CN114064915A - Domain knowledge graph construction method and system based on rules and deep learning - Google Patents
Domain knowledge graph construction method and system based on rules and deep learning Download PDFInfo
- Publication number
- CN114064915A CN114064915A CN202111230908.8A CN202111230908A CN114064915A CN 114064915 A CN114064915 A CN 114064915A CN 202111230908 A CN202111230908 A CN 202111230908A CN 114064915 A CN114064915 A CN 114064915A
- Authority
- CN
- China
- Prior art keywords
- extraction
- text
- module
- deep learning
- entity
- 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
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
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Animal Behavior & Ethology (AREA)
- Databases & Information Systems (AREA)
- Machine Translation (AREA)
Abstract
The embodiment of the application discloses a method and a system for establishing a domain knowledge graph based on rules and deep learning, wherein the method comprises the following steps: s1, inputting a document; s2, classifying and screening the documents to obtain structured texts and unstructured texts; s3, carrying out rule-based knowledge extraction on the structured text to obtain a first triple; s4, carrying out serialized labeling on the unstructured text to obtain a labeled text; s5, performing knowledge extraction based on deep learning on the annotated text to obtain a second triple; and S6, constructing a domain knowledge graph according to the first triples and the second triples. The method realizes the construction of the domain knowledge graph by fusing the extraction method based on the rules and the deep learning, and solves the problems of low generalization capability of the extraction method based on the rules and low accuracy of the extraction method based on the deep learning.
Description
Technical Field
The application relates to the technical field of natural language processing, in particular to a method and a system for establishing a domain knowledge graph based on rules and deep learning.
Background
With the explosive growth of data, the traditional manual retrieval mode cannot meet the requirement of rapid capture and understanding of knowledge by human beings. The knowledge graph has great advantages in the aspects of solving the precision and the expandability of knowledge query. Knowledge-maps provide an ability to better organize, manage, and understand data information by expressing a large amount of data information in a form that more closely resembles human cognition. Since the concept of knowledge graph was proposed by Google 2012, more and more large-scale universal knowledge graphs were constructed, such as: freebase, Wikidata, DBpedia and the like are widely applied to the fields of personalized recommendation, intelligent question answering, information retrieval and the like.
A knowledge graph is essentially a structured knowledge base and is also a huge structured knowledge network, and entities in the network represent things or concepts in the real world and are generally represented in the form of triples (head entities, relations, tail entities). The process of constructing the knowledge graph refers to a process of extracting entities and relations from structured or unstructured text, namely a process of extracting knowledge from documents.
The construction method of the general knowledge graph is mature, the extraction of entities and relations is realized by deep learning, but the domain knowledge graph is different from the general knowledge graph, wherein the general knowledge graph faces to the general field, the general knowledge is based on the common knowledge, the wider the knowledge coverage field is, the better the domain knowledge graph faces to the specific field, and the user aims at experts with different levels in the field, so the domain knowledge graph has more strict quality requirements, higher expert participation degree and lower automation degree compared with the general knowledge graph, and the high-quality domain knowledge graph construction is difficult to realize by directly using the deep learning method. In addition, in the industry, due to the fact that message communication among all departments is not timely, internal structures of documents generated in the enterprise are different, and it is difficult to completely extract all entities and relations by directly using the rule-based extraction method.
Disclosure of Invention
Therefore, an extraction method based on the combination of rules and deep learning is needed to construct the domain knowledge graph.
The invention aims to provide a domain knowledge graph construction method and system based on rules and deep learning, and in order to achieve at least one of the purposes, the technical scheme is as follows:
the application provides a domain knowledge graph construction method based on rules and deep learning in a first aspect, which comprises the following steps:
s1, inputting a document;
s2, classifying and screening the documents to obtain structured texts and unstructured texts;
s3, carrying out rule-based knowledge extraction on the structured text to obtain a first triple;
s4, carrying out serialized labeling on the unstructured text to obtain a labeled text;
s5, performing knowledge extraction based on deep learning on the annotated text to obtain a second triple;
and S6, constructing a domain knowledge graph according to the first triples and the second triples.
In a specific embodiment, the S2 further includes:
s21, classifying the documents according to the document types to obtain classified documents;
s22, screening the classified documents by using a text screening mode to screen out structured texts and unstructured texts;
wherein the content of the first and second substances,
the document type is judged according to the document title.
In a specific embodiment, the S3 further includes:
making an extraction rule according to the definition of an expert and extracting;
the rule-based knowledge extraction includes: extracting knowledge by taking an article structure as a characteristic and extracting index knowledge in a professional field;
wherein, the structure of the index is as follows: index name: and (4) index value.
In a specific embodiment, the S5 includes:
s51, performing entity extraction on the annotated text;
and S52, extracting the relation of the annotated text.
In a specific embodiment, the S51 includes:
s511, preprocessing the annotated text through a BERT pre-training language model to obtain a word vector;
s512, processing the word vector through a bidirectional GRU module to obtain the context information of the entity;
s513, decoding the context information through a full-connection module to obtain the probability distribution of the entity;
s514, correcting the probability distribution through a conditional random field model;
and S515, extracting and classifying the entities to complete entity extraction.
In a specific embodiment, the S52 includes:
s521, preprocessing the annotated text through a BERT pre-training language model to obtain a word vector;
s522, processing the word vector through a bidirectional GRU module to obtain context information of an entity;
s523, acquiring a head entity vector and a tail entity vector of the relation through a multilayer perceptron module;
and S524, judging the relation between the head entity vector and the tail entity vector through a bilinear model, and finishing the relation extraction.
The second aspect of the present application provides a domain knowledge graph building system based on rules and deep learning, including:
the input module is used for inputting a document;
the classification and screening module is used for classifying and screening the documents to obtain structured texts and unstructured texts;
a rule extraction module for extracting the knowledge based on the rule from the structured text to obtain a first triple
The labeling module is used for carrying out serialized labeling on the unstructured text to obtain a labeled text deep learning extraction module and carrying out knowledge extraction based on deep learning on the labeled text to obtain a second triple;
and the output domain knowledge graph module is used for constructing a domain knowledge graph according to the first triad and the second triad.
In a specific embodiment, the deep learning module further includes:
the entity extraction module is used for performing entity extraction on the annotated text;
and the relation extraction module is used for extracting the relation of the labeled text.
A third aspect of the application provides a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the above method when executing the program.
A fourth aspect of the present application provides a computer readable medium having stored thereon a computer program which, when executed by a processor, performs the above-described method.
The beneficial effect of this application is as follows:
the method and the system realize the construction of the domain knowledge map by fusing the extraction method based on the rules and the deep learning, solve the problems of low generalization capability of the extraction method based on the rules and low accuracy of the extraction method based on the deep learning, can obtain higher accuracy and certain automation degree, and reduce the participation degree of experts. The method and the device realize effective management and visual display of the domain knowledge, and provide support for realizing intelligent retrieval and quick capture and understanding of the knowledge.
Drawings
The following describes embodiments of the present invention in further detail with reference to the accompanying drawings.
Fig. 1 is a diagram illustrating steps of a domain knowledge graph construction method based on rules and deep learning in an embodiment of the present application.
FIG. 2 shows a block diagram of a domain knowledge graph building system based on rules and deep learning in an embodiment of the present application.
Fig. 3 illustrates a flow diagram for deep learning based entity extraction in one embodiment of the present application.
Fig. 4 shows a relationship extraction flow chart based on deep learning in an embodiment of the present application.
Detailed Description
In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of one or more embodiments. It may be evident, however, that such embodiment(s) may be practiced without these specific details.
In the description of the present application, it should be noted that the terms "upper", "lower", and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, which are only for convenience in describing the present application and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation, and operate, and thus, should not be construed as limiting the present application. Unless expressly stated or limited otherwise, the terms "mounted," "connected," and "connected" are intended to be inclusive and mean, for example, that they may be fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meaning of the above terms in the present application can be understood by those of ordinary skill in the art as appropriate.
It is further noted that, in the description of the present application, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
At present, there are two main methods for extracting knowledge that are widely used: a rule-based knowledge extraction method and a deep learning-based knowledge extraction method.
The traditional rule-based knowledge extraction method comprises the following steps: the method comprises the steps of defining a character, grammar or semantic mode of a relation expressed in a text, and using the matching of the mode and the text as a main means to realize the acquisition of a relation example.
The traditional knowledge extraction method based on deep learning is as follows: by labeling each word in the sentence, entities in the sentence and relationships between pairs of entities are identified.
Both of these extraction methods have their own limitations. The rules play an important role in early knowledge extraction, the extraction based on the rules has higher accuracy, but needs experts to participate in the whole process, the generalization capability is low, only several manually recognized patterns can be matched, the exceeding pattern rules cannot be matched, namely the rules cannot be applied to phrases with the same semantics but different expressions, and therefore the rules are only suitable for extracting the knowledge with stronger structuralization. The deep learning can match two sentences with different expressions but similar semantics based on expression learning, the degree of automation is high, the manual participation is less, the deep learning seriously depends on the natural language processing labels such as part of speech labels and syntax analysis to provide classification features, a large number of errors often exist in the natural language processing labels, and the errors are continuously propagated and amplified in a knowledge extraction system, so that the knowledge extraction effect is finally influenced. The two extraction methods are shown in the following table:
therefore, in order to obtain a higher accuracy and a certain degree of automation, reduce the participation of experts, and solve the problem of low generalization ability, an embodiment of the present application provides a domain knowledge graph construction method based on rules and deep learning, as shown in fig. 1, including:
s1, inputting a document;
s2, classifying and screening the documents to obtain structured texts and unstructured texts;
in a specific embodiment, the S2 further includes:
s21, classifying the documents according to the document types to obtain classified documents;
generally, different document types have different document structures, and documents of the same type have the same document structure, that is, the titles are consistent, so preferably, the documents are classified according to the titles of the documents to obtain classified documents;
s22, screening the classified documents by using a text screening mode to screen out structured texts and unstructured texts;
preferably, some screening rules are manually made to screen the classified documents, usually rules made by experts; through observation, the same chapters in the same type of documents are either all unstructured or all structured, so that an unstructured text can be matched by adopting a keyword matching method;
further screening by utilizing the proportion of the structured text, preferably solving the proportion of the structured text by adopting a counting method; it should be noted that the judgment standard of the occupied specific gravity is evaluated by an expert;
preferably, if the proportion of the structured text is high, the knowledge map is constructed by directly performing rule-based knowledge extraction on the structured text.
S3, carrying out rule-based knowledge extraction on the structured text to obtain a first triple;
in a particular embodiment, the rule-based knowledge extraction includes: extracting knowledge by taking an article structure as a characteristic and extracting index knowledge in a professional field; the description of the entity 'index' in the document has the same rule characteristics, so that the extraction rule of the structured knowledge is formulated according to the definition of experts, and the extraction of the structured knowledge is realized;
wherein, the structure of the index is as follows: index name: and (4) index value.
S4, carrying out serialized labeling on the unstructured text to obtain a labeled text;
s5, performing knowledge extraction based on deep learning on the annotated text to obtain a second triple;
in a specific embodiment, the S5 further includes:
s51, performing entity extraction on the annotated text;
and S52, extracting the relation of the annotated text.
In an embodiment, taking "our love beijing tiananmen" as an example, the step of S51 is shown in fig. 3, and includes:
s511, preprocessing the 'love Beijing Tiananmen' through a BERT pre-training language model to obtain word vectors;
s512, processing the word vector through a bidirectional GRU module to obtain the context information of the entity;
s513, decoding the context information through a full-connection module to obtain the probability distribution of the entity;
s514, correcting the probability distribution through a conditional random field model;
and S515, extracting and classifying the entities to complete entity extraction.
The step of S52 is shown in fig. 4, and includes:
s521, preprocessing the 'love Beijing Tiananmen' through a BERT pre-training language model to obtain word vectors;
s522, processing the word vector through a bidirectional GRU module to obtain context information of an entity;
s523, acquiring a head entity vector and a tail entity vector of the relation through a multilayer perceptron module;
s524, judging the relation between the head entity vector and the tail entity vector through a bilinear model to complete relation extraction;
wherein, S in FIG. 2ij arcAn energy function for evaluating a head entity vector and the tail entity vector relationship.
And S6, constructing a domain knowledge graph according to the first triples and the second triples.
In a second aspect, an embodiment of the present application further provides a domain knowledge graph building system based on rules and deep learning, as shown in fig. 2, including:
the input module is used for inputting a document;
the classification and screening module is used for classifying and screening the documents to obtain structured texts and unstructured texts;
the rule extraction module is used for carrying out rule-based knowledge extraction on the structured text to obtain a first triple;
the labeling module is used for carrying out serialized labeling on the unstructured text to obtain a labeled text;
the deep learning extraction module is used for performing deep learning-based knowledge extraction on the labeled text to obtain a second triple;
and the output domain knowledge graph module is used for constructing a domain knowledge graph according to the first triad and the second triad.
In a specific embodiment, the deep learning module further includes:
the entity extraction module is used for performing entity extraction on the annotated text;
and the relation extraction module is used for extracting the relation of the labeled text.
In a specific embodiment, the entity extraction module further includes:
the BERT pre-processing module is used for processing the annotated text through a BERT pre-training language model to obtain a word vector;
the GRU module is used for processing the word vector to obtain the context information of the entity;
a full connection module, configured to decode the context information to obtain a probability distribution of the entity;
and the conditional random field module is used for correcting the probability distribution.
The relationship extraction module further comprises:
the BERT pre-processing module is used for processing the annotated text through a BERT pre-training language model to obtain a word vector;
the GRU module is used for processing the word vector to obtain the context information of the entity;
the multilayer perceptron module is used for acquiring a head entity vector and a tail entity vector of the relationship;
and the bilinear module is used for judging the relation between the head entity vector and the tail entity vector and finishing the relation extraction.
In a third aspect, an embodiment of the present application further provides a computer device, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor implements the above method when executing the program.
In a fourth aspect, an embodiment of the present application further provides a computer readable medium, on which a computer program is stored, which when executed by a processor implements the above method.
The method and the device realize the construction of the domain knowledge map by fusing the extraction method based on the rules and the deep learning, solve the problems of low generalization capability of the extraction method based on the rules and low accuracy of the extraction method based on the deep learning, can obtain higher accuracy and certain degree of automation, and reduce the participation of experts.
It should be understood that the above-mentioned embodiments of the present invention are only examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention, and it will be obvious to those skilled in the art that other variations or modifications may be made on the basis of the above description, and all embodiments may not be exhaustive, and all obvious variations or modifications may be included within the scope of the present invention.
Claims (10)
1. A domain knowledge graph construction method based on rules and deep learning is characterized by comprising the following steps:
s1, inputting a document;
s2, classifying and screening the documents to obtain structured texts and unstructured texts;
s3, carrying out rule-based knowledge extraction on the structured text to obtain a first triple;
s4, carrying out serialized labeling on the unstructured text to obtain a labeled text;
s5, performing knowledge extraction based on deep learning on the annotated text to obtain a second triple;
and S6, constructing a domain knowledge graph according to the first triples and the second triples.
2. The method according to claim 1, wherein the S2 further comprises:
s21, classifying the documents according to the document types to obtain classified documents;
s22, screening the classified documents by using a text screening mode to screen out structured texts and unstructured texts;
wherein the content of the first and second substances,
the document type is judged according to the document title.
3. The method according to claim 1, wherein the S3 further comprises:
making an extraction rule according to the definition of an expert and extracting;
the rule-based knowledge extraction includes: extracting knowledge by taking an article structure as a characteristic and extracting index knowledge in a professional field;
wherein, the structure of the index is as follows: index name: and (4) index value.
4. The method according to claim 1, wherein the S5 includes:
s51, performing entity extraction on the annotated text;
and S52, extracting the relation of the annotated text.
5. The method according to claim 4, wherein the S51 includes:
s511, preprocessing the annotated text through a BERT pre-training language model to obtain a word vector;
s512, processing the word vector through a bidirectional GRU module to obtain the context information of the entity;
s513, decoding the context information through a full-connection module to obtain the probability distribution of the entity;
s514, correcting the probability distribution through a conditional random field model;
and S515, extracting and classifying the entities to complete entity extraction.
6. The method according to claim 4, wherein the S52 includes:
s521, preprocessing the annotated text through a BERT pre-training language model to obtain a word vector;
s522, processing the word vector through a bidirectional GRU module to obtain context information of an entity;
s523, acquiring a head entity vector and a tail entity vector of the relation through a multilayer perceptron module;
and S524, judging the relation between the head entity vector and the tail entity vector through a bilinear model, and finishing the relation extraction.
7. A domain knowledge graph construction system based on rules and deep learning is characterized by comprising the following steps:
the input module is used for inputting a document;
the classification and screening module is used for classifying and screening the documents to obtain structured texts and unstructured texts;
the rule extraction module is used for carrying out rule-based knowledge extraction on the structured text to obtain a first triple;
the labeling module is used for carrying out serialized labeling on the unstructured text to obtain a labeled text;
the deep learning extraction module is used for performing deep learning-based knowledge extraction on the labeled text to obtain a second triple;
and the output domain knowledge graph module is used for constructing a domain knowledge graph according to the first triad and the second triad.
8. The system of claim 7, wherein the deep learning module further comprises:
the entity extraction module is used for performing entity extraction on the annotated text;
and the relation extraction module is used for extracting the relation of the labeled text.
9. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor,
the processor, when executing the program, implements the method of any of claims 1-6.
10. A computer-readable medium, having stored thereon a computer program,
the program when executed by a processor implementing the method according to any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111230908.8A CN114064915A (en) | 2021-10-22 | 2021-10-22 | Domain knowledge graph construction method and system based on rules and deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111230908.8A CN114064915A (en) | 2021-10-22 | 2021-10-22 | Domain knowledge graph construction method and system based on rules and deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114064915A true CN114064915A (en) | 2022-02-18 |
Family
ID=80235142
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111230908.8A Pending CN114064915A (en) | 2021-10-22 | 2021-10-22 | Domain knowledge graph construction method and system based on rules and deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114064915A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115114495A (en) * | 2022-08-31 | 2022-09-27 | 青岛民航凯亚系统集成有限公司 | Airworthiness data management auxiliary method and system based on deep learning |
| CN116090560A (en) * | 2023-04-06 | 2023-05-09 | 北京大学深圳研究生院 | Knowledge graph establishment method, device and system based on teaching materials |
-
2021
- 2021-10-22 CN CN202111230908.8A patent/CN114064915A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115114495A (en) * | 2022-08-31 | 2022-09-27 | 青岛民航凯亚系统集成有限公司 | Airworthiness data management auxiliary method and system based on deep learning |
| CN115114495B (en) * | 2022-08-31 | 2022-11-22 | 青岛民航凯亚系统集成有限公司 | Airworthiness data management auxiliary method and system based on deep learning |
| CN116090560A (en) * | 2023-04-06 | 2023-05-09 | 北京大学深圳研究生院 | Knowledge graph establishment method, device and system based on teaching materials |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110298033B (en) | Keyword corpus labeling training extraction system | |
| Kettinger et al. | The infological equation extended: towards conceptual clarity in the relationship between data, information and knowledge | |
| CN110727779A (en) | Question-answering method and system based on multi-model fusion | |
| CN114064918B (en) | Multi-modal event knowledge graph construction method | |
| CN110110054A (en) | A method of obtaining question and answer pair in the slave non-structured text based on deep learning | |
| CN109543034B (en) | Text clustering method and device based on knowledge graph and readable storage medium | |
| CN113127624B (en) | Question-answer model training method and device | |
| CN114064915A (en) | Domain knowledge graph construction method and system based on rules and deep learning | |
| CN110750648A (en) | Text emotion classification method based on deep learning and feature fusion | |
| WO2023040493A1 (en) | Event detection | |
| CN111625622B (en) | Domain ontology construction method and device, electronic equipment and storage medium | |
| CN111324691A (en) | Intelligent question-answering method for minority nationality field based on knowledge graph | |
| CN114580639A (en) | Knowledge graph construction method based on automatic extraction and alignment of government affair triples | |
| CN112650845A (en) | Question-answering system and method based on BERT and knowledge representation learning | |
| CN116383352A (en) | Knowledge graph-based method for constructing field intelligent question-answering system by using zero samples | |
| CN116244412A (en) | Multi-intention recognition method and device | |
| CN111666374A (en) | Method for integrating additional knowledge information into deep language model | |
| Samih et al. | Enhanced sentiment analysis based on improved word embeddings and XGboost. | |
| CN114238653A (en) | Method for establishing, complementing and intelligently asking and answering knowledge graph of programming education | |
| CN116522165B (en) | Public opinion text matching system and method based on twin structure | |
| CN111046168B (en) | Method, apparatus, electronic device and medium for generating patent summary information | |
| CN112417170A (en) | Relation linking method for incomplete knowledge graph | |
| CN111951079A (en) | Credit rating method and device based on knowledge graph and electronic equipment | |
| CN116522945A (en) | Model and method for identifying named entities in food safety field | |
| CN117009503A (en) | Text classification method and device |
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 |