CN114428864A - Knowledge graph construction method and device, electronic equipment and medium - Google Patents
Knowledge graph construction method and device, electronic equipment and medium Download PDFInfo
- Publication number
- CN114428864A CN114428864A CN202210337302.2A CN202210337302A CN114428864A CN 114428864 A CN114428864 A CN 114428864A CN 202210337302 A CN202210337302 A CN 202210337302A CN 114428864 A CN114428864 A CN 114428864A
- Authority
- CN
- China
- Prior art keywords
- graph
- enterprise
- entity
- knowledge graph
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000010276 construction Methods 0.000 title abstract description 18
- 238000000034 method Methods 0.000 claims abstract description 66
- 238000000605 extraction Methods 0.000 claims description 33
- 238000001228 spectrum Methods 0.000 claims description 30
- 238000012545 processing Methods 0.000 claims description 25
- 238000011144 upstream manufacturing Methods 0.000 claims description 25
- 238000005065 mining Methods 0.000 claims description 23
- 230000004927 fusion Effects 0.000 claims description 20
- 238000013441 quality evaluation Methods 0.000 claims description 18
- 238000000354 decomposition reaction Methods 0.000 claims description 13
- 239000013598 vector Substances 0.000 claims description 8
- 238000013136 deep learning model Methods 0.000 claims description 5
- 230000009466 transformation Effects 0.000 claims description 4
- 238000005516 engineering process Methods 0.000 abstract description 11
- 238000004519 manufacturing process Methods 0.000 abstract description 5
- 230000008520 organization Effects 0.000 description 11
- 230000008569 process Effects 0.000 description 10
- 238000010586 diagram Methods 0.000 description 6
- 239000003795 chemical substances by application Substances 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000001149 cognitive effect Effects 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000019771 cognition Effects 0.000 description 2
- 230000005484 gravity Effects 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 239000002994 raw material Substances 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000004931 aggregating effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 229940003372 compro Drugs 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000008094 contradictory effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- WIKYUJGCLQQFNW-UHFFFAOYSA-N prochlorperazine Chemical compound C1CN(C)CCN1CCCN1C2=CC(Cl)=CC=C2SC2=CC=CC=C21 WIKYUJGCLQQFNW-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/08—Auctions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2216/00—Indexing scheme relating to additional aspects of information retrieval not explicitly covered by G06F16/00 and subgroups
- G06F2216/03—Data mining
Abstract
The application discloses a method and a device for constructing a knowledge graph, electronic equipment and a medium. By applying the technical scheme of the application, a knowledge graph which can be applied to all traditional manufacturing industries can be constructed, wherein the knowledge graph comprises the bidding data, the bid winning data, the contact information data and the graph data of the enterprise operation range data, so that the problems of low identification accuracy caused by fine granularity and narrow coverage of the existing industry graph construction method in the related technology are solved.
Description
Technical Field
The present application relates to data processing technologies, and in particular, to a method and an apparatus for constructing a knowledge graph, an electronic device, and a medium.
Background
The rise of the big data era and the fact that the research center of gravity of artificial intelligence is transited from perception intelligence to cognition intelligence push the enthusiasm of the knowledge graph.
The knowledge graph is a semantic network for describing knowledge and modeling relations by using an image model and serves as a bottom support of cognitive intelligence, and the industry knowledge graph has great significance for upgrading AI enabling traditional industries and plays an important role in landing and industrial intelligence of the traditional industries.
However, the existing industry map construction method is fine in granularity, so that the coverage is narrow, and multi-scenario enterprise services cannot be processed.
Disclosure of Invention
The embodiment of the application provides a method and a device for constructing a knowledge graph, electronic equipment and a medium. The method is used for solving the problems that the existing industry map construction method is fine in granularity and narrow in coverage area, and accordingly recognition accuracy is low in the related technology.
According to an aspect of the embodiments of the present application, a method for constructing a knowledge graph is provided, including:
constructing an ontology graph, comprising: constructing an ontology graph comprising enterprise bidding contacts, upstream and downstream of enterprises, enterprise contact ways and industry-enterprise-product sub-graph spectrum layers which take an industrial chain as a guide;
entity extraction is carried out on multiple data sources by utilizing multiple entity identification models to obtain a sample entity set, wherein the multiple data sources are at least one of bid inviting data, bid winning data, contact information data and enterprise operation range data corresponding to enterprise information;
inputting the sample entity set into the ontology graph to obtain an initial knowledge graph;
and carrying out knowledge fusion on the entity information in the initial knowledge graph to obtain a to-be-mined knowledge graph, and carrying out entity correlation mining on the to-be-mined knowledge graph to obtain a target knowledge graph for processing enterprise services.
Optionally, in another embodiment based on the above method of the present application, the constructing the ontology graph includes:
constructing a sub-graph spectrum layer for reflecting the incidence relation between the user class body and the corresponding enterprise class body, wherein the user class body comprises enterprise bidding contacts, and the enterprise class body comprises bidding enterprises, bidding organizations and agencies; and the number of the first and second groups,
constructing a sub-graph spectrum layer for reflecting the incidence relation among the bidding enterprise ontologies, wherein the bidding enterprise ontologies comprise bidding enterprises/organizations, agencies, bidding enterprises and bidding enterprises; and the number of the first and second groups,
constructing a sub-graph spectrum layer for each enterprise contact way; and (c) a second step of,
and constructing a sub-graph spectrum layer for the incidence relation among the enterprise class body, the industry class body and the corresponding product class body, wherein the product class body comprises an upstream product body, a midstream product body and a downstream product body, and the industry class body comprises an upstream industry, a midstream industry and a downstream industry.
Optionally, in another embodiment based on the above method of the present application, the performing entity extraction for multiple data sources by using multiple entity identification models to obtain a sample entity set includes:
if the data source corresponds to the bid inviting data and the bid winning data of the enterprise information, extracting a bid inviting enterprise, a bid bidding enterprise, a bid winning enterprise and an agency included in the data source by using an enterprise entity extraction model; extracting bidding contact users, bidding project responsible users and agency contact users by using a contact person information extraction model to obtain the sample entity set;
and/or the presence of a gas in the gas,
if the data source corresponds to the enterprise operation range data of the enterprise information, extracting enterprise entities included in the data source and a contact way corresponding to the enterprise by using an enterprise contact information extraction model to obtain a sample entity set;
and/or the presence of a gas in the gas,
and if the data source corresponds to the contact information data of the enterprise information, extracting the product entities included in the data source by using a deep learning model to obtain the sample entity set.
Optionally, in another embodiment based on the above method of the present application, the inputting the sample entity set into the ontology graph to obtain an initial knowledge-graph includes:
and inputting the sample data in the sample entity set into a structure corresponding to the ontology graph according to the ontology structure of the ontology graph to obtain the initial knowledge graph.
Optionally, in another embodiment based on the foregoing method of the present application, after the obtaining the initial knowledge-graph, the method further includes:
and performing quality evaluation on the initial knowledge graph, and performing knowledge fusion on entity information in the initial knowledge graph after the initial knowledge graph passes the quality evaluation, wherein the quality evaluation corresponds to determining the relationship among the entity information and the attribute of the entity information.
Optionally, in another embodiment based on the foregoing method of the present application, the performing knowledge fusion on the entity information in the initial knowledge graph to obtain a to-be-mined knowledge graph includes:
and carrying out entity decomposition and entity combination on the entity information in the initial knowledge graph to obtain the knowledge graph to be mined, wherein the entity decomposition is used for processing the screened entity information with the same name but different meanings, and the entity combination is used for processing the screened entity information with the same meaning but different names.
Optionally, in another embodiment based on the foregoing method of the present application, after the mining of the entity interrelation of the knowledge graph to be mined, the obtaining of the target knowledge graph for processing the enterprise service includes:
and acquiring original relations among the entities, and performing vector transformation on the entities and the original relations among the entities by using a graph model to realize entity mutual relation mining on the knowledge graph to be mined, so as to obtain the target knowledge graph.
According to another aspect of the embodiments of the present application, there is provided an apparatus for constructing a knowledge graph, including:
a build module configured to build an ontology graph, comprising: constructing an ontology graph comprising enterprise bidding contacts, upstream and downstream of enterprises, enterprise contact ways and industry-enterprise-product sub-graph spectrum layers which take an industrial chain as a guide;
the generating module is configured to perform entity extraction on multiple data sources by using multiple entity identification models to obtain a sample entity set, wherein the multiple data sources are at least one of bid inviting data, bid winning data, contact information data and enterprise operation range data corresponding to enterprise information;
an input module configured to input the sample entity set into the ontology graph, resulting in an initial knowledge-graph;
the generating module is configured to perform knowledge fusion on entity information in the initial knowledge graph to obtain a to-be-mined knowledge graph, and obtain a target knowledge graph for processing enterprise business after performing entity mutual relation mining on the to-be-mined knowledge graph.
According to another aspect of the embodiments of the present application, there is provided an electronic device including:
a memory for storing executable instructions; and
a display for communicating with the memory to execute the executable instructions to perform any of the above-described methods of constructing a knowledge-graph.
According to a further aspect of the embodiments of the present application, there is provided a computer-readable storage medium for storing computer-readable instructions, which, when executed, perform the operations of any one of the above-mentioned methods for constructing a knowledge graph.
According to the method, an ontology graph comprising enterprise bidding contacts, upstream and downstream of enterprises, enterprise contact ways and industry-enterprise-product sub-graph spectrum layers which are oriented by an industrial chain can be constructed, and entity extraction is performed on multiple data sources by utilizing multiple entity recognition models to obtain a sample entity set; inputting the sample entity set into the ontology graph to obtain an initial knowledge graph; and carrying out knowledge fusion on entity information in the initial knowledge graph to obtain a to-be-mined knowledge graph, and carrying out entity correlation mining on the to-be-mined knowledge graph to obtain a target knowledge graph for processing enterprise services. By applying the technical scheme of the application, a knowledge graph which can be applied to all traditional manufacturing industries can be constructed, wherein the knowledge graph comprises the bidding data, the bid winning data, the contact information data and the graph data of the enterprise operation range data, so that the problems of low identification accuracy caused by fine granularity and narrow coverage of the existing industry graph construction method in the related technology are solved.
The technical solution of the present application is further described in detail by the accompanying drawings and examples.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description, serve to explain the principles of the application.
The present application may be more clearly understood from the following detailed description with reference to the accompanying drawings, in which:
FIG. 1 is a schematic diagram of a method for constructing a knowledge graph according to the present application;
FIGS. 2-3 are schematic diagrams of the overall architecture of the knowledge-graph proposed in the present application;
FIG. 4 is a schematic diagram of an electronic apparatus for constructing a knowledge graph according to the present application;
fig. 5 is a schematic structural diagram of an electronic device for constructing a knowledge graph according to the present application.
Detailed Description
Various exemplary embodiments of the present application will now be described in detail with reference to the accompanying drawings. It should be noted that: the relative arrangement of the components and steps, the numerical expressions, and numerical values set forth in these embodiments do not limit the scope of the present application unless specifically stated otherwise.
Meanwhile, it should be understood that the sizes of the respective portions shown in the drawings are not drawn in an actual proportional relationship for the convenience of description.
The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the application, its application, or uses.
Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail but are intended to be part of the specification where appropriate.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be discussed further in subsequent figures.
In addition, technical solutions between the various embodiments of the present application may be combined with each other, but it must be based on the realization of the technical solutions by a person skilled in the art, and when the technical solutions are contradictory or cannot be realized, such a combination of technical solutions should be considered to be absent and not within the protection scope of the present application.
It should be noted that all the directional indicators (such as upper, lower, left, right, front and rear … …) in the embodiment of the present application are only used to explain the relative position relationship between the components, the motion situation, etc. in a specific posture (as shown in the drawings), and if the specific posture is changed, the directional indicator is changed accordingly.
A method for constructing a knowledge graph according to an exemplary embodiment of the present application is described below with reference to fig. 1 to 3. It should be noted that the following application scenarios are merely illustrated for facilitating understanding of the spirit and principles of the present application, and the embodiments of the present application are not limited in any way in this respect. Rather, embodiments of the present application may be applied to any scenario where applicable.
The application also provides a method and a device for constructing the knowledge graph, electronic equipment and a medium.
Fig. 1 schematically shows a flow chart of a method for constructing a knowledge graph according to an embodiment of the present application. As shown in fig. 1, the method includes:
s101, constructing an ontology graph, comprising: and constructing an ontology graph comprising enterprise bidding contacts, upstream and downstream of the enterprise, enterprise contact ways and industry-enterprise-product sub-graph spectrum layers oriented by an industrial chain.
And S102, performing entity extraction on multiple data sources by using multiple entity identification models to obtain a sample entity set, wherein the multiple data sources are at least one of bid inviting data, bid winning data, contact information data and enterprise business range data corresponding to enterprise information.
S103, inputting the sample entity set into the ontology graph to obtain an initial knowledge graph.
And S104, carrying out knowledge fusion on entity information in the initial knowledge graph to obtain a knowledge graph to be mined, and carrying out entity correlation mining on the knowledge graph to be mined to obtain a target knowledge graph for processing enterprise services.
In the related art, the rise of the big data era and the fact that the research center of gravity of artificial intelligence is transited from perception intelligence to cognition intelligence push the enthusiasm of the knowledge graph. The knowledge graph is a semantic network for describing knowledge and modeling relations by using an image model and serves as a bottom support of cognitive intelligence, and the industry knowledge graph has great significance for upgrading AI enabling the traditional industry and plays an important role in landing and industrial intelligence of the traditional industry.
The knowledge graph can convert massive structured and unstructured data into a mesh knowledge structure, nodes in the network represent entities (entities) or concepts (concepts), edges represent various real semantic relations between the entities and the concepts, the mapping from the big data to the entities, the concepts and the relations of the knowledge graph is realized in the process, the converted structured knowledge endows a machine with the capability of understanding the data, and the machine can learn high-precision knowledge of a designated industry based on the machine.
In addition, the relationship in the knowledge graph has great value, and the machine can be endowed with cognitive intelligent reasoning ability by combining logic rules, statistics and deep learning to reason out the implicit relationship between entities or concepts. And dividing according to the coverage of the knowledge graph, including a general knowledge graph and an industry knowledge graph. The industry knowledge graph is started later, the requirement on accuracy is higher, a strict data mode with industry significance is provided, and the construction of a mode layer needs a large amount of industry knowledge, so that expert assistance in the industry is needed in the initial construction stage, and the industry knowledge graph in most fields belongs to the exploration stage at present.
Most of the existing technologies deposited by the relevant scholars aim at a specific field, for example, knowledge maps are constructed aiming at the financial field, or coarse-grained methods without considering the field knowledge characteristics, for example, a generalized knowledge map construction method is provided, and the knowledge maps cannot be supported in the business enabling the traditional industry.
Further, as shown in fig. 2 to fig. 3, first, ontology modeling may be performed on a mode layer of the industry-wide graph to obtain an ontology graph. Secondly, based on the ontology graph, a plurality of entity identification models are collected in a data-driven mode to extract entities aiming at a plurality of data sources to complete entity expansion, and the parallel relation and the upper-lower relation of the entities are obtained simultaneously. And then, completing multi-source knowledge fusion by utilizing an entity decomposition and entity combination technology.
And finally, the quality of the atlas is ensured by giving up the knowledge with lower confidence coefficient, and the process is a quality evaluation stage. And finally, obtaining an initial knowledge graph, and carrying out graph calculation and mining on new relations based on the initial knowledge graph in the follow-up process to finish knowledge reasoning and mining of the graph. And performing quality evaluation on the new inferred relation again, and incrementally updating the initial knowledge graph by using the screened relation with higher quality to finally obtain the target knowledge graph.
According to the method, an ontology graph comprising enterprise bidding contacts, upstream and downstream of enterprises, enterprise contact ways and industry-enterprise-product sub-graph spectrum layers which are oriented by an industrial chain can be constructed, and entity extraction is performed on multiple data sources by utilizing multiple entity recognition models to obtain a sample entity set; inputting the sample entity set into the ontology graph to obtain an initial knowledge graph; and carrying out knowledge fusion on entity information in the initial knowledge graph to obtain a to-be-mined knowledge graph, and carrying out entity correlation mining on the to-be-mined knowledge graph to obtain a target knowledge graph for processing enterprise services. By applying the technical scheme of the application, a knowledge graph which can be applied to all traditional manufacturing industries can be constructed, wherein the knowledge graph comprises the bidding data, the bid winning data, the contact information data and the graph data of the enterprise operation range data, so that the problems of low identification accuracy caused by fine granularity and narrow coverage of the existing industry graph construction method in the related technology are solved.
Optionally, in another embodiment based on the above method of the present application, the constructing the ontology graph includes:
constructing a sub-graph spectrum layer for reflecting the incidence relation between the user class body and the corresponding enterprise class body, wherein the user class body comprises enterprise bidding contacts, and the enterprise class body comprises bidding enterprises, bidding organizations and agencies; and the number of the first and second groups,
constructing a sub-graph spectrum layer for reflecting the incidence relation among the bidding enterprise ontologies, wherein the bidding enterprise ontologies comprise bidding enterprises/organizations, agencies, bidding enterprises and bidding enterprises; and (c) a second step of,
constructing a sub-graph spectrum layer for each enterprise contact way; and the number of the first and second groups,
and constructing a sub-graph spectrum layer for the incidence relation among the enterprise class body, the industry class body and the corresponding product class body, wherein the product class body comprises an upstream product body, a midstream product body and a downstream product body, and the industry class body comprises an upstream industry, a midstream industry and a downstream industry.
Optionally, in another embodiment based on the above method of the present application, the performing entity extraction for multiple data sources by using multiple entity identification models to obtain a sample entity set includes:
if the data source corresponds to the bid inviting data and the bid winning data of the enterprise information, extracting a bid inviting enterprise, a bid bidding enterprise, a bid winning enterprise and an agency included in the data source by using an enterprise entity extraction model; extracting bidding contact users, bidding project responsible users and agency contact users by using a contact person information extraction model to obtain the sample entity set;
and/or the presence of a gas in the gas,
if the data source corresponds to the enterprise operation range data of the enterprise information, extracting enterprise entities included in the data source and a contact way corresponding to the enterprise by using an enterprise contact information extraction model to obtain a sample entity set;
and/or the presence of a gas in the gas,
and if the data source corresponds to the contact information data of the enterprise information, extracting the product entities included in the data source by using a deep learning model to obtain the sample entity set.
Optionally, in another embodiment based on the above method of the present application, the inputting the sample entity set into the ontology graph to obtain an initial knowledge-graph includes:
and inputting the sample data in the sample entity set into a structure corresponding to the ontology graph according to the ontology structure of the ontology graph to obtain the initial knowledge graph.
Optionally, in another embodiment based on the foregoing method of the present application, after the obtaining the initial knowledge-graph, the method further includes:
and performing quality evaluation on the initial knowledge graph, and performing knowledge fusion on entity information in the initial knowledge graph after the initial knowledge graph passes the quality evaluation, wherein the quality evaluation corresponds to determining the relationship among the entity information and the attribute of the entity information.
Optionally, in another embodiment of the method based on the present application, the performing knowledge fusion on the entity information in the initial knowledge graph to obtain a knowledge graph to be mined includes:
and carrying out entity decomposition and entity combination on the entity information in the initial knowledge graph to obtain the knowledge graph to be mined, wherein the entity decomposition is used for processing the screened entity information with the same name but different meanings, and the entity combination is used for processing the screened entity information with the same meaning but different names.
Optionally, in another embodiment based on the foregoing method of the present application, after the mining of the entity interrelation of the knowledge graph to be mined, the obtaining of the target knowledge graph for processing the enterprise service includes:
and acquiring original relations among the entities, and performing vector conversion on the entities and the original relations among the entities by using a graph model to realize entity mutual relation mining on the knowledge graph to be mined, so as to obtain the target knowledge graph.
Further, for constructing the ontology graph in the present application, the following steps may be included:
in one mode, the ontology graph modeling method comprises four parts, namely: the method comprises the steps of establishing an ontology of a bidding contact person sub-map layer, establishing an ontology of an upstream and downstream sub-map layer, establishing an ontology of an enterprise contact mode sub-map layer, and establishing an ontology of an industry-enterprise-product sub-map layer by taking an industry chain as a guide.
For the ontology construction of the bidding contact person subgraph spectrum layer, the ontology construction can include three enterprise/organization class ontologies: namely bidding enterprises, bidding organizations, agencies. And, three classes of human ontology are included: namely a bid inviting item responsible person, a bid inviting contact person and an agent contact person. And, including two business attributes: the system comprises a telephone and a mailbox, wherein the attributes are attached to three ontologies, namely a bidding project responsible person, a bidding contact person and an agent contact person. It should be noted that the sub-graph contains a type of relationship, contact _ in, which indicates that the contact user is at the business, and associates the human ontology class and the business/organization ontology class.
The construction of the upstream and downstream sub-map ontology for bidding can include four types of ontologies of a bidding enterprise/organization, an agency, a bidding enterprise and a bid winning enterprise. The four types of relationships are respectively: the tb relation of the bidding enterprise pointing to the bidding enterprise/organization indicates that the bidding enterprise has participated in the bidding project of the bidding enterprise/organization, the tw relation of the bidding enterprise pointing to the bidding enterprise/organization indicates that the bidding enterprise has participated in the bidding project of the bidding enterprise/organization and bids, the ta relation of the agency pointing to the bidding enterprise/organization indicates that the agency has proxied the bidding project of the bidding enterprise/organization, and the th relation between the bidding enterprises indicates that two bidding enterprises have jointly bid on the bidding project of the same bidding enterprise/organization.
For the construction of the enterprise contact way subgraph spectrum, the enterprise contact way subgraph spectrum can comprise an upstream enterprise ontology, a midstream enterprise ontology, a downstream enterprise ontology, a bidding enterprise/organization ontology, an agency ontology, a bidding enterprise ontology and a winning enterprise ontology, wherein each ontology has five attributes of a telephone, a mailbox, a QQ, a wechat Website and a website official website.
For the construction of the industry-enterprise-product sub-graph spectrum spanning three fields guided by an industry chain, the enterprise type body is divided into an upstream enterprise body, a midstream enterprise body and a downstream enterprise body according to the industry chain, the product type body is correspondingly divided into an upstream product body, a midstream product body and a downstream product body, and the industry type body comprises an upstream industry, a midstream industry and a downstream industry. The relationship includes four categories, which are: the com _ up _ down relation represents the upstream and downstream relation of the enterprise in the industrial chain; product relation, which is used for describing the fact that the enterprise and the product of the enterprise operation range are in fact; material relationships, representing raw material relationships, e.g., product A directed to product B, representing a raw material for which A is B; the pro _ indu relationship describes the fact that a specific product belongs to a specific industry; the com _ pro relationship indicates that the enterprise belongs to the directed industry.
Further, in the process of extracting entities from multiple data sources by using multiple entity identification models to obtain a sample entity set, the method may include:
first, the multiple data sources in the present application may be at least one of bid bidding data, bid winning data, contact information data, and enterprise business scope data corresponding to enterprise information. Further, in the process of extracting entities by using a plurality of entity identification models for a plurality of data sources to obtain a sample entity set, different model extraction entities can be used for different data sources. The device comprises at least one of the following four parts:
a first part:
and extracting the bidding enterprises, the bidding enterprises and the agencies by using the enterprise entity extraction model for the bidding data, and extracting the bidding contact users, the bidding project responsible users and the agencies to contact the users by using the contact person information extraction model to further obtain the sample entity set.
A second part:
and extracting bidding enterprises and agencies by using the enterprise entity extraction model consistent with the bidding data, and extracting contacts, bidding project responsible persons and bidding contacts of the agencies by using the contact person information extraction model consistent with the bidding data, so as to obtain the sample entity set.
And a third part:
if the data source corresponds to contact information data of the enterprise information, such as an enterprise official network, an e-commerce platform and the like, the enterprise entity and attributes of a telephone, a mailbox, a QQ, a web Wechat, a website and the like corresponding to the enterprise are extracted by using an enterprise contact information extraction model, and then the sample entity set is obtained.
The fourth part:
if the data source corresponds to enterprise operation range data of enterprise information, the enterprise operation range data are enterprise official nets, hundred-degree encyclopedia brief introduction, all E-commerce platforms, vertical industry sites, financial reports and the like, sentences containing operation ranges are extracted by using the deep learning model, then product entities contained in the sentences are extracted, and further the sample entity set is obtained.
Further, if it is determined that the entity extraction is completed, the defined relationship in the ontology structure can be specified for the corresponding entity according to the modeled ontology structure to obtain an initial knowledge graph.
In one mode, after the initial knowledge graph is obtained, the quality of the constructed knowledge graph needs to be evaluated, and the quality of the knowledge graph is ensured by discarding the knowledge with lower confidence coefficient.
Further, the process of the quality evaluation of the knowledge graph in the application mainly comprises the following three types: the upper and lower relations of the entity information error, the entity information attribute have deviation, and the entity information relation has logic error.
For example, if business entity A does not have a bid relationship ring directed to itself, if A agent B does not have the possibility of B agent A, and both A and B have a relationship with phone p, then it is likely that there is not a relationship. In the quality evaluation process, besides a rule reasoning method, the relationship with lower confidence coefficient can be abandoned through the confidence coefficient of each entity relationship, and the relationship with high confidence coefficient is reserved.
In addition, because the industry knowledge graph needs to support scenes such as data decision and the like, the quality of the knowledge graph has high requirements, so that the quality evaluation stage plays an important role in the construction process of the industry knowledge graph, the quality evaluation is needed every time the graph data expansion is carried out, the quality of the control graph can be strictly controlled to ensure that the knowledge graph falls on the ground in industry applications such as intelligent decision analysis based on big data, AI retrieval based on the knowledge graph, intelligent recommendation and the like, and the realization of enabling the traditional industry is really realized.
In one mode, in the process of performing knowledge fusion on entity information in an initial knowledge graph to obtain a to-be-mined knowledge graph, the method may include performing entity decomposition and entity merging on the entity information in the initial knowledge graph, specifically:
the scenario of entity decomposition is that a named entity expresses different meanings in different contexts, entities with the same name and different synonyms can be understood as correct meanings to be linked to correct entities by using an entity decomposition technology, for example, "apple" can represent a fruit and can also represent a technology company, and the entity decomposition can link the named entity to correct entities according to the context meanings.
The entity merging refers to aggregating entities having a common reference relationship in a text into one entity, for example, "cross split refrigerator", "double-open four-door refrigerator", and "cross four-door refrigerator" refer to "cross split four-door refrigerator", and the "cross split refrigerator", "double-open four-door refrigerator", and "cross four-door refrigerator" need to be merged into the entity "cross split four-door refrigerator" by using rules and corresponding entity merging algorithms. Thus, business entities and product entities from multiple data sources can be fused.
In one mode, in the process of obtaining a target knowledge graph for processing enterprise services after mining the entity interrelation of the knowledge graph to be mined, the method may include:
and acquiring original relations among the entities, and performing vector transformation on the entities and the original relations among the entities by using a graph model to realize entity mutual relation mining on the knowledge graph to be mined, so as to obtain the target knowledge graph.
Specifically, if there is a tb (bid) relationship with Business entity A pointing to Business entity C and a tb (bid) relationship with Business entity B pointing to Business entity C, then the fact that there is a th (in-line) relationship between Business entity A and Business entity B can be inferred. Such as the tw (bid) relationship that business entity a points to business entity B, the fact that business entity a points to the tb (bid) relationship of business entity B can be inferred.
In addition, an inference method based on knowledge graph representation learning, such as TransE, can be used, a graph model can be used for relationship completion, such as GCN and RGCN, the method represents the entities and the relationships as vectors, the calculation between the vectors replaces graph traversal and search to predict the relationship existing between the two entities (namely, the existence of a prediction triple), and the method has the advantages that the vector representation of the entities and the relationship contains the original semantic information of the entities, and meanwhile, the graph structure is used for completing knowledge inference and mining.
By applying the technical scheme of the application, a knowledge graph which can be applied to all traditional manufacturing industries can be constructed, wherein the knowledge graph comprises the bidding data, the bid winning data, the contact information data and the graph data of the enterprise operation range data, so that the problems of low identification accuracy caused by fine granularity and narrow coverage of the existing industry graph construction method in the related technology are solved.
Optionally, in another embodiment of the present application, as shown in fig. 4, the present application further provides a knowledge graph constructing apparatus. Which comprises the following steps:
a building module 201 configured to build an ontology graph, comprising: constructing an ontology graph comprising enterprise bidding contacts, upstream and downstream of enterprises, enterprise contact ways and industry-enterprise-product sub-graph spectrum layers which take an industrial chain as a guide;
the generating module 202 is configured to perform entity extraction on multiple data sources by using multiple entity identification models to obtain a sample entity set, where the multiple data sources are at least one of bid inviting data, bid winning data, contact information data and enterprise operation range data corresponding to enterprise information;
an input module 203 configured to input the sample entity set into the ontology graph, resulting in an initial knowledge-graph;
the generating module 202 is configured to perform knowledge fusion on the entity information in the initial knowledge graph to obtain a to-be-mined knowledge graph, and obtain a target knowledge graph for processing enterprise services after performing entity correlation mining on the to-be-mined knowledge graph.
According to the method, after the ontology graph can be constructed, entity extraction is carried out on multiple data sources by using multiple entity identification models to obtain a sample entity set; inputting the sample entity set into the ontology graph to obtain an initial knowledge graph; and carrying out knowledge fusion on entity information in the initial knowledge graph to obtain a knowledge graph to be mined, and carrying out entity correlation mining on the knowledge graph to be mined to obtain a target knowledge graph for processing enterprise services. By applying the technical scheme of the application, a knowledge graph which can be applied to all traditional manufacturing industries can be constructed, wherein the knowledge graph comprises the bidding data, the bid winning data, the contact information data and the graph data of the enterprise operation range data, so that the problems of low identification accuracy caused by fine granularity and narrow coverage of the existing industry graph construction method in the related technology are solved.
In another embodiment of the present application, the building module 201 is configured to perform the steps including:
constructing a sub-graph spectrum layer for reflecting the incidence relation between the user class body and the corresponding enterprise class body, wherein the user class body comprises enterprise bidding contacts, and the enterprise class body comprises bidding enterprises, bidding organizations and agencies; and the number of the first and second groups,
constructing a sub-graph spectrum layer for reflecting the incidence relation among the bidding enterprise ontologies, wherein the bidding enterprise ontologies comprise bidding enterprises/organizations, agency organizations, bidding enterprises and bid winning enterprises; and the number of the first and second groups,
constructing a sub-graph spectrum layer for each enterprise contact way; and (c) a second step of,
and constructing a sub-graph spectrum layer for the incidence relation among the enterprise class body, the industry class body and the corresponding product class body, wherein the product class body comprises an upstream product body, a midstream product body and a downstream product body, and the industry class body comprises an upstream industry, a midstream industry and a downstream industry.
In another embodiment of the present application, the building module 201 is configured to perform the steps including:
if the data source corresponds to the bid inviting data and the bid winning data of the enterprise information, extracting a bid inviting enterprise, a bid bidding enterprise, a bid winning enterprise and an agency included in the data source by using an enterprise entity extraction model; extracting bidding contact users, bidding project responsible users and agency contact users by using a contact person information extraction model to obtain the sample entity set;
and/or the presence of a gas in the gas,
if the data source corresponds to the enterprise operation range data of the enterprise information, extracting enterprise entities included in the data source and a contact way corresponding to the enterprise by using an enterprise contact information extraction model to obtain a sample entity set;
and/or the presence of a gas in the gas,
and if the data source corresponds to the contact information data of the enterprise information, extracting the product entities included in the data source by using a deep learning model to obtain the sample entity set.
In another embodiment of the present application, the building module 201 is configured to perform the steps including:
and inputting the sample data in the sample entity set into a structure corresponding to the ontology graph according to the ontology structure of the ontology graph to obtain the initial knowledge graph.
In another embodiment of the present application, the building module 201 is configured to perform the steps including:
and performing quality evaluation on the initial knowledge graph, and performing knowledge fusion on entity information in the initial knowledge graph after the initial knowledge graph passes the quality evaluation, wherein the quality evaluation corresponds to determining the relationship among the entity information and the attribute of the entity information.
In another embodiment of the present application, the building module 201 is configured to perform the steps including:
and carrying out entity decomposition and entity combination on the entity information in the initial knowledge graph to obtain the knowledge graph to be mined, wherein the entity decomposition is used for processing the screened entity information with the same name but different meanings, and the entity combination is used for processing the screened entity information with the same meaning but different names.
In another embodiment of the present application, the building module 201 is configured to perform the steps including:
and acquiring original relations among the entities, and performing vector transformation on the entities and the original relations among the entities by using a graph model to realize entity mutual relation mining on the knowledge graph to be mined, so as to obtain the target knowledge graph.
Fig. 5 is a block diagram illustrating a logical structure of an electronic device in accordance with an exemplary embodiment. For example, the electronic device 300 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, an exercise device, a personal digital assistant, and the like.
In an exemplary embodiment, there is also provided a non-transitory computer readable storage medium, such as a memory, including instructions executable by a processor of an electronic device to perform a method of constructing a knowledge-graph as described above, the method including: constructing an ontology graph, comprising: constructing an ontology graph comprising enterprise bidding contacts, upstream and downstream enterprises, enterprise contact ways and industry-enterprise-product sub-graph spectrum layers guided by an industry chain; entity extraction is carried out on multiple data sources by utilizing multiple entity identification models to obtain a sample entity set, wherein the multiple data sources are at least one of bid inviting data, bid winning data, contact information data and enterprise operation range data corresponding to enterprise information; inputting the sample entity set into the ontology graph to obtain an initial knowledge graph; and carrying out knowledge fusion on the entity information in the initial knowledge graph to obtain a to-be-mined knowledge graph, and carrying out entity correlation mining on the to-be-mined knowledge graph to obtain a target knowledge graph for processing enterprise services. Optionally, the instructions may also be executable by a processor of the electronic device to perform other steps involved in the exemplary embodiments described above. For example, the non-transitory computer readable storage medium may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.
In an exemplary embodiment, there is also provided an application/computer program product including one or more instructions executable by a processor of an electronic device to perform the above method of constructing a knowledge-graph, the method comprising: constructing an ontology graph, comprising: constructing an ontology graph comprising enterprise bidding contacts, upstream and downstream of enterprises, enterprise contact ways and industry-enterprise-product sub-graph spectrum layers which take an industrial chain as a guide; entity extraction is carried out on multiple data sources by utilizing multiple entity identification models to obtain a sample entity set, wherein the multiple data sources are at least one of bid inviting data, bid winning data, contact information data and enterprise operation range data corresponding to enterprise information; inputting the sample entity set into the ontology graph to obtain an initial knowledge graph; and carrying out knowledge fusion on entity information in the initial knowledge graph to obtain a knowledge graph to be mined, and carrying out entity correlation mining on the knowledge graph to be mined to obtain a target knowledge graph for processing enterprise services. Optionally, the instructions may also be executable by a processor of the electronic device to perform other steps involved in the exemplary embodiments described above.
Those skilled in the art will appreciate that the schematic diagram 5 is merely an example of the electronic device 300 and does not constitute a limitation of the electronic device 300 and may include more or less components than those shown, or combine certain components, or different components, e.g., the electronic device 300 may also include input-output devices, network access devices, buses, etc.
The Processor 302 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. The general purpose processor may be a microprocessor or the processor 302 may be any conventional processor or the like, and the processor 302 is the control center of the electronic device 300 and connects the various parts of the entire electronic device 300 using various interfaces and lines.
The memory 301 may be used to store computer readable instructions and the processor 302 may implement various functions of the electronic device 300 by executing or executing computer readable instructions or modules stored in the memory 301 and by invoking data stored in the memory 301. The memory 301 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data created according to the use of the electronic device 300, and the like. In addition, the Memory 301 may include a hard disk, a Memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Memory Card (Flash Card), at least one disk storage device, a Flash Memory device, a Read-Only Memory (ROM), a Random Access Memory (RAM), or other non-volatile/volatile storage devices.
The modules integrated by the electronic device 300 may be stored in a computer-readable storage medium if they are implemented in the form of software functional modules and sold or used as separate products. Based on such understanding, all or part of the flow in the method of the embodiments described above can be realized by the present application, and can also be realized by hardware related to computer readable instructions, which can be stored in a computer readable storage medium, and when the computer readable instructions are executed by a processor, the steps of the above described method embodiments can be realized.
Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.
It will be understood that the present application is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the application is limited only by the appended claims.
Claims (10)
1. A method for constructing a knowledge graph, comprising:
constructing an ontology graph, comprising: constructing an ontology graph comprising enterprise bidding contacts, upstream and downstream of enterprises, enterprise contact ways and industry-enterprise-product sub-graph spectrum layers which take an industrial chain as a guide;
entity extraction is carried out on multiple data sources by utilizing multiple entity identification models to obtain a sample entity set, wherein the multiple data sources are at least one of bid inviting data, bid winning data, contact information data and enterprise operation range data corresponding to enterprise information;
inputting the sample entity set into the ontology graph to obtain an initial knowledge graph;
and carrying out knowledge fusion on the entity information in the initial knowledge graph to obtain a to-be-mined knowledge graph, and carrying out entity correlation mining on the to-be-mined knowledge graph to obtain a target knowledge graph for processing enterprise services.
2. The method of claim 1, wherein the constructing the ontology graph comprises:
constructing a sub-graph spectrum layer for reflecting the incidence relation between the user class body and the corresponding enterprise class body, wherein the user class body comprises enterprise bidding contacts, and the enterprise class body comprises bidding enterprises, bidding organizations and agencies; and the number of the first and second groups,
constructing a sub-graph spectrum layer for reflecting the incidence relation among the bidding enterprise ontologies, wherein the bidding enterprise ontologies comprise bidding enterprises/organizations, agencies, bidding enterprises and bidding enterprises; and the number of the first and second groups,
constructing a sub-graph spectrum layer for each enterprise contact way; and the number of the first and second groups,
and constructing a sub-graph spectrum layer for the incidence relation among the enterprise class body, the industry class body and the corresponding product class body, wherein the product class body comprises an upstream product body, a midstream product body and a downstream product body, and the industry class body comprises an upstream industry, a midstream industry and a downstream industry.
3. The method of claim 1, wherein said utilizing a plurality of entity identification models for entity extraction for multiple data sources to obtain a sample entity set comprises:
if the data source corresponds to the bid inviting data and the bid winning data of the enterprise information, extracting a bid inviting enterprise, a bid bidding enterprise, a bid winning enterprise and an agency included in the data source by using an enterprise entity extraction model; extracting bidding contact users, bidding project responsible users and agency contact users by using a contact person information extraction model to obtain the sample entity set;
and/or the presence of a gas in the gas,
if the data source corresponds to the enterprise operation range data of the enterprise information, extracting enterprise entities included in the data source and a contact way corresponding to the enterprise by using an enterprise contact information extraction model to obtain a sample entity set;
and/or the presence of a gas in the gas,
and if the data source corresponds to the contact information data of the enterprise information, extracting the product entities included in the data source by using a deep learning model to obtain the sample entity set.
4. The method of claim 1 or 3, wherein the inputting the sample entity set into the ontology graph results in an initial knowledge-graph comprising:
and inputting the sample data in the sample entity set into a structure corresponding to the ontology graph according to the ontology structure of the ontology graph to obtain the initial knowledge graph.
5. The method of claim 4, after said obtaining the initial knowledge-graph, further comprising:
and performing quality evaluation on the initial knowledge graph, and performing knowledge fusion on entity information in the initial knowledge graph after the initial knowledge graph passes the quality evaluation, wherein the quality evaluation corresponds to determining the relationship among the entity information and the attribute of the entity information.
6. The method of claim 1, wherein the performing knowledge fusion on the entity information in the initial knowledge-graph to obtain a knowledge-graph to be mined comprises:
and carrying out entity decomposition and entity combination on the entity information in the initial knowledge graph to obtain the knowledge graph to be mined, wherein the entity decomposition is used for processing the screened entity information with the same name but different meanings, and the entity combination is used for processing the screened entity information with the same meaning but different names.
7. The method of claim 1, wherein the mining the mutual relationship of the entities of the to-be-mined knowledge graph to obtain a target knowledge graph for processing business of an enterprise comprises:
and acquiring original relations among the entities, and performing vector transformation on the entities and the original relations among the entities by using a graph model to realize entity mutual relation mining on the knowledge graph to be mined, so as to obtain the target knowledge graph.
8. An apparatus for constructing a knowledge graph, comprising:
a build module configured to build an ontology graph, comprising: constructing an ontology graph comprising enterprise bidding contacts, upstream and downstream of enterprises, enterprise contact ways and industry-enterprise-product sub-graph spectrum layers which take an industrial chain as a guide;
the generating module is configured to perform entity extraction on multiple data sources by using multiple entity identification models to obtain a sample entity set, wherein the multiple data sources are at least one of bid inviting data, bid winning data, contact information data and enterprise operation range data corresponding to enterprise information;
an input module configured to input the sample entity set into the ontology graph, resulting in an initial knowledge-graph;
the generating module is configured to perform knowledge fusion on entity information in the initial knowledge graph to obtain a to-be-mined knowledge graph, and obtain a target knowledge graph for processing enterprise business after performing entity mutual relation mining on the to-be-mined knowledge graph.
9. An electronic device, comprising:
a memory for storing executable instructions; and the number of the first and second groups,
a processor for executing the executable instructions with the memory to perform the operations of the method of constructing a knowledge-graph of any one of claims 1-7.
10. A computer-readable storage medium storing computer-readable instructions that, when executed, perform the operations of the method of constructing a knowledge graph of any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210337302.2A CN114428864A (en) | 2022-04-01 | 2022-04-01 | Knowledge graph construction method and device, electronic equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210337302.2A CN114428864A (en) | 2022-04-01 | 2022-04-01 | Knowledge graph construction method and device, electronic equipment and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114428864A true CN114428864A (en) | 2022-05-03 |
Family
ID=81314392
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210337302.2A Pending CN114428864A (en) | 2022-04-01 | 2022-04-01 | Knowledge graph construction method and device, electronic equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114428864A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116881430A (en) * | 2023-09-07 | 2023-10-13 | 北京上奇数字科技有限公司 | Industrial chain identification method and device, electronic equipment and readable storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109255034A (en) * | 2018-08-08 | 2019-01-22 | 数据地平线（广州）科技有限公司 | A kind of domain knowledge map construction method based on industrial chain |
| CN110737774A (en) * | 2018-07-03 | 2020-01-31 | 百度在线网络技术（北京）有限公司 | Book knowledge graph construction method, book recommendation method, device, equipment and medium |
| CN112836019A (en) * | 2021-02-19 | 2021-05-25 | 中国科学院新疆理化技术研究所 | Public health and public health named entity identification and entity linking method and device, electronic equipment and storage medium |
| CN113010688A (en) * | 2021-03-05 | 2021-06-22 | 北京信息科技大学 | Knowledge graph construction method, device and equipment and computer readable storage medium |
| CN113392229A (en) * | 2021-08-13 | 2021-09-14 | 四川新龟科技有限公司 | Supply chain relation construction and prediction method, device, equipment and storage medium |
| CN114254126A (en) * | 2021-12-21 | 2022-03-29 | 钛镕智能科技(苏州)有限公司 | Supply chain knowledge graph analysis method based on big data |
-
2022
- 2022-04-01 CN CN202210337302.2A patent/CN114428864A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110737774A (en) * | 2018-07-03 | 2020-01-31 | 百度在线网络技术（北京）有限公司 | Book knowledge graph construction method, book recommendation method, device, equipment and medium |
| CN109255034A (en) * | 2018-08-08 | 2019-01-22 | 数据地平线（广州）科技有限公司 | A kind of domain knowledge map construction method based on industrial chain |
| CN112836019A (en) * | 2021-02-19 | 2021-05-25 | 中国科学院新疆理化技术研究所 | Public health and public health named entity identification and entity linking method and device, electronic equipment and storage medium |
| CN113010688A (en) * | 2021-03-05 | 2021-06-22 | 北京信息科技大学 | Knowledge graph construction method, device and equipment and computer readable storage medium |
| CN113392229A (en) * | 2021-08-13 | 2021-09-14 | 四川新龟科技有限公司 | Supply chain relation construction and prediction method, device, equipment and storage medium |
| CN114254126A (en) * | 2021-12-21 | 2022-03-29 | 钛镕智能科技(苏州)有限公司 | Supply chain knowledge graph analysis method based on big data |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116881430A (en) * | 2023-09-07 | 2023-10-13 | 北京上奇数字科技有限公司 | Industrial chain identification method and device, electronic equipment and readable storage medium |
| CN116881430B (en) * | 2023-09-07 | 2023-12-12 | 北京上奇数字科技有限公司 | Industrial chain identification method and device, electronic equipment and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bharadiya | A comparative study of business intelligence and artificial intelligence with big data analytics | |
| US11775771B2 (en) | Enabling rhetorical analysis via the use of communicative discourse trees | |
| CN113990495B (en) | Disease diagnosis prediction system based on graph neural network | |
| Cioffi-Revilla et al. | Computation and social science | |
| CN111026842B (en) | Natural language processing method, natural language processing device and intelligent question-answering system | |
| CN112106056A (en) | Constructing fictitious utterance trees to improve the ability to answer convergent questions | |
| WO2021093821A1 (en) | Intelligent assistant evaluation and recommendation methods, system, terminal, and readable storage medium | |
| CN111222681A (en) | Data processing method, device, equipment and storage medium for enterprise bankruptcy risk prediction | |
| WO2019152426A1 (en) | Using communicative discourse trees to detect a request for an explanation | |
| CN110968701A (en) | Relationship map establishing method, device and equipment for graph neural network | |
| CN111538844B (en) | Target field knowledge base generation and problem solution method and device | |
| CN106407999A (en) | Rule combined machine learning method and system | |
| CN110674312B (en) | Method, device and medium for constructing knowledge graph and electronic equipment | |
| US20220253611A1 (en) | Techniques for maintaining rhetorical flow | |
| CN109299258A (en) | A kind of public sentiment event detecting method, device and equipment | |
| Seebacher | Predictive intelligence for data-driven managers | |
| CN113672599A (en) | Visual aid decision-making method for realizing government affair informatization project construction management by creating domain knowledge graph | |
| CN114428864A (en) | Knowledge graph construction method and device, electronic equipment and medium | |
| Kortum et al. | Dissection of AI job advertisements: A text mining-based analysis of employee skills in the disciplines computer vision and natural language processing | |
| Tan et al. | A model-based approach to generate dynamic synthetic test data: A conceptual model | |
| Liu et al. | Text classification in tourism and hospitality–a deep learning perspective | |
| Addi et al. | An ontology-based model for credit scoring knowledge in microfinance: Towards a better decision making | |
| CN115129863A (en) | Intention recognition method, device, equipment, storage medium and computer program product | |
| Vergara et al. | A Schematic Review of Knowledge Reasoning Approaches Based on the Knowledge Graph | |
| KR102630611B1 (en) | System for providing map based franchise foundation simulation service using market analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220503 |