] A public opinion field knowledge graph corresponding to the public opinion field is constructed according to the target public opinion information.

[0077] In actual applications, each piece of public opinion information will correspond to at least one public opinion field. Correspondingly, each public opinion field will correspond to multiple public opinion information. First, we must determine the target public opinion information corresponding to each public opinion field. For example, for the product field, the corresponding public opinion information is the public opinion information about the product in the industry dynamics. For the expert field, the corresponding public opinion information is the public opinion information interpreted by experts, and so on. Determining the target public opinion information corresponding to each public opinion field can help find the target data in a large amount of data information, saving time and improving efficiency for the subsequent construction of the knowledge graph.

[0078] After determining the target public opinion information, build a public opinion domain knowledge graph for the public opinion field based on the target public opinion information. Figure 2 , Figure 2 The flowchart of the method for constructing a knowledge graph in the public opinion field in the knowledge base construction method is shown. The public opinion field knowledge graph corresponding to the public opinion field is constructed according to the target public opinion information, including the following steps 202 to 206:

[0079] Step 202: Perform knowledge processing on the target public opinion information to obtain candidate data.

[0080] Knowledge processing, also known as information extraction, is the first step in building a knowledge graph. Information extraction is a technology that automatically extracts structured information such as entities, relationships, and entity attributes from unstructured data and semi-structured data. Unstructured data is data with irregular or incomplete structure, no predefined data model, and is not convenient to represent using a two-dimensional logical table in a database. It includes office documents of all formats, text, pictures, HTML, various reports, images, audio and video information, etc. Semi-structured data refers to data that has a certain degree of structure compared to unstructured data, such as XML files, JSON files, encyclopedias, etc.

[0081] Optionally, performing knowledge processing on the target public opinion information to obtain candidate data includes:

[0082] Entity extraction, relationship extraction and attribute extraction are performed on the target public opinion information to obtain candidate entities, relationships and attribute information of candidate entities in the target public opinion information.

[0083] Entity extraction, also known as named entity recognition (NER), refers to the automatic identification of named entities from text datasets. For example, for a text "A Group was established in 2009, and Mr. Wang has been working diligently since he took office as chairman...", through entity extraction, we can obtain the entities "A Group" and "Mr. Wang".

[0084] After entity extraction of the target audit information, a series of discrete named entities are obtained. In order to obtain semantic information, it is also necessary to extract the association relationship between entities from the relevant corpus and connect the entities through relationships to form a network knowledge structure. This is what relationship extraction needs to do. Taking the above text as an example, "A Group was established in 2009. Since Wang took office as chairman, he has been working diligently..." The relationship extracted from it is "Wang" is the "Chairman" of "A Group".

[0085] The purpose of attribute extraction is to collect attribute information of specific entities from different information sources. For example, the nickname, birthday, nationality, educational background, etc. of a public figure or well-known enterprise can be obtained from the public information on the Internet. Continuing with the above example, for the chairman of "Group A", "Mr. Wang", the public information on the Internet can be used to find out that "Mr. Wang" was born in 1968, is from City C, and graduated from graduate school, and other attribute information related to Mr. Wang.

[0086] Candidate data refers to the entities, relationships, and entity attribute information obtained after knowledge processing of the target audit information.

[0087] Step 204: Perform knowledge fusion on the candidate data to obtain structured data.

[0088] Through knowledge processing, we can obtain entities, relationships, and attribute information of entities from the original target audit information. However, there are many candidate data obtained from the target audit information, which are scattered, repeated, and even interfered by some wrong data. Therefore, we need to fuse the candidate data, exclude the wrong information, normalize the repeated data, and clarify the information of the unclear data.

[0089] Optionally, performing knowledge fusion on the candidate data to obtain structured data includes:

[0090] Perform entity disambiguation, entity normalization, and coreference resolution on the candidate entities to determine a target entity;

[0091] The corresponding target relationship and attribute information of the target entity are determined according to the target entity.

[0092] In practical applications, the same word may express different meanings in different contexts, so entity disambiguation is needed. The purpose of entity disambiguation is to match the same word with different entities according to different contexts. For example, for James, if he appears in the context of basketball, he can be determined as an NBA star, and if he appears in the context of movies, he can be determined as a movie director.

[0093] Similarly, in actual applications, there may be a situation where two words correspond to the same entity, such as "Beijing" and "the capital of the motherland". Although the two words are literally two different entities, they actually refer to the same entity, and it is necessary to perform entity resolution operations on multiple candidate entities.

[0094] Co-reference resolution is also an important step in knowledge fusion. In the target audit information, there are usually many pronouns such as "he", "it", "they", etc. Knowledge fusion also needs to determine the entity corresponding to each pronoun. For example, for a sentence "Zhang San did not come to work yesterday because he attended Li Si's hearing, but he came to work today." After co-reference resolution, it can be determined that the "he" specifically refers to "Zhang San" rather than "Li Si".

[0095] After performing entity disambiguation, entity normalization, and coreference resolution on candidate entities, the target entity can be determined. Based on the target entity, the relationship corresponding to the target entity and the attribute information of the target entity can be determined. At this point, the entities, relationships, and attribute information of the entities for constructing the knowledge graph have been prepared, and structured data has been constructed.

[0096] Step 206: Perform knowledge reasoning on the structured data to obtain a public opinion knowledge graph in the public opinion field.

[0097] After the above series of operations, we get a series of basic factual expressions. However, facts themselves are not equivalent to knowledge. If we want to finally obtain a knowledge graph, we also need to perform knowledge reasoning on the data to obtain the association information between entities, the causal relationship between events, etc.

[0098] Optionally, performing knowledge reasoning on the structured data to obtain an audit domain knowledge graph of the audit domain includes:

[0099] Constructing the ontology of the public opinion field according to the target entity and the target relationship;

[0100] A public opinion domain knowledge graph of the public opinion domain is generated according to the attribute information of the target entity and the ontology of the public opinion domain.

[0101] Ontology (Schema) refers to a set of concepts and a conceptual framework. Ontology can be constructed manually by human editing or in a data-driven automated way. Specifically, the process of automated ontology construction includes three stages: calculation of similarity of parallel relationships between entities, extraction of hierarchical and subordinate relationships between entities, and generation of ontology.

[0102] After obtaining multiple entities, first convert the entities into vectors, then calculate the similarity between multiple entities, determine the entities with similarity greater than the threshold as similar entities, and then classify the entities into upper and lower levels based on their corresponding target relationships, and then construct the corresponding ontology, and then correspond the entity attribute information corresponding to each entity to generate a knowledge graph for the corresponding field.

[0103] It should be noted that in actual applications, most of the relationships between the knowledge graphs generated at this time are still incomplete. They can be further improved based on the relationships between different entities. For example, if A and B are father and son, B is the chairman of C Group, and C Group is located in City D, then it can be inferred that A lives in City D. Then, the incomplete relationships between the knowledge graphs can be filled in to construct a complete knowledge graph.

[0104] In this specification, a public opinion field knowledge graph corresponding to each public opinion field is constructed based on the public opinion information, such as an expert opinion knowledge graph, a product knowledge graph, a website disclosure knowledge graph, an industry company knowledge graph, etc.

[0105] Step 108: Generate a public opinion knowledge base based on the knowledge graph of each public opinion field.

[0106] Multiple public opinion knowledge graphs are combined together to generate a public opinion knowledge base. The public opinion knowledge base contains public opinion knowledge graphs in different fields. The public opinion knowledge base combines public opinion knowledge graphs in multiple fields, making it easier to quickly, accurately and effectively understand changes in public opinion information, and efficiently adjust business content in a timely manner according to changes in public opinion information.

[0107] Optionally, after generating the public opinion knowledge base, the method further includes:

[0108] A public opinion consultation platform is constructed based on the public opinion knowledge base.

[0109] In practical applications, the obtained public opinion knowledge base is relatively abstract and can be in the form of a graph database. However, users without computer knowledge cannot effectively utilize the public opinion knowledge base. Therefore, after generating the public opinion knowledge base, a public opinion consultation platform can be built for the public opinion knowledge base to visualize and concretize the form of the graph database. Users can quickly and directly query the contents of the public opinion knowledge base through a graphical interface. At the same time, the public opinion platform also saves public opinion information, that is, users can not only directly use the public opinion knowledge base through the public opinion platform, but also query the content of public opinion information in the public opinion platform, which is convenient for users to use and improves the user experience.

[0110] Optionally, the method further includes:

[0111] Receive updated public opinion information, and update the public opinion knowledge base according to the updated public opinion information.

[0112] Updating public opinion information specifically refers to new public opinion information, such as new laws and regulations, expert opinions, industry trends, etc. In actual applications, the public opinion knowledge base is not fixed after it is generated, and it is necessary to update the public opinion knowledge base at any time according to the updated public opinion information. Updating the public opinion knowledge base according to the new public opinion information can timely obtain new public opinion information.

[0113] The knowledge base construction method provided in this specification obtains public opinion information, and performs semantic analysis on the public opinion information to determine at least one public opinion keyword corresponding to the public opinion information; determines the public opinion field corresponding to each public opinion keyword; constructs a public opinion field knowledge graph corresponding to each public opinion field based on the public opinion information; and generates a public opinion knowledge base based on each public opinion field knowledge graph. This method performs semantic analysis on the public opinion information to determine the public opinion keywords corresponding to the public opinion information, and constructs a public opinion knowledge graph of the public opinion field corresponding to the public opinion keywords based on the public opinion information, and then generates a public opinion knowledge base. This method can simplify the content of complex and cumbersome public opinion information, extract effective information, and then accurately and effectively understand the changes in public opinion, saving time and energy.

[0114] The following combination Figure 3 , the knowledge base construction method is further described. Figure 3 A processing flow chart of a method for constructing a knowledge base applied to the field of public opinion provided in an embodiment of this specification is shown, which specifically includes the following steps:

[0115] Step 302: Obtain public opinion information, and perform semantic analysis on the public opinion information to determine at least one public opinion keyword corresponding to the public opinion information.

[0116] In the embodiments provided in this specification, the knowledge base construction method provided in this specification is further explained by taking the public opinion information of "1. Partner M released new product A" and "2. N experts made comments on product A at a conference" as an example. The above public opinion information is semantically analyzed to determine that the keywords of the first public opinion information are partner M (manufacturer M) and new product A, and the keywords of the second public opinion information are N experts and product A evaluation.

[0117] Step 304: Determine the public opinion field corresponding to each public opinion keyword.

[0118] In the embodiment provided in this specification, the public opinion field corresponding to the manufacturer M is determined as the merchant field, the public opinion field corresponding to the N experts is determined as the expert field, and the public opinion field corresponding to the new product A is determined as the product field.

[0119] Step 306: Determine the target public opinion information corresponding to each public opinion field.

[0120] In the embodiments provided in this specification, the target public opinion information corresponding to the merchant field is determined as "1. Partner M has released a new product A", the target public opinion information corresponding to the expert field is determined as "2. Information on the remarks made by N experts on A at a certain conference", and the target public opinion information corresponding to the product field is determined as "1. Partner M has released a new product A" and "2. Information on the remarks made by N experts on A at a certain conference".

[0121] Step 308: Perform knowledge processing on the target public opinion information to obtain candidate data.

[0122] In the embodiments provided in this specification, taking the business field as an example, knowledge processing is performed on the target public opinion information "1. Partner M releases new product A" to obtain a candidate data set.

[0123] Step 310: Perform knowledge fusion on the candidate data to obtain structured data.

[0124] In the embodiments provided in this specification, knowledge fusion is performed on the candidate data set to obtain structured data in the public opinion field of the business field.

[0125] Step 312: Perform knowledge reasoning on the structured data to obtain a public opinion knowledge graph in the public opinion field.

[0126] In the embodiments provided in this specification, knowledge reasoning is performed on structured data in the field of public opinion in the merchant field to obtain a knowledge graph of public opinion in the merchant field, knowledge reasoning is performed on structured data in the field of public opinion in the expert field to obtain a knowledge graph of public opinion in the expert field, and knowledge reasoning is performed on structured data in the field of public opinion in the product field to obtain a knowledge graph of public opinion in the product field.

[0127] Step 314: Generate a public opinion knowledge base based on the knowledge graph of each public opinion field.

[0128] In the embodiments provided in this specification, a public opinion knowledge base is constructed based on the business field public opinion knowledge graph, the expert field public opinion knowledge graph, and the product field public opinion knowledge graph.

[0129] Step 316: Construct a public opinion consultation platform based on the public opinion knowledge base.

[0130] In the embodiments provided in this specification, a corresponding public opinion consultation platform is constructed based on the public opinion knowledge base for users to use.

[0131] The knowledge base construction method provided in this specification obtains public opinion information, and performs semantic analysis on the public opinion information to determine at least one public opinion keyword corresponding to the public opinion information; determines the public opinion field corresponding to each public opinion keyword; constructs a public opinion field knowledge graph corresponding to each public opinion field based on the public opinion information; and generates a public opinion knowledge base based on each public opinion field knowledge graph. This method performs semantic analysis on the public opinion information to determine the public opinion keywords corresponding to the public opinion information, and constructs a public opinion knowledge graph of the public opinion field corresponding to the public opinion keywords based on the public opinion information, and then generates a public opinion knowledge base. This method can simplify the content of complex and cumbersome public opinion information, extract effective information, and then accurately and effectively understand the changes in public opinion, saving time and energy.

[0132] Corresponding to the above method embodiment, this specification also provides a knowledge base construction device embodiment, Figure 4 FIG. 1 shows a schematic diagram of a knowledge base construction device provided in an embodiment of the present specification. Figure 4 As shown, the device comprises:

[0133] The acquisition module 402 is configured to acquire public opinion information, and perform semantic analysis on the public opinion information to determine at least one public opinion keyword corresponding to the public opinion information;

[0134] A determination module 404 is configured to determine the public opinion field corresponding to each public opinion keyword;

[0135] A construction module 406 is configured to construct a public opinion field knowledge graph corresponding to each public opinion field according to the public opinion information;

[0136] The generation module 408 is configured to generate a public opinion knowledge base according to each public opinion field knowledge graph.

[0137] Optionally, the construction module 406 includes:

[0138] A determination unit, configured to determine target public opinion information corresponding to each public opinion field;

[0139] The construction unit is configured to construct a public opinion field knowledge graph corresponding to the public opinion field according to the target public opinion information.

[0140] Optionally, the construction unit includes:

[0141] A knowledge processing subunit is configured to perform knowledge processing on the target public opinion information to obtain candidate data;

[0142] a knowledge fusion subunit, configured to perform knowledge fusion on the candidate data to obtain structured data;

[0143] The knowledge reasoning subunit is configured to perform knowledge reasoning on the structured data to obtain a public opinion field knowledge graph in the public opinion field.

[0144] Optionally, the knowledge processing subunit is further configured to:

[0145] Entity extraction, relationship extraction and attribute extraction are performed on the target public opinion information to obtain candidate entities, relationships and attribute information of candidate entities in the target public opinion information.

[0146] Optionally, the knowledge fusion subunit is further configured as follows:

[0147] Perform entity disambiguation, entity normalization, and coreference resolution on the candidate entities to determine a target entity;

[0148] The corresponding target relationship and attribute information of the target entity are determined according to the target entity.

[0149] Optionally, the knowledge reasoning subunit is further configured as follows:

[0150] Constructing the ontology of the public opinion field according to the target entity and the target relationship;

[0151] A public opinion domain knowledge graph of the public opinion domain is generated according to the attribute information of the target entity and the ontology of the public opinion domain.

[0152] Optionally, the acquisition module 402 is further configured to:

[0153] Obtain at least one of public opinion news information, authoritative speech information, public regulations information, and industry dynamics information.

[0154] Optionally, the device further comprises:

[0155] The platform construction module is configured to construct a public opinion consultation platform based on the public opinion knowledge base.

[0156] Optionally, the device further comprises:

[0157] The updating module is configured to receive updated public opinion information and update the public opinion knowledge base according to the updated public opinion information.

[0158] The knowledge base construction device provided in the present specification obtains public opinion information, and performs semantic analysis on the public opinion information to determine at least one public opinion keyword corresponding to the public opinion information; determines the public opinion field corresponding to each public opinion keyword; constructs a public opinion field knowledge graph corresponding to each public opinion field based on the public opinion information; and generates a public opinion knowledge base based on each public opinion field knowledge graph. The device performs semantic analysis on the public opinion information to determine the public opinion keywords corresponding to the public opinion information, and constructs a public opinion knowledge graph of the public opinion field corresponding to the public opinion keywords based on the public opinion information, and then generates a public opinion knowledge base. The device can simplify the content of complex and cumbersome public opinion information, extract effective information, and then accurately and effectively understand the changes in public opinion, saving time and energy.

[0159] The above is a schematic scheme of a knowledge base construction device of this embodiment. It should be noted that the technical scheme of the knowledge base construction device and the technical scheme of the above-mentioned knowledge base construction method belong to the same concept, and the details not described in detail in the technical scheme of the knowledge base construction device can be referred to the description of the technical scheme of the above-mentioned knowledge base construction method.

[0160] Figure 5The structure block diagram of a computing device 500 provided according to an embodiment of the present specification is shown. The components of the computing device 500 include but are not limited to a memory 510 and a processor 520. The processor 520 is connected to the memory 510 via a bus 530, and the database 550 is used to store data.

[0161] The computing device 500 also includes an access device 540 that enables the computing device 500 to communicate via one or more networks 560. Examples of these networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 540 may include one or more of any type of network interface (e.g., a network interface card (NIC)) whether wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a World Wide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, a near field communication (NFC) interface, and the like.

[0162] In one embodiment of the present specification, the above components of the computing device 500 and Figure 5 Other components not shown in the figure may also be connected to each other, for example, via a bus. It should be understood that Figure 5 The computing device structure block diagram shown is only for the purpose of illustration, and is not intended to limit the scope of this specification. Those skilled in the art can add or replace other components as needed.

[0163] The computing device 500 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smart phone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or PC. The computing device 500 may also be a mobile or stationary server.

[0164] The processor 520 is used to execute the following computer executable instructions:

[0165] Acquire public opinion information, and perform semantic analysis on the public opinion information to determine at least one public opinion keyword corresponding to the public opinion information;

[0166] Determine the public opinion field corresponding to each public opinion keyword;

[0167] Constructing a public opinion field knowledge graph corresponding to each public opinion field according to the public opinion information;

[0168] Generate a public opinion knowledge base based on the knowledge graph of each public opinion field.

[0169] The above is a schematic scheme of a computing device of this embodiment. It should be noted that the technical scheme of the computing device and the technical scheme of the above-mentioned knowledge base construction method belong to the same concept, and the details not described in detail in the technical scheme of the computing device can be referred to the description of the technical scheme of the above-mentioned knowledge base construction method.

[0170] An embodiment of the present specification further provides a computer-readable storage medium storing computer instructions, which are used when executed by a processor to:

[0171] Acquire public opinion information, and perform semantic analysis on the public opinion information to determine at least one public opinion keyword corresponding to the public opinion information;

[0172] Determine the public opinion field corresponding to each public opinion keyword;

[0173] Constructing a public opinion field knowledge graph corresponding to each public opinion field according to the public opinion information;

[0174] Generate a public opinion knowledge base based on the knowledge graph of each public opinion field.

[0175] The above is a schematic scheme of a computer-readable storage medium of this embodiment. It should be noted that the technical scheme of the storage medium and the technical scheme of the above-mentioned knowledge base construction method belong to the same concept, and the details not described in detail in the technical scheme of the storage medium can be referred to the description of the technical scheme of the above-mentioned knowledge base construction method.

[0176] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0177] The computer instructions include computer program codes, which may be in source code form, object code form, executable files or some intermediate forms, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0178] It should be noted that, for the convenience of description, the aforementioned method embodiments are all described as a series of action combinations, but those skilled in the art should be aware that this specification is not limited by the order of the actions described, because according to this specification, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this specification.

[0179] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0180] The preferred embodiments of this specification disclosed above are only used to help explain this specification. The optional embodiments do not describe all the details in detail, nor do they limit the invention to only the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of this specification, so that technicians in the relevant technical field can well understand and use this specification. This specification is limited only by the claims and their full scope and equivalents.

Claims

1. A knowledge base construction method, comprising: Acquire public opinion information, and perform semantic analysis on the public opinion information to determine at least one public opinion keyword corresponding to the public opinion information; Determine the public opinion field corresponding to each public opinion keyword; Constructing a public opinion field knowledge graph corresponding to each of the public opinion fields according to the public opinion information, wherein the constructing a public opinion field knowledge graph corresponding to each of the public opinion fields according to the public opinion information includes: constructing an ontology of the public opinion field corresponding to each public opinion keyword according to a target entity and a target relationship, wherein the target entity is determined based on a candidate entity, the target relationship is determined based on the target entity, and the public opinion field knowledge graph corresponding to the public opinion field is generated based on the attribute information of the target entity and the ontology corresponding to the public opinion field corresponding to each public opinion keyword; Generate a public opinion knowledge base according to each public opinion field knowledge graph, so that the public opinion knowledge base combines the public opinion field knowledge graphs of multiple public opinion fields; A public opinion consultation platform is constructed based on the public opinion knowledge base.

2. The knowledge base construction method according to claim 1, constructing a public opinion field knowledge graph corresponding to each public opinion field according to the public opinion information, comprising: Determine the target public opinion information corresponding to each public opinion field; A public opinion field knowledge graph corresponding to the public opinion field is constructed according to the target public opinion information.

3. The knowledge base construction method according to claim 2, constructing a public opinion field knowledge graph corresponding to the public opinion field according to the target public opinion information, comprising: Performing knowledge processing on the target public opinion information to obtain candidate data; Performing knowledge fusion on the candidate data to obtain structured data; Knowledge reasoning is performed on the structured data to obtain a public opinion field knowledge graph in the public opinion field.

4. The knowledge base construction method according to claim 3, performing knowledge processing on the target public opinion information to obtain candidate data, comprising: Entity extraction, relationship extraction and attribute extraction are performed on the target public opinion information to obtain candidate entities, relationships and attribute information of candidate entities in the target public opinion information.

5. The knowledge base construction method according to claim 4, wherein the candidate data is subjected to knowledge fusion to obtain structured data, comprising: Perform entity disambiguation, entity normalization, and coreference resolution on the candidate entities to determine a target entity; The corresponding target relationship and attribute information of the target entity are determined according to the target entity.

6. The knowledge base construction method according to claim 1, obtaining public opinion information, comprising: Obtain at least one of public opinion news information, authoritative speech information, public regulations information, and industry dynamics information.

7. The knowledge base construction method according to claim 1, further comprising: Receive updated public opinion information, and update the public opinion knowledge base according to the updated public opinion information.

8. A knowledge base construction device, comprising: An acquisition module is configured to acquire public opinion information, and perform semantic analysis on the public opinion information to determine at least one public opinion keyword corresponding to the public opinion information; A determination module is configured to determine the public opinion field corresponding to each public opinion keyword; A construction module is configured to construct a public opinion field knowledge graph corresponding to each of the public opinion fields according to the public opinion information, wherein the construction of the public opinion field knowledge graph corresponding to each of the public opinion fields according to the public opinion information includes: constructing an ontology corresponding to multiple entities in the public opinion field corresponding to each public opinion keyword according to a target entity and a target relationship, wherein the target entity is determined based on a candidate entity, the target relationship is determined based on the target entity, and a public opinion field knowledge graph corresponding to the public opinion field is generated based on the attribute information of the target entity and the ontology corresponding to the public opinion field corresponding to each public opinion keyword; A generation module is configured to generate a public opinion knowledge base according to each public opinion field knowledge graph, so that the public opinion knowledge base combines the public opinion field knowledge graphs of multiple public opinion fields; The platform construction module is configured to construct a public opinion consultation platform based on the public opinion knowledge base.

9. A computing device comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the following method: Acquire public opinion information, and perform semantic analysis on the public opinion information to determine at least one public opinion keyword corresponding to the public opinion information; Determine the public opinion field corresponding to each public opinion keyword; Constructing a public opinion field knowledge graph corresponding to each of the public opinion fields according to the public opinion information, wherein constructing a public opinion field knowledge graph corresponding to each of the public opinion fields according to the public opinion information comprises: constructing an ontology corresponding to multiple entities in the public opinion field corresponding to each public opinion keyword according to a target entity and a target relationship, wherein the target entity is determined based on a candidate entity, the target relationship is determined based on the target entity, and a public opinion field knowledge graph corresponding to the public opinion field is generated based on the attribute information of the target entity and the ontology corresponding to the public opinion field corresponding to each public opinion keyword; Generate a public opinion knowledge base according to each public opinion field knowledge graph, so that the public opinion knowledge base combines the public opinion field knowledge graphs of multiple public opinion fields; A public opinion consultation platform is constructed based on the public opinion knowledge base.

10. A computer-readable storage medium storing computer instructions, which, when executed by a processor, implement the steps of the knowledge base construction method described in any one of claims 1 to 7.

Citation Information

Patent Citations

  • Public opinion association analysis method and system based on knowledge graph

    CN110413784A

  • Automobile public opinion analysis method based on knowledge graph and deep learning

    CN110633373A

  • College online public opinion risk assessment method

    CN110866126A

  • Industry process field knowledge graph construction method and device

    CN111444351A

  • Public opinion analysis method and device, computer equipment and storage medium

    CN111611408A