CN112199516A - Method, device, terminal and storage medium for constructing knowledge graph - Google Patents
Method, device, terminal and storage medium for constructing knowledge graph Download PDFInfo
- Publication number
- CN112199516A CN112199516A CN202011289180.1A CN202011289180A CN112199516A CN 112199516 A CN112199516 A CN 112199516A CN 202011289180 A CN202011289180 A CN 202011289180A CN 112199516 A CN112199516 A CN 112199516A
- Authority
- CN
- China
- Prior art keywords
- target
- entity
- entities
- 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
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000012216 screening Methods 0.000 claims abstract description 40
- 238000010276 construction Methods 0.000 claims abstract description 32
- 238000007781 pre-processing Methods 0.000 claims abstract description 13
- 238000012545 processing Methods 0.000 claims description 13
- 230000011218 segmentation Effects 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 7
- 238000004891 communication Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 239000003086 colorant Substances 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000002699 waste material Substances 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
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses a method, a device, a terminal and a storage medium for constructing a knowledge graph, wherein the method comprises the steps of obtaining associated data associated with a target entity from a data source, preprocessing the associated data to obtain a reference associated entity set associated with the target entity in the associated data, determining the correlation between each reference associated entity in the reference associated entity set and the target entity, screening out at least one target associated entity from the reference associated entity set based on each correlation and a preset screening dimension, and constructing the target knowledge graph based on the target entity and each target associated entity. By implementing the method, the relevant knowledge graph can be automatically constructed based on the collected data, the correlation among the data can be embodied in different display forms in the knowledge graph, and the intelligence and the construction efficiency of the knowledge graph construction are improved.
Description
Technical Field
The invention relates to the technical field of information processing, in particular to a method, a device, a terminal and a storage medium for constructing a knowledge graph.
Background
The knowledge graph is a formal description framework based on semantic knowledge, generally adopts nodes to represent semantic symbols, can utilize connecting lines between the nodes to represent semantic relations between the symbols, provides an effective mode for expression, organization, management and the like of mass data on the Internet, and is widely applied to various fields.
In the existing knowledge graph construction mode, the incidence relation between different data is usually obtained manually, and a knowledge graph is drawn based on the incidence relation, however, the requirement on manpower is high and time consumption is long when the incidence relation between the data is judged manually and the graph is drawn, and the incidence relation is judged by a person subjectively, so that the judgment basis of the correlation between the data cannot be embodied in the knowledge graph, and therefore, the intelligence and the efficiency of constructing the knowledge graph are low at present.
Disclosure of Invention
The embodiment of the invention provides a method, a device, a terminal and a storage medium for constructing a knowledge graph, which can automatically construct a related knowledge graph based on collected data, embody the correlation among the data in different display forms in the knowledge graph, and improve the intelligence and the construction efficiency of the knowledge graph construction.
In one aspect, an embodiment of the present invention provides a method for constructing a knowledge graph, where the method includes:
obtaining association data associated with a target entity from a data source;
preprocessing the associated data to obtain a reference associated entity set associated with the target entity in the associated data, wherein the reference associated entity set comprises at least one reference associated entity;
determining a correlation between each reference associated entity in the set of reference associated entities and the target entity;
screening out at least one target associated entity from the reference associated entity set based on the correlation between each reference associated entity and the target entity and a preset screening dimension;
constructing a target knowledge graph based on the target entities and the target associated entities so as to analyze the target entities based on the target knowledge graph, wherein the relative display mode of the target entities and the target associated entities in the target knowledge graph is determined by the correlation between the target entities and the target associated entities.
In one aspect, an embodiment of the present invention provides an apparatus for constructing a knowledge graph, where the apparatus includes:
the acquisition module is used for acquiring the associated data associated with the target entity from the data source;
a preprocessing module, configured to preprocess the associated data to obtain a reference associated entity set associated with the target entity in the associated data, where the reference associated entity set includes at least one reference associated entity;
a determining module, configured to determine a correlation between each reference associated entity in the reference associated entity set and the target entity;
a screening module configured to screen out at least one target associated entity from the reference associated entity set based on a correlation between each reference associated entity and the target entity and a preset screening dimension;
the construction module is used for constructing a target knowledge graph based on the target entities and the target associated entities so as to analyze the target entities based on the target knowledge graph, wherein the relative display mode of the target entities and the target associated entities in the target knowledge graph is determined by the correlation between the target entities and the target associated entities.
In one aspect, an embodiment of the present invention provides a terminal, including a processor, an input interface, an output interface, and a memory, where the processor, the input interface, the output interface, and the memory are connected to each other, where the memory is used to store a computer program, and the computer program includes program instructions, and the processor is configured to call the program instructions to execute the method for constructing a knowledge graph.
In one aspect, an embodiment of the present invention provides a computer-readable storage medium, which stores a computer program, the computer program comprising program instructions, which, when executed by a processor, cause the processor to execute the method for constructing a knowledge graph.
In the embodiment of the invention, a terminal acquires associated data associated with a target entity from a data source, preprocesses the associated data to obtain a reference associated entity set associated with the target entity in the associated data, wherein the reference associated entity set comprises at least one reference associated entity, determines the correlation between each reference associated entity in the reference associated entity set and the target entity, screens out at least one target associated entity from the reference associated entity set based on the correlation between each reference associated entity and the target entity and a preset screening dimension, constructs a target knowledge graph based on the target entity and each target associated entity, and analyzes the target entity based on the target knowledge graph, wherein the relative display mode of the target entity and the target associated entity in the target knowledge graph is realized by the target entity and the target associated entity Correlation between the volumes is determined. By implementing the method, the relevant knowledge graph can be automatically constructed based on the collected data, the correlation among the data can be embodied in different display forms in the knowledge graph, and the intelligence and the construction efficiency of the knowledge graph construction are improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic flow chart of a method for constructing a knowledge graph according to an embodiment of the present invention;
FIG. 2a is a schematic view of a knowledge-graph according to an embodiment of the present invention;
FIG. 2b is a schematic illustration of another knowledge-graph provided by an embodiment of the present invention;
FIG. 2c is a schematic illustration of another knowledge-graph provided by an embodiment of the present invention;
FIG. 3 is a schematic flow chart of another method for constructing a knowledge graph according to an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of an apparatus for constructing a knowledge graph according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a terminal according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should also be understood that the term "and/or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.
The embodiment of the invention provides a method for constructing a knowledge graph, which can automatically construct a related knowledge graph based on collected data, embody the correlation among the data in different display forms in the knowledge graph, and improve the intelligence and the construction efficiency of the knowledge graph construction.
The embodiment of the invention provides a method for constructing a knowledge graph, which is implemented in a terminal, wherein the terminal comprises electronic equipment such as a smart phone, a tablet computer, a notebook computer, a palm computer, a Portable Media Player (PMP), a Personal Digital Assistant (PDA), a Digital audio and video Player, an electronic reader, a handheld game console or vehicle-mounted electronic equipment.
Fig. 1 is a schematic flow chart of a method for constructing a knowledge graph according to an embodiment of the present invention, and as shown in fig. 1, the flow of the method for constructing a knowledge graph in the embodiment may include:
s101, acquiring associated data associated with the target entity from a data source.
In the embodiment of the invention, the data source can be composed of a self-built data source and an internet open source data source, the self-built data source can be a database which is pre-built by research personnel, various entities and related information of the entities are stored in the database, and the internet open source data source specifically stores data which can be inquired from the internet. The terminal may obtain association data associated with the target entity from a data source. Specifically, the terminal may obtain a large amount of unprocessed associated data from the data source based on the preliminary information of the target entity. The preliminary information of the target entity may refer to a name of the target entity, an owner of the target entity, and the like, the associated data may specifically be text information, and the text information may be an article, a paragraph, a phrase, and the like.
In an implementation manner, the preliminary information of the target entity is a name of the target entity, and a specific manner of acquiring, by the terminal, the associated data associated with the target entity from the data source may be that the terminal acquires text information including the name of the target entity and determines the acquired text information as the associated data associated with the target entity.
In an implementation manner, if the preliminary information of the target entity is the owner of the target entity, the specific manner of acquiring, by the terminal, the associated data associated with the target entity from the data source may be that the terminal acquires text information including the name of the owner of the target entity and acquires the related data of the owner of the target entity as the associated data associated with the target entity. The related data of the owner comprises the relatives information, credit information, historical behavior information and the like of the owner.
For example, if the target entity is a property asset, the associated data acquired by the terminal may include section information of the property asset, owner information, auction information, other property information under the owner name, and the like.
S102, preprocessing the associated data to obtain a reference associated entity set associated with the target entity in the associated data.
In the embodiment of the present invention, after the terminal acquires the associated data, the associated data may be preprocessed to obtain a reference associated entity set associated with the target entity in the associated data, where the reference associated entity set includes at least one reference associated entity.
In an embodiment, the associated data may refer to text information including a target entity, and the specific manner in which the terminal preprocesses the associated data based on the obtained text information to obtain a reference associated entity set may be that the text information is subjected to word segmentation to obtain at least one word group, the terminal selects a target word group satisfying a preset rule from the at least one word group, determines a reference entity set corresponding to the target word group and a relationship between each reference entity in the reference entity set and the target entity, selects at least one reference entity from the reference entity set based on the relationship between each reference entity and the target entity, and uses the selected reference entity as the reference associated entity to construct a reference associated entity set associated with the target entity. The preset rule may be that the part of speech is a preset part of speech, and if the preset part of speech is a noun, the terminal screens out noun phrases from at least one phrase obtained by word segmentation, and the noun phrases are used as target phrases meeting the preset rule. The reference entity set corresponding to the target phrase is specifically obtained from a text containing the target phrase, and specifically, the reference entity set corresponding to the target phrase may be obtained by performing word segmentation processing on the text containing the target phrase to obtain a plurality of phrases, taking each phrase in the plurality of phrases as a reference entity corresponding to the target phrase, and obtaining by the terminal that the reference entity set is constructed based on the plurality of reference entities. The relationship between the reference entity and the target entity can be obtained by analyzing some key verbs in the text information, and the terminal extracts the relationship between the reference entity and the target entity from the text information. For example, the target entity is a property B, and the text information content obtained from the associated data is: "the belonger C of the property B located at the position a has also purchased the property D", performs word segmentation processing on the text information to obtain phrases "located", "position a", "property B", "belonger C", "purchase", "property D", screens a plurality of phrases by preset rules to obtain a target phrase, determines that a reference entity set corresponding to the target phrase is "position a", "property B", "belonger C", "property D", extracts key verbs "located", "purchase" in the text information, analyzes the key verbs to obtain relationships between the target entity and each reference entity in the reference entity set, further, can analyze the relationships between each reference entity in the reference entity set and the target entity as direct relationships or indirect relationships, the extracted relationships can be shown in a table form, specifically, as shown in table 1:
TABLE 1
In one embodiment, the associated data may refer to structured data and unstructured data related to the target entity, the structured data may refer to data of a relational database, and the unstructured data may refer to unstructured text data, tables, HTML files, and the like. The specific way of preprocessing the associated data based on the acquired structured data and unstructured data to obtain a reference associated entity set may be that the structured data and the unstructured data are preprocessed to obtain a target triple representing the relationship among the reference entity, the target entity and the entities, the relationship among the entities is specifically the relationship between the reference entity and the target entity, at least one reference entity is screened out from each reference entity based on the relationship among the entities in the triple, the screened reference entity is used as the reference associated entity, and the reference associated entity set associated with the target entity is constructed. The structured data can be preprocessed by converting the structured data into resource description frame data by using a data conversion tool, and converting the resource description frame data into structured data triples, the structured data triples include reference entities, target entities and relationships between the reference entities and the target entities, the structured data triples include reference entities, the target entities and the relationships between the reference entities and the target entities, the unstructured data triples include unstructured data triples, the unstructured data triples include reference entities, target entities and relationships between the reference entities and the target entities, the reference entities can be obtained by using an entity identification model by using an entity extraction technology, the relationships between the reference entities and the target entities can be obtained by using a model in a relationship extraction frame by using a relationship extraction technology, and combining the obtained structured data triples and the unstructured data triples, and integrating the reference entities with the same meaning to obtain target triples simultaneously containing the structured data triples and the unstructured data triples, namely performing entity alignment, wherein the technology for performing entity alignment can be entity unification and reference resolution, and further, the obtained target triples can be subjected to knowledge reasoning and quality evaluation to ensure the quality of information contained in the target triples.
S103, determining the correlation between each reference associated entity in the reference associated entity set and the target entity.
In the embodiment of the present invention, after the terminal acquires the reference associated entity set, the correlation between each reference associated entity in the reference associated entity set and the target entity may be further determined based on the relationship between each reference associated entity in the reference associated entity set and the target entity, and based on the magnitude of the correlation, the terminal may determine the strength of the correlation between each reference associated entity in the reference associated entity set and the target entity.
In one embodiment, the correlation between the reference associated entity and the target entity is determined by an acquisition source of the reference associated entity, and the manner for the terminal to determine the correlation between one target reference associated entity in the reference associated entity set and the target entity may be that the acquisition source of the target reference associated entity is determined, the correlation between the target reference associated entity and the target entity is determined according to a corresponding relationship between the acquisition source and the correlation, where the acquisition source may include an official database, a certified agency database, or an unauthenticated agency database, if the target reference associated entity is acquired from the official database, the correlation size is determined as a first correlation value, if the target reference associated entity is acquired from the certified agency database, the correlation size is determined as a second correlation value, if the target reference entity is acquired from the unauthenticated agency database, and determining the correlation size as a third correlation value, wherein the first correlation value is greater than the second correlation value, and the second correlation value is greater than the third correlation value, the official database may be a database corresponding to data disclosed by a government agency, the certified agency may have an agency of a certificate issued by the certification agency, the database corresponding to the disclosed data is a certified agency database, and the unauthenticated agency database may be a database storing data issued by each user in the internet. The above approach may determine the correlation between the reference associated entity and the target entity based on different data sources.
In an embodiment, a specific way for the terminal to determine the correlation between any one target reference associated entity in the reference associated entity set and the target entity may be that the terminal obtains a co-occurrence probability of the target reference associated entity and the target entity in the associated data, determines a relationship type corresponding to a relationship between the target reference associated entity and the target entity, determines a weight of the target reference associated entity according to the relationship type, and performs weighting processing on the co-occurrence probability of the target reference associated entity and the target entity in the associated data according to the weight of the target reference associated entity to obtain the correlation between the target reference associated entity and the target entity. The specific determination method of the co-occurrence probability of the target reference associated entity and the target entity in the associated data may be that the terminal obtains text information from the associated data, determines the co-occurrence frequency of the reference associated entity and the target entity in the same text information and the occurrence frequency of the target reference associated entity in the associated data, and determines the ratio of the co-occurrence frequency to the occurrence frequency as the co-occurrence probability of the target reference associated entity and the target entity in the associated data, for example, in one text information, the co-occurrence frequency x of the target reference associated entity j and the target entity iij1, the total frequency x of occurrence of the target reference associated entity in the associated datajIs 5, thus resulting in a co-occurrence probability of 0.2 for the target reference associated entity j and the target entity i in the associated data. The relationship type corresponding to the relationship between the target reference associated entity and the target entity mayThe method is divided into direct relation and indirect relation, and is specifically shown in the corresponding examples in table 1. Optionally, the relationship type corresponding to the relationship between the target reference associated entity and the target entity may also be determined by a distance between the target reference associated entity and the target entity in the same text, if the distance between the target reference associated entity and the target entity in the text is less than a preset distance, the relationship type is determined to be a direct relationship, if the distance is greater than or equal to the preset distance, the relationship type is determined to be an indirect relationship, and the distance may specifically be determined by the number of characters spaced between the target reference associated entity and the target entity in the text. Optionally, the relationship type corresponding to the relationship between the target reference associated entity and the target entity may also be determined based on the type of the target reference associated entity, if the type of the target reference associated entity is a name, the relationship type is determined to be a direct relationship, and if the type of the target reference associated entity is a number, the relationship type is determined to be an indirect relationship. And aiming at different relation types, different weights are correspondingly arranged, if the weight corresponding to the direct relation is 2 and the weight corresponding to the indirect relation is 1, when the relation type corresponding to the relation between the target reference associated entity and the target entity is the direct relation and the co-occurrence probability of the target reference associated entity and the target entity in the associated data is 0.2, the correlation between the target reference associated entity and the target entity is determined to be 0.4.
In an embodiment, a specific way for the terminal to obtain the co-occurrence probability of a target reference associated entity and a target entity in the associated data in the reference associated entity set may also be to establish a co-occurrence matrix of the target entity and the target reference associated entity, obtain text information in the associated data, preset a range, indicate that the reference associated entity k appears in the environment of the target entity i once if the distance between the target reference associated entity k and the target entity i in the text information is within the preset range, and set xikReferring to the number of times that the associated entity k appears in the environment for all positions of the text information where the target entity i exists, and co-occurrence matrix x corresponding to the reference associated entity k appears in the environment of the target entity i once every timeikThe numerical value plus one, set xiFor all x in co-occurrence matrixikAndi.e. by
The co-occurrence probability of the reference association entity k and the target entity i in the association data is Pk＝xik/xi。
Further, the terminal may determine the weight of the reference associated entity according to the obtained relationship type by manually assigning a weight value. For example, the weight value given to the owner of the target entity is 1, the weight value given to the position of the target entity is 0.7, the weight values given to other entities under the name of the owner are 0.5, and the weight values given to other entity related information under the name of the owner are 0.3; or the weight of the reference associated entity which has a direct relationship with the target entity is 1, and the weight of the reference associated entity which has an indirect relationship with the target entity is 0.5. The specific way of obtaining the correlation by performing weighting processing on each reference associated entity and the target entity may be to obtain the correlation between the corresponding reference associated entity and the target entity by multiplying the co-occurrence probability of each reference associated entity and the target entity in the associated data by the weighted value given to the reference associated entity. For example, the reference associated entity set includes a reference associated entity 1, a reference associated entity 2, and a reference associated entity 3, the co-occurrence probabilities are 0.7, 0.5, and 0.2, respectively, and the weight values are 0.5, 0.7, and 0.3, respectively, and based on the co-occurrence probabilities and the correlations of the respective reference associated entities, the correlation between the reference associated entity 1 and the target entity is 0.7 × 0.5 — 0.35, the correlation between the reference associated entity 2 and the target entity is 0.7 × 0.5 — 0.35, and the correlation between the reference associated entity 1 and the target entity is 0.2 × 0.3 — 0.06.
And S104, screening out at least one target associated entity from the reference associated entity set based on the correlation between each reference associated entity and the target entity and a preset screening dimension.
In the embodiment of the present invention, the terminal obtains the association degree between each reference associated entity in the reference associated entity set and the target entity based on the obtained correlation between each reference associated entity and the target entity, and further screens each reference associated entity to obtain a reference associated entity with a strong association degree.
In an embodiment, the preset screening dimension rule may be a preset threshold, and the terminal screens out, from the reference associated entity set, a reference associated entity whose correlation with the target entity is greater than the threshold, and uses the reference associated entity as the target associated entity, specifically, the preset threshold may be directly set manually, for example, if the correlation of the reference associated entity 1 in the reference associated entity set is 0.7, the correlation of the reference associated entity 2 is 0.3, the correlation of the reference associated entity 3 is 0.5, and the manually set threshold is 0.5, then the reference associated entities screened out as the target associated entity are the reference associated entity 1 and the reference associated entity 3. The preset threshold may also be calculated by a formula based on a plurality of correlations, inputting a plurality of correlation values into an average formula,
further, the threshold values are obtained, for example, the correlation of the reference associated entity 1, the correlation of the reference associated entity 2 and the correlation of the reference associated entity 3 in the reference associated entity set are 0.7, 0.3 and 0.5 respectively, and the average value formula is calculated
The value of the correlation is calculated to obtain a threshold value of
Then the reference associated entities as the target associated entities are screened out as the reference associated entity 1 and the reference associated entity 3. The preset screening dimension rule may also be that each reference associated entity in the reference associated entity set is sorted from high to low according to the corresponding correlation, a required dimension number of the reference associated entity is preset, and the terminal selects the reference associated entity from the sorted reference associated entity set from high to low according to the set dimension number as the target associated entityAnd the number of the selected reference associated entities is determined by the preset dimension number.
And S105, constructing a target knowledge graph based on the target entities and the target associated entities.
In the embodiment of the invention, the terminal constructs a multi-dimensional target knowledge graph based on the target entity and each target associated entity, the constructed target knowledge graph takes the target entity as the center, the target entity is respectively connected with each target associated entity, after the construction is finished, the terminal can analyze the target entity based on the target knowledge graph, and store the constructed target knowledge graph and information obtained based on the analysis of the target knowledge graph into the database.
In an embodiment, a specific manner of connecting the target entity with each target associated entity may be that the terminal acquires a correlation between the target entity and each target associated entity, and determines a corresponding map construction manner based on the correlation. The map construction method may be that, according to the magnitude of the correlation, the connection distance between each target associated entity and the target entity in the target knowledge map is determined, and specifically, the distances of line segments used for connecting the target entity and each target associated entity are different according to the difference in the correlation. Optionally, according to the size of the correlation, display colors of the target associated entities in the target knowledge graph are determined, that is, the correlations are different, and the colors used for representing the entities are different, for example, the target associated entity with the highest correlation with the target entity is represented by green in the target knowledge graph, and the target associated entity with the second highest correlation with the target entity is represented by blue in the target knowledge graph, as shown in fig. 2 a.
In an embodiment, after the target knowledge graph is constructed, the target knowledge graph may be further expanded, the knowledge graph of the extended associated entities in the target associated entities is continuously constructed, and the constructed knowledge graph of the extended associated entities is stored in the database. After the terminal obtains the extended associated entities, based on the preliminary information of the extended associated entities, obtaining extended associated data associated with the extended associated entities from the data source, the terminal obtains a reference extended associated entity set associated with the extended associated data based on the extended associated data, the specific way of obtaining the reference extended associated entity set may be to obtain text information including the extended associated entities in the reference extended associated data, and perform word segmentation processing on the text information to obtain at least one word group, the terminal selects an extended target word group satisfying a preset rule from the at least one word group, and determines a reference extended entity set corresponding to the extended target word group and a relationship between each reference extended entity in the reference extended entity set and the extended associated entities, the terminal selects at least one reference extended entity from the reference extended entity set based on the relationship between each reference extended entity and the extended associated entities, and constructing a reference extension associated entity set associated with the extension associated entities by taking the screened reference extension entities as reference extension associated entities. The preset rule may be to filter out words with the same part of speech as the extended associated entity, for example, the extended associated entity is a noun, and the preset rule may be to filter out noun phrases. Determining the relationship between the reference extended entity and the extended associated entity may be analyzing the relationship between the reference extended entity and the extended associated entity according to some key verbs in the text information, extracting the relationship between the reference extended entity and the extended associated entity from the text information, and screening at least one reference extended entity from the reference extended entity set based on the relationship between each reference extended entity and the extended associated entity in the reference extended entity set, which may mean screening the reference extended entity from the reference extended entity set based on the extracted relationship, and constructing a reference extended associated entity set associated with the extended associated entity by using the screened reference extended entity as the reference extended associated entity. The terminal obtains a reference extended associated entity set, and constructs a knowledge sub-map based on the extended associated entity and each reference extended associated entity in the reference extended associated entity set, where the specific construction mode may be that the extended associated entity is used as a center, and the extended associated entity is connected to each reference extended associated entity, as shown in fig. 2 b.
In an embodiment, the terminal obtains the knowledge sub-map, may combine the knowledge sub-map with the target knowledge map to obtain an extended knowledge map for the target associated entity, and store the obtained extended knowledge map in the database, and the specific manner may be that the knowledge sub-map is superimposed on the corresponding target associated entity in the target knowledge map according to the correspondence between the extended associated entity in the knowledge sub-map and the target associated entity in the target knowledge map, as shown in fig. 2 c.
In the embodiment of the invention, a terminal acquires associated data associated with a target entity from a data source, preprocesses the associated data to obtain a reference associated entity set associated with the target entity in the associated data, determines the correlation between each reference associated entity in the reference associated entity set and the target entity, screens out at least one target associated entity from the reference associated entity set based on the correlation between each reference associated entity and the target entity and a preset screening dimension, and constructs a target knowledge graph based on the target entity and each target associated entity. By implementing the method, a reference associated entity set can be obtained by preprocessing the associated data, the target associated entity highly related to the target entity can be obtained by calculating the correlation between the reference associated entity in the reference associated entity set and the target entity, and the high-correlation knowledge map can be constructed on the basis of the target associated entity and the target entity, so that a large amount of redundant information is abandoned to a certain extent, the construction efficiency is improved, and the constructed knowledge map is more accurate.
Fig. 3 is a schematic flowchart of a method for constructing a knowledge graph according to an embodiment of the present invention, and as shown in fig. 3, a flowchart of the method for constructing a knowledge graph in the embodiment may include:
s301, when a target entity knowledge graph construction instruction is received, whether the knowledge graph of the target entity is stored in the database or not is detected.
In the embodiment of the invention, when a knowledge graph is constructed, the terminal stores the knowledge graph in the database, when the terminal receives a target entity knowledge graph construction instruction, whether the corresponding target entity knowledge graph is stored in the database or not is detected, if the corresponding target entity knowledge graph is stored in the database, S302 is executed, and if the corresponding target entity knowledge graph is not stored in the database, S303 is executed.
S302, if the target entity knowledge graph is stored in the database, whether the knowledge graph updating condition is met is detected.
In the embodiment of the invention, when a terminal detects that a corresponding target entity knowledge graph is stored in a database, the target entity knowledge graph is detected, whether a knowledge graph updating condition is met or not is judged, if the knowledge graph updating condition is met, the target entity knowledge graph needs to be updated, S304 is executed, and if the knowledge graph updating condition is not met, the target entity knowledge graph does not need to be updated, and S305 is executed. The specific way of judging whether the knowledge graph updating condition is met or not can be to judge whether the construction time of the target entity knowledge graph exceeds a preset effective time range or not, if the construction time of the target entity knowledge graph exceeds the effective time range, the knowledge graph updating condition is met, and if the construction time of the target entity knowledge graph does not exceed the effective time range, the knowledge graph updating condition is not met.
And S303, if the target entity knowledge graph is not stored in the database, constructing the target entity knowledge graph and storing the target entity knowledge graph in the database.
In the embodiment of the invention, if the terminal detects that the target entity knowledge graph is not stored in the database, the target entity is a new target entity, the construction of the knowledge graph of the target entity is required, the specific construction mode can be as shown in S101-S105, and the target entity knowledge graph is stored in the database after the construction is completed, so that the target entity knowledge graph can be detected or called when the terminal receives the target entity knowledge graph construction instruction next time.
And S304, if the knowledge graph updating condition is not met, directly calling the target entity knowledge graph.
In the embodiment of the invention, if the terminal detects that the target entity knowledge graph does not meet the knowledge graph updating condition, the target entity knowledge graph is directly called for display without updating the target entity knowledge graph.
And S305, if the knowledge graph updating condition is met, updating the target entity knowledge graph and storing the target entity knowledge graph in a database.
In the embodiment of the invention, if the terminal detects that the target entity knowledge graph meets the knowledge graph updating condition, the target entity knowledge graph is updated, the specific updating mode can be to re-construct the knowledge graph for the target entity, and the specific constructing mode can be as shown in S101-S105. The specific way of updating may also be to acquire new associated data related to the target entity and process the new associated data to obtain a new target associated entity and a new correlation, and the specific way may be as shown in S101-S104, and compare the original target associated entity and the correlation of the target entity knowledge graph with the new target associated entity and the correlation, output the target associated entity or the correlation having the difference, and update the target entity knowledge graph correspondingly based on the target associated entity or the correlation having the difference. The updated knowledge graph is stored in the database and used for detection or calling when the terminal receives the target entity knowledge graph construction instruction next time.
In the embodiment of the invention, when a terminal receives a target entity knowledge graph construction instruction, whether the knowledge graph of a target entity is stored in a database is detected, if the target entity knowledge graph is stored in the database, the terminal detects whether a knowledge graph updating condition is met, if the knowledge graph updating condition is not met, the terminal directly calls the target entity knowledge graph, if the knowledge graph updating condition is met, the terminal updates the target entity knowledge graph and stores the target entity knowledge graph in the database, and if the target entity knowledge graph is not stored in the database, the terminal constructs the target entity knowledge graph and stores the target entity knowledge graph in the database. By implementing the method, the efficiency of obtaining the target knowledge graph can be improved to a certain extent, and the waste of resources caused by constructing the knowledge graph of the same target entity for multiple times is avoided.
An apparatus for constructing a knowledge graph according to an embodiment of the present invention will be described in detail with reference to fig. 4. It should be noted that the apparatus for constructing a knowledge graph shown in fig. 4 is used for executing the method of the embodiment shown in fig. 1 and 3 of the present invention, and for convenience of description, only the portion related to the embodiment of the present invention is shown, and the specific technical details are not disclosed, and reference is made to the embodiment shown in fig. 1 and 3 of the present invention.
Referring to fig. 4, a schematic structural diagram of an apparatus for constructing a knowledge graph according to the present invention is shown, where the apparatus 400 for constructing a knowledge graph may include: the system comprises an acquisition module 401, a preprocessing module 402, a determination module 403, a screening module 404 and a construction module 405.
An obtaining module 401, configured to obtain association data associated with a target entity from a data source;
a preprocessing module 402, configured to preprocess the associated data to obtain a reference associated entity set associated with the target entity in the associated data, where the reference associated entity set includes at least one reference associated entity;
a determining module 403, configured to determine a correlation between each reference associated entity in the reference associated entity set and the target entity;
a screening module 404, configured to screen out at least one target associated entity from the reference associated entity set based on a correlation between each reference associated entity and the target entity and a preset screening dimension;
a constructing module 405, configured to construct a target knowledge graph based on the target entity and each target associated entity, so as to analyze the target entity based on the target knowledge graph, where a relative display manner of the target entity and the target associated entity in the target knowledge graph is determined by a correlation between the target entity and the target associated entity.
In one embodiment, the associated data includes text information including the target entity, and the preprocessing module 402 is specifically configured to perform word segmentation processing on the text information to obtain at least one word group; screening target phrases meeting preset rules from the at least one phrase; determining a reference entity set corresponding to the target phrase and a relationship between each reference entity in the reference entity set and the target entity; and screening at least one reference entity from the reference entity set based on the relation between each reference entity and the target entity, and constructing a reference associated entity set associated with the target entity.
In an embodiment, the determining module 403 is specifically configured to obtain a co-occurrence probability of the target reference associated entity and the target entity in the associated data, where the target reference associated entity includes any reference associated entity in the reference associated entity set; determining a relation type corresponding to the relation between the target reference associated entity and the target entity, and determining the weight of the target reference associated entity according to the relation type; and according to the weight of the target reference associated entity, carrying out weighting processing on the co-occurrence probability of the target reference associated entity and the target entity in the associated data to obtain the correlation between the target reference associated entity and the target entity.
In an embodiment, the associated data includes at least one text, and the determining module 403 is specifically configured to obtain a co-occurrence frequency of the target reference associated entity and the target entity in the same text in the associated data, and an occurrence frequency of the target reference associated entity in the associated data; and determining the ratio of the co-occurrence frequency to the occurrence frequency as the co-occurrence probability of the target reference associated entity and the target entity in the associated data.
In an embodiment, the preset screening dimension includes a preset threshold, and the screening module 404 is specifically configured to screen, from the reference associated entity set, a reference associated entity whose correlation with the target entity is greater than the preset threshold, as the target associated entity.
In an embodiment, the constructing module 405 is specifically configured to obtain a correlation between the target entity and each target associated entity, and determine, based on the correlation, a relative display manner of the target entity and the target associated entity in the knowledge graph, where the relative display manner includes a connection distance between each target associated entity and the target entity in the target knowledge graph and a display color of each target associated entity in the target knowledge graph.
In an embodiment, after the constructing module 405, an extended associated entity is screened from the target associated entities, extended associated data associated with the extended associated entity is obtained from the data source, a reference extended associated entity set associated with the extended associated entity is obtained based on the extended associated data, an extended knowledge sub-graph is constructed based on the extended associated entity and the reference extended associated entity set, and the target knowledge graph and the extended knowledge sub-graph are combined to obtain an extended knowledge graph for the target associated entity.
In the embodiment of the present invention, an obtaining module 401 obtains associated data associated with a target entity from a data source, a preprocessing module 402 preprocesses the associated data to obtain a reference associated entity set associated with the target entity in the associated data, where the reference associated entity set includes at least one reference associated entity, a determining module 403 is configured to determine a correlation between each reference associated entity in the reference associated entity set and the target entity, a screening module 404 screens out at least one target associated entity from the reference associated entity set based on the correlation between each reference associated entity and the target entity and a preset screening dimension, a constructing module 405 constructs a target knowledge graph based on the target entity and each target associated entity to analyze the target entity based on the target knowledge graph, wherein the relative display manner of the target entity and the target associated entity in the target knowledge graph is determined by the correlation between the target entity and the target associated entity. By implementing the method, the relevant knowledge graph can be automatically constructed based on the collected data, the correlation among the data is embodied in the knowledge graph in different display forms, and the intelligence and the construction efficiency of the knowledge graph construction are improved to a certain extent.
Fig. 5 is a schematic structural diagram of a terminal according to an embodiment of the present invention. As shown in fig. 5, the terminal 500 includes: at least one processor 501, an input device 502, an output device 503, a memory 504, at least one communication bus 505. The input device 502 may be a control panel or a microphone, and the output device 503 may be a display screen. The memory 504 may be a high-speed RAM memory or a non-volatile memory (non-volatile memory), such as at least one disk memory. Wherein a communication bus 505 is used to enable connectivity communication between these components. The memory 504 may optionally be at least one storage device located remotely from the processor 501. Wherein the processor 501 may be combined with the apparatus described in fig. 4, the memory 504 stores a set of program codes, and the processor 501, the input device 502, and the output device 503 call the program codes stored in the memory 504 to perform the following operations:
a processor 501, configured to obtain association data associated with a target entity from a data source;
a processor 501, configured to pre-process the associated data to obtain a reference associated entity set associated with the target entity in the associated data, where the reference associated entity set includes at least one reference associated entity;
a processor 501, configured to determine a correlation between each reference associated entity in the reference associated entity set and the target entity;
a processor 501, configured to screen out at least one target associated entity from the reference associated entity set based on a correlation between each reference associated entity and the target entity and a preset screening dimension;
a processor 501, configured to construct a target knowledge graph based on the target entities and the target associated entities, so as to analyze the target entities based on the target knowledge graph, where a relative display manner of the target entities and the target associated entities in the target knowledge graph is determined by a correlation between the target entities and the target associated entities.
In an embodiment, the associated data includes text information including the target entity, and the processor 501 is specifically configured to:
performing word segmentation processing on the text information to obtain at least one word group;
screening target phrases meeting preset rules from the at least one phrase;
determining a reference entity set corresponding to the target phrase and a relationship between each reference entity in the reference entity set and the target entity;
and screening at least one reference entity from the reference entity set based on the relation between each reference entity and the target entity, and constructing a reference associated entity set associated with the target entity.
In an embodiment, the processor 501 is specifically configured to:
acquiring the co-occurrence probability of the target reference associated entity and the target entity in the associated data, wherein the target reference associated entity comprises any one reference associated entity in the reference associated entity set;
determining a relation type corresponding to the relation between the target reference associated entity and the target entity, and determining the weight of the target reference associated entity according to the relation type;
and according to the weight of the target reference associated entity, carrying out weighting processing on the co-occurrence probability of the target reference associated entity and the target entity in the associated data to obtain the correlation between the target reference associated entity and the target entity.
In an embodiment, the associated data includes at least one text, and the processor 501 is specifically configured to:
acquiring the co-occurrence frequency of the target reference associated entity and the target entity in the same text in the associated data and the occurrence frequency of the target reference associated entity in the associated data;
and determining the ratio of the co-occurrence frequency to the occurrence frequency as the co-occurrence probability of the target reference associated entity and the target entity in the associated data.
In an embodiment, the processor 501 is specifically configured to: and screening out the reference associated entities with the correlation with the target entity larger than the preset threshold value from the reference associated entity set as target associated entities.
In an embodiment, the processor 501 is specifically configured to:
acquiring the correlation between the target entity and each target associated entity;
determining a relative display mode of the target entity and the target associated entity in the knowledge graph based on the correlation, wherein the relative display mode comprises a connection distance between each target associated entity and the target entity in the target knowledge graph and a display color of each target associated entity in the target knowledge graph.
In an embodiment, after the target knowledge-graph is constructed based on the target entities and the target associated entities, the processor 501 is specifically configured to:
screening out an expansion associated entity from each target associated entity;
acquiring extension associated data associated with the extension associated entity from the data source;
obtaining a reference extended associated entity set associated with the extended associated entity based on the extended associated data;
and constructing an extended knowledge sub-graph based on the extended associated entity and the reference extended associated entity set, and combining the target knowledge graph and the extended knowledge sub-graph to obtain an extended knowledge graph for the target associated entity.
In the embodiment of the present invention, the processor 501 obtains associated data associated with a target entity from a data source, the processor 501 pre-processes the associated data to obtain a reference associated entity set associated with the target entity in the associated data, the reference associated entity set comprises at least one reference associated entity, the processor 501 determines the correlation between each reference associated entity in the reference associated entity set and the target entity, the processor 501 screens out at least one target associated entity from the reference associated entity set based on the correlation between each reference associated entity and the target entity and a preset screening dimension, and the processor 501 constructs a target knowledge graph based on the target entity and each target associated entity so as to analyze the target entity based on the target knowledge graph. By implementing the method, the relevant knowledge graph can be automatically constructed based on the collected data, the correlation among the data is embodied in different display forms in the knowledge graph, and the intelligence and the construction efficiency of the knowledge graph construction are improved.
The module in the embodiment of the present invention may be implemented by a general-purpose Integrated Circuit, such as a CPU (central Processing Unit), or an ASIC (application Specific Integrated Circuit).
It should be understood that, in the embodiment of the present invention, the Processor 501 may be a Central Processing Unit (CPU), and may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The communication bus 505 may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended ISA (EISA) bus, or the like, and the communication bus 505 may be divided into an address bus, a data bus, a control bus, or the like, and fig. 5 shows only one thick line for convenience of illustration, but does not show only one bus or one type of bus.
It will be understood by those skilled in the art that all or part of the processes of the methods of the above embodiments may be implemented by a computer program, which may be stored in a computer storage medium and may include the processes of the embodiments of the methods described above when executed. The computer storage medium may be a magnetic disk, an optical disk, a Read-only Memory (ROM), a Random Access Memory (RAM), or the like.
While the present embodiments have been described with reference to the accompanying drawings, it is to be understood that the invention is not limited to the precise embodiments described above, which are meant to be illustrative and not restrictive, and that various changes may be made therein by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. A method of constructing a knowledge graph, comprising:
obtaining association data associated with a target entity from a data source;
preprocessing the associated data to obtain a reference associated entity set associated with the target entity in the associated data, wherein the reference associated entity set comprises at least one reference associated entity;
determining a correlation between each reference associated entity in the set of reference associated entities and the target entity;
screening out at least one target associated entity from the reference associated entity set based on the correlation between each reference associated entity and the target entity and a preset screening dimension;
constructing a target knowledge graph based on the target entities and the target associated entities so as to analyze the target entities based on the target knowledge graph, wherein the relative display mode of the target entities and the target associated entities in the target knowledge graph is determined by the correlation between the target entities and the target associated entities.
2. The method of claim 1, wherein the association data includes textual information including the target entity, and wherein pre-processing the association data to obtain a reference set of associated entities in the association data that are associated with the target entity comprises:
performing word segmentation processing on the text information to obtain at least one word group;
screening target phrases meeting preset rules from the at least one phrase;
determining a reference entity set corresponding to the target phrase and a relationship between each reference entity in the reference entity set and the target entity;
and screening at least one reference entity from the reference entity set based on the relation between each reference entity and the target entity, and constructing a reference associated entity set associated with the target entity.
3. The method of claim 1, wherein determining the relevance between a target reference associated entity of the set of reference associated entities and the target entity comprises:
acquiring the co-occurrence probability of the target reference associated entity and the target entity in the associated data, wherein the target reference associated entity comprises any one reference associated entity in the reference associated entity set;
determining a relation type corresponding to the relation between the target reference associated entity and the target entity, and determining the weight of the target reference associated entity according to the relation type;
and according to the weight of the target reference associated entity, carrying out weighting processing on the co-occurrence probability of the target reference associated entity and the target entity in the associated data to obtain the correlation between the target reference associated entity and the target entity.
4. The method of claim 3, wherein the association data includes at least one text, the obtaining a probability of co-occurrence of the target reference associated entity with the target entity in the association data comprising:
acquiring the co-occurrence frequency of the target reference associated entity and the target entity in the same text in the associated data and the occurrence frequency of the target reference associated entity in the associated data;
and determining the ratio of the co-occurrence frequency to the occurrence frequency as the co-occurrence probability of the target reference associated entity and the target entity in the associated data.
5. The method of claim 1, wherein the preset screening dimension comprises a preset threshold, the screening at least one target associated entity from the set of reference associated entities based on the correlation between the respective reference associated entity and the target entity and a preset screening dimension comprising:
and screening out the reference associated entities with the correlation with the target entity larger than the preset threshold value from the reference associated entity set as target associated entities.
6. The method of claim 1, wherein the constructing a target knowledge-graph based on the target entities and the respective target associated entities comprises:
acquiring the correlation between the target entity and each target associated entity;
determining a relative display mode of the target entity and the target associated entity in the knowledge graph based on the correlation, wherein the relative display mode comprises a connection distance between each target associated entity and the target entity in the target knowledge graph and a display color of each target associated entity in the target knowledge graph.
7. The method of claim 1, wherein after constructing a target knowledge-graph based on the target entities and the respective target associated entities, further comprising:
screening out an expansion associated entity from each target associated entity;
acquiring extension associated data associated with the extension associated entity from the data source;
obtaining a reference extended associated entity set associated with the extended associated entity based on the extended associated data;
and constructing an extended knowledge sub-graph based on the extended associated entity and the reference extended associated entity set, and combining the target knowledge graph and the extended knowledge sub-graph to obtain an extended knowledge graph for the target associated entity.
8. An apparatus for constructing a knowledge graph, comprising:
the acquisition module is used for acquiring the associated data associated with the target entity from the data source;
a preprocessing module, configured to preprocess the associated data to obtain a reference associated entity set associated with the target entity in the associated data, where the reference associated entity set includes at least one reference associated entity;
a determining module, configured to determine a correlation between each reference associated entity in the reference associated entity set and the target entity;
a screening module configured to screen out at least one target associated entity from the reference associated entity set based on a correlation between each reference associated entity and the target entity and a preset screening dimension;
the construction module is used for constructing a target knowledge graph based on the target entities and the target associated entities so as to analyze the target entities based on the target knowledge graph, wherein the relative display mode of the target entities and the target associated entities in the target knowledge graph is determined by the correlation between the target entities and the target associated entities.
9. A terminal, comprising a processor, an input interface, an output interface, and a memory, the processor, the input interface, the output interface, and the memory being interconnected, wherein the memory is configured to store a computer program comprising program instructions, the processor being configured to invoke the program instructions to perform the method of any of claims 1-7.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program comprising program instructions that, when executed by a processor, cause the processor to carry out the method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011289180.1A CN112199516A (en) | 2020-11-17 | 2020-11-17 | Method, device, terminal and storage medium for constructing knowledge graph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011289180.1A CN112199516A (en) | 2020-11-17 | 2020-11-17 | Method, device, terminal and storage medium for constructing knowledge graph |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112199516A true CN112199516A (en) | 2021-01-08 |
Family
ID=74034457
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011289180.1A Pending CN112199516A (en) | 2020-11-17 | 2020-11-17 | Method, device, terminal and storage medium for constructing knowledge graph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112199516A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113836313A (en) * | 2021-09-13 | 2021-12-24 | 北京信息科技大学 | Audit information identification method and system based on map |
| CN115618018A (en) * | 2022-10-31 | 2023-01-17 | 福州果集信息科技有限公司 | Knowledge graph construction method based on SPU and storage medium |
| CN115631495A (en) * | 2022-10-31 | 2023-01-20 | 福州果集信息科技有限公司 | SPU (SPU) acquisition method based on page analysis and storage medium |
| WO2023216671A1 (en) * | 2022-05-07 | 2023-11-16 | 北京金堤科技有限公司 | Graph display method and apparatus, storage medium and electronic device |
-
2020
- 2020-11-17 CN CN202011289180.1A patent/CN112199516A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113836313A (en) * | 2021-09-13 | 2021-12-24 | 北京信息科技大学 | Audit information identification method and system based on map |
| WO2023216671A1 (en) * | 2022-05-07 | 2023-11-16 | 北京金堤科技有限公司 | Graph display method and apparatus, storage medium and electronic device |
| CN115618018A (en) * | 2022-10-31 | 2023-01-17 | 福州果集信息科技有限公司 | Knowledge graph construction method based on SPU and storage medium |
| CN115631495A (en) * | 2022-10-31 | 2023-01-20 | 福州果集信息科技有限公司 | SPU (SPU) acquisition method based on page analysis and storage medium |
| CN115631495B (en) * | 2022-10-31 | 2023-08-22 | 福州果集信息科技有限公司 | SPU acquisition method based on page analysis and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112199516A (en) | Method, device, terminal and storage medium for constructing knowledge graph | |
| CN110555372A (en) | Data entry method, device, equipment and storage medium | |
| CN106874253A (en) | Recognize the method and device of sensitive information | |
| CN110362601B (en) | Metadata standard mapping method, device, equipment and storage medium | |
| WO2017198031A1 (en) | Semantic parsing method and apparatus | |
| CN106897290B (en) | Method and device for establishing keyword model | |
| CN112434188B (en) | Data integration method, device and storage medium of heterogeneous database | |
| CN111382248A (en) | Question reply method and device, storage medium and terminal equipment | |
| CN112784066A (en) | Information feedback method, device, terminal and storage medium based on knowledge graph | |
| CN112199588A (en) | Public opinion text screening method and device | |
| CN112001170A (en) | Method and system for recognizing deformed sensitive words | |
| CN111159404A (en) | Text classification method and device | |
| CN111429184A (en) | User portrait extraction method based on text information | |
| CN114722199A (en) | Risk identification method and device based on call recording, computer equipment and medium | |
| CN117763126A (en) | Knowledge retrieval method, device, storage medium and apparatus | |
| CN112579781A (en) | Text classification method and device, electronic equipment and medium | |
| CN110347934B (en) | Text data filtering method, device and medium | |
| CN110597985A (en) | Data classification method, device, terminal and medium based on data analysis | |
| CN111125379A (en) | Knowledge base expansion method and device, electronic equipment and storage medium | |
| CN113836297B (en) | Training method and device for text emotion analysis model | |
| CN109961801A (en) | Intelligent Service evaluation method, computer readable storage medium and terminal device | |
| CN115544204A (en) | Bad corpus filtering method and system | |
| CN114417010A (en) | Knowledge graph construction method and device for real-time workflow and storage medium | |
| CN114117057A (en) | Keyword extraction method of product feedback information and terminal equipment | |
| CN113159178A (en) | Problem expansion method, device, server and medium |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20210108 |