CN110674274A - Knowledge graph construction method for food safety regulation question-answering system - Google Patents
Knowledge graph construction method for food safety regulation question-answering system Download PDFInfo
- Publication number
- CN110674274A CN110674274A CN201910898625.7A CN201910898625A CN110674274A CN 110674274 A CN110674274 A CN 110674274A CN 201910898625 A CN201910898625 A CN 201910898625A CN 110674274 A CN110674274 A CN 110674274A
- Authority
- CN
- China
- Prior art keywords
- food safety
- entity
- regulation
- knowledge graph
- regulations
- 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.)
- Granted
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
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- 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
- G06F16/35—Clustering; Classification
-
- 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
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/374—Thesaurus
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Human Computer Interaction (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a knowledge graph construction method for a question-answering system in the field of food safety regulations, which is used for acquiring food safety problems, classifying the food safety problems according to different properties of the problems, obtaining an irrevocable entity, constructing a problem management knowledge graph and constructing a field dictionary; obtaining food safety regulations, extracting regulation category information and chapter, strip and section information of the regulations as management classes of a regulation knowledge map, preprocessing, segmenting and part-of-speech tagging are carried out on text corpora of the regulations by utilizing a word segmentation tool and a TF-IDF method, and subject words are counted by word frequency to carry out entity tagging, entity relation extraction and attribute extraction so as to construct a regulation mode knowledge map; matching the entities in the obtained problem management type knowledge graph with the entities in the food safety regulation knowledge graph; and storing the acquired entities, entity relations and entity attributes in a Neo4j database for visualization. The method can accelerate the speed and the accuracy rate of information retrieval and answer extraction in a question-answering system.
Description
The technical field is as follows:
the invention relates to the technical field of artificial intelligence and computers, in particular to a knowledge graph construction method for a question-answering system in the field of food safety regulations.
Background art:
in order to ensure food safety and ensure public health and life safety, a series of food safety laws and regulations are made by the country and various departments, and 3500 food safety laws and regulations are currently used. The regulation of Chinese food safety law includes nearly 3 characters, which are divided into 154 regulations in 10 chapters, so that more food safety regulations aim at standardizing food safety in various links of food processing, selling, storing and transporting in the food field and guaranteeing personal safety of food consumers. However, many food safety regulations allow government regulators, food enterprise operators and food consumers to understand and master the content and specifications of the laws and regulations. For example, a person responsible for food safety management in a government department can quickly master the core of food safety regulations so as to supervise food enterprises; the food enterprise operator can master laws and regulations and operate the enterprise according to the requirements of the laws and regulations so as to avoid the legal risk; consumers can know food safety laws and protect own rights and interests; correctly combing the relationship between laws and regulations: including hierarchical relationships, category relationships, extended relationships, and the like. For example, they are guiding and compendial laws and regulations, and they are legal regulations and regulations. Some regulations emphasize which type of food is managed, which type of food is processed, and which type of food is regulated and safe for food materials. The core content and the regulation content of each type and each food safety regulation can be rapidly known, and the problems of establishment basis, development direction, trend and the like of the food safety regulations can be answered, so that the establishment of an intelligent food safety regulation management system is necessary.
With the rapid development and popularization of network technologies, people increasingly use networks to seek help and knowledge sharing, however, when a user uses a traditional search engine, most of feedback is a large number of webpages, the webpages also contain a lot of contents irrelevant to query intentions, and the webpages also have a lot of misleading information, which is an understanding opinion of some individuals on laws, but the authorities peculiar to laws do not allow the individuals to read laws and regulations, so that a question-answering system in the field of food safety regulations is built, and specific regulations are used as questions and answers of people, so that the authority of laws can be maintained, the actual life problems of common people can be effectively solved, and great convenience is brought to the lives of people.
The knowledge base of the question-answering system has a variety of possible data sources. Conventional data sources include web documents, search engines, encyclopedia descriptions, question-and-answer communities, and the like. Without exception, these data sources are unstructured plain text data. There are a number of information retrieval based methods that are dedicated to the study of knowledge extraction and answers from plain text data. In recent years, knowledge-graph-based question-answering systems have become a hot direction for research and application in various industries. Knowledge maps express knowledge in triplets (entities, relationships, attribute values) in a human-understandable organizational format, and are called knowledge maps because they use graphs as data structures to represent knowledge. The concept and entity of the objective world or their attribute value are represented by nodes of the graph, the concept and the actual relation or attribute are represented by edges between the nodes, and the node-edge-node constitutes statement sentences representing knowledge and facts. And the knowledge map is used for expressing knowledge and facts of an objective world at a semantic level, so that various intelligent applications can be established, and the method has the characteristics of integration and accumulation. The question-answering system constructed based on the knowledge graph has the following advantages in data: (1) solving the problem of intelligent degree of semantic understanding by using the data association degree; (2) solving the question of answer accuracy by using data precision; (3) the problem retrieval efficiency is improved by using the data structuring of the triples.
In order to make the knowledge graph better adapt to the question-answering system, the obtained questions are classified from the perspective of users, and are further refined into question subclasses according to the classification of the questions, and the questions can be continuously refined according to the nature of the questions until the questions are irrevocable. On the basis of classification, the classes are utilized to construct a question management class knowledge graph and a rule mode knowledge graph to establish a matching relation, and a knowledge graph capable of answering questions is provided.
The invention content is as follows:
the invention aims to provide a method for constructing a knowledge graph of a question-answering system in the field of food safety regulations. In order to solve the problems and enable the constructed knowledge graph to better serve the question-answering system, the embodiment of the invention provides a knowledge graph construction method for the question-answering system in the field of food safety regulations, which comprises the following steps:
the method comprises the steps of obtaining food safety regulation problems proposed by people on a food safety regulation authority website, and filtering subject matters of question sentences, so that the problems are classified from the perspective of users and are divided into 10 categories including food safety risk monitoring, food safety standards, food production and management, food inspection, food import and export, food safety accident handling, supervision and management, legal responsibility and general knowledge of food safety regulations.
According to the obtained problem classification, setting the problem classification as a primary classification standard, further dividing primary classification results to obtain secondary, tertiary or quaternary classification results in order to classify the problems to specific entities, until irreparable entities are obtained, and constructing a problem management class knowledge graph;
according to the obtained problem classification result, dividing the field by taking the first-level classification as a reference, and counting the specific entity names of the rest of each-level classification to be used as a dictionary of the field;
downloading the current effective food safety regulation in China on a food safety regulation website to be used as basic corpus for constructing a food safety regulation knowledge map;
according to the downloaded laws and regulations, classifying and managing the laws and regulations according to two aspects of a law management department level and a related field level, wherein the management department level can be divided into a country level, a related department level, a province level, a city level and a district (county), the field level can be divided into a food category, a check category, a standard category and the like, and a management law body class is constructed;
displaying the content of the laws and regulations according to the downloaded laws and regulations and the level sequence of the names of the laws and regulations, wherein one part of the laws and regulations is divided into a plurality of chapters, each chapter comprises a plurality of sections, each section comprises a plurality of laws, each law comprises a plurality of entities, and the management body class of each law is constructed;
extracting each rule keyword by using a TF-IDF statistical method according to the downloaded basic corpus texts of a plurality of rules, extracting the name, chapter name and title of each rule by using a regular expression to match the Chinese character string with the characteristics of the rule document, and using the obtained keyword and the name and title obtained by regular matching as a dictionary in the field of food safety rules;
loading the obtained domain dictionary into a word segmentation tool, and performing word removal, word segmentation and part-of-speech tagging on each rule basic corpus by using a jieba word segmentation tool;
according to the obtained domain dictionary, entity extraction and relationship extraction based on a dependency relationship model are utilized, and according to the obtained part-of-speech labeled regulation document, dependency relationship paths among different entities are extracted, and meanwhile, the entities and the relationships are extracted;
and according to the information such as the entity, the relation, the entity attribute and the like obtained by information extraction, carrying out knowledge fusion on the information: combining entity link and knowledge, eliminating concept ambiguity through knowledge fusion, and eliminating redundant and wrong concepts;
constructing a knowledge graph of each rule according to the obtained entity, relationship and attribute subjected to knowledge fusion and the obtained rule body class;
according to the obtained entities, relations and attributes, associating the management type knowledge graph obtained by problem classification with the food safety regulation knowledge graph through entity similarity calculation, and thus constructing a knowledge graph for a food safety regulation question-answering system;
storing the obtained entities, entity relations and entity attributes in the knowledge graph into a Neo4j database to realize visualization;
preferably, the method comprises the steps of extracting specified rule keywords by using a TF-IDF statistical method, converting a text into a csv format, performing weight calculation by using a TF-IDF algorithm through constructing a word frequency matrix, sequencing the calculated weights, and selecting the weights with larger weights as the keywords;
the TF-IDF algorithm is calculated as follows:
TF-IDF＝TF×IDF
preferably, the rule document features are matched with the regular expression to obtain the chapter name of the food safety rule as the corpus content of the food safety rule, wherein the regular expression is a logic formula for operating on the character string, that is, a certain specific character and a combination of the specific characters are defined in advance to form a "rule character string", and the "rule character string" is used for expressing a filtering logic for the character string.
Preferably, the named entity identification and entity relationship extraction are performed, and a dependency relationship path between different parts of speech is extracted according to the obtained part of speech tagged rule document based on a dependency relationship model;
for the relation extraction problem, in the relation extraction model based on the dependency relation, the relation words are not preset categories, but exist in the current sentence. The relationship extraction algorithm based on the dependency relationship comprises the following steps:
(1) obtaining a sentence;
(2) constructing a dependency relationship path, and labeling the path according to a dependency labeling table;
(3) extracting core words;
(4) constructing a bingo structure, and finding a subject and an object as two entities by taking the core words as the relationship.
It can be found that not only the relationship can be extracted based on the dependency relationship, but also the corresponding entity can be extracted.
Preferably, the words in the domain dictionary obtained by classification are stored as entities, entity alignment is performed on the words and the entities obtained by named entity recognition through entity similarity calculation, and the entity similarity is calculated by using a SingeLinkage algorithm in a hierarchical clustering method. Establishing a relation between the problem management knowledge graph and the regulation knowledge graph;
preferably, the hierarchical clustering Singe Linkage algorithm is to input N physical objects to be calculated and a distance matrix of N × N, each class contains only one object, the distance between the classes is the distance between the objects contained in the classes, then the calculated values are sorted by calculating the distance between each class, the two classes with the closest distance are merged, so that the total class is one less, the physical objects in the two classes are merged together, and then the distance between the new class and the old class is recalculated.
The distance calculation formula is defined as follows:
wherein C isi，CjBeing two different classes, distmin(Ci，Cj) Denotes the minimum of the distance between classes, p being of CiEntity objects in a class, p' being of the type CjIs the distance between the entity objects in the two categories, | p-p' |.
Drawings
FIG. 1 is a flow chart illustrating a method for constructing a knowledge graph for a question-and-answer system in the field of food safety regulations according to a preferred embodiment of the present invention;
FIG. 2 is a schematic diagram of a problem classification part of a knowledge graph construction method for a question-answering system in the field of food safety regulations according to a preferred embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating a detailed classification of a food safety standard part of a knowledge map construction method for a question-answering system in the field of food safety regulations according to a preferred embodiment of the present invention;
fig. 4 is a fourth chapter of the food production and management part regulation "food safety law of the people's republic of china" for a method for constructing a knowledge graph of a question-answer system in the field of food safety regulations according to a preferred embodiment of the present invention: a schematic representation of a knowledge map of the general contents of "food vendors" in the first general section of food production operations;
FIG. 5 is a schematic view illustrating classification management of food safety regulations with respect to a method for constructing a knowledge map of a question-and-answer system in the field of food safety regulations according to a preferred embodiment of the present invention;
fig. 6 is a schematic view of the management class of the food safety regulation itself according to a method for constructing a knowledge map of a question-and-answer system in the field of the food safety regulation according to a preferred embodiment of the present invention.
Detailed Description
The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
Fig. 1 is a schematic flow chart of a method for constructing a knowledge graph of a question-answering system in the field of food safety regulations according to a preferred embodiment of the present invention, and as shown in fig. 1, the present invention provides a method for constructing a knowledge graph of a question-answering system in food safety regulations, including:
s101, obtaining food safety problems, classifying the food safety problems according to different problems, thinning the problems step by step until irreparable entities are obtained, constructing a management class knowledge graph and constructing a domain dictionary;
s102, obtaining the current effective food safety regulation in China on a website for downloading the food safety regulation, not only obtaining the management class of the regulation category information and the chapter, bar and section information regulation knowledge map of the regulation, namely the body class of the management regulation, but also preprocessing, segmenting and part-of-speech tagging the text corpus of the regulation by using a word segmentation tool and a TF-IDF method, counting subject words by word frequency, and then performing entity tagging, entity relationship extraction, attribute extraction and construction of the regulation knowledge map;
s103, matching the entities in the problem management type knowledge graph and the entities in the food safety regulation knowledge graph through entity similarity calculation;
s104, storing the obtained entities, entity relationships and entity attributes into a Neo4j database to realize visualization;
in this embodiment, the food safety laws, rules and standards are downloaded and obtained from the official network of the relevant national department, and the contents are serious and proper, so that the food safety laws, rules and standards can be used as the entity concept of the knowledge graph of the food safety laws and rules.
In this embodiment, the food safety regulation questions put forward by people on the food safety regulation authority website are crawled, filtered, and classified by judging the subject purpose of the question sentence, as shown in fig. 2, into 10 categories of food safety risk monitoring, food safety standards, food production and management, food inspection, food import and export, food safety accident handling, supervision and management, legal responsibility, and general knowledge of food safety regulations;
in this embodiment, the problem classification result is set as a primary classification standard according to the problem classification result, and in order to obtain a specific entity, the primary classification result is refined step by step to obtain a secondary, tertiary or quaternary classification result until a non-differentiable entity is obtained;
in the embodiment, according to the obtained problem classification result, the domain is divided by taking the first-level classification as a reference, and the specific entity names of the rest of each-level classification are counted to be used as the domain dictionary;
in the embodiment, the obtained food safety regulation document is obtained, a specified regulation keyword is extracted by using a TF-IDF statistical method, the text is converted into a csv format, a word frequency matrix is constructed, weight calculation is performed by using a TF-IDF algorithm, the calculated weights are sequenced, and the weights with larger weights are selected as the keywords;
in this embodiment, the names of the laws and regulations, the chapter names, the department names, and the like in the food safety laws and regulations are obtained by regular expression matching using the characteristics of the laws and regulations documents, and the names obtained by matching are used as a dictionary in the field of the food safety laws and regulations;
in this embodiment, the downloaded food safety regulations classify and manage the regulations from two aspects, namely, a regulation management department level and a related field level, wherein the management department level may be classified into a country level, a related department, a province, a city, and a district (county), the field level may be classified into a food category, a test category, a standard category, and the like, and specific examples are shown in fig. 5;
in this embodiment, the downloaded food safety regulations show the contents of the laws and regulations according to the order of the names of the regulations, the names of the chapters and the names of the regulations, wherein one regulation is divided into a plurality of chapters, each chapter includes a plurality of sections, each section includes a plurality of laws, each regulation includes a plurality of entities, and a management body class of each regulation is constructed, for example, as shown in fig. 6;
in this embodiment, according to the obtained dictionary in the food safety regulation field, the operation of word segmentation, part of speech tagging and the like is performed on the food safety regulation document by a jieba word segmentation tool and loading the dictionary in the field;
in this embodiment, the dependency relationship path between different parts of speech is extracted based on the relationship extraction of the dependency relationship model according to the part of speech tagged rule document, the dependency relationship path is constructed, and the path is tagged according to the dependency tagging table; extracting core words; constructing a bingo structure, and finding a subject and an object as two entities by taking the core words as the relationship. Therefore, it can be found that not only the relationship can be extracted based on the dependency relationship, but also the corresponding entity can be extracted.
In this embodiment, the knowledge graph of each rule is constructed by retaining the attribute of the chapter in the rule according to the corpus of the rule subjected to entity extraction and attribute extraction;
in this embodiment, the entities in the problem management class knowledge graph and the entities in the food safety regulation knowledge graph calculate the entity similarity by using the Singe link algorithm in the hierarchical clustering method, associate the related entities, wherein the Singe link algorithm is a distance matrix into which N entity objects and N × N are input, each class includes only one object, and the distance between the classes is the distance between the objects included in the classes, and then sort the calculated values by calculating the distance between the classes, merge the two classes closest to each other, and thus the entity objects in the two classes are merged together, thereby constructing the knowledge graph for the food safety regulation question-answer system.
The invention provides a method for constructing a knowledge graph of a question-answering system in the field of food safety regulations, which is used for acquiring questions proposed by the question-answering system, firstly classifying the questions according to divided categories, matching the questions to specific entities in a first-level and first-level downward manner, and displaying the relations and attributes of the entities through the knowledge graph, as shown in figure 3, when the questions are about the use standard of hydrogen peroxide or sulfur, the hydrogen peroxide entities and the sulfur entities can be found by judging that the questions are under the bleaching agent under the use standard of an additive under the food safety standard, because the two entities are respectively corresponding to the regulatory entities: the food additive use standards are related, so that the related information of the two additives in the food additive use standards can be clearly displayed through a knowledge graph. According to the mode, the knowledge map of the food safety regulation question-answering system is constructed, and the question-answering system is more effectively served for information retrieval.
Finally, the method of the present invention is only a preferred embodiment and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (7)
1. A knowledge graph construction method for a question-answering system in the field of food safety regulations is characterized by comprising the following steps:
the method comprises the steps of obtaining food safety regulation problems proposed by people on a food safety regulation authority website, and filtering subject matters of question sentences, so that the problems are divided into 10 categories including food safety risk monitoring, food safety standards, food production and management, food inspection, food import and export, food safety accident handling, supervision and management, legal responsibility and general knowledge of food safety regulations;
setting a primary classification result according to the obtained problem classification, further dividing the primary classification result to obtain a secondary, tertiary or quaternary classification result in order to classify the problem to a specific entity, and constructing a problem management class knowledge graph until an irrevocable entity is obtained;
obtaining the current effective food safety regulation in China on a downloaded food safety regulation website, and obtaining regulation category information and information such as chapter, bar, section and the like of the regulation as the management class of a regulation knowledge map, namely the body class of the management regulation;
obtaining the chapter, strip and section names of each rule through regular matching by utilizing the orderliness of each rule chapter;
preprocessing text corpora of laws and regulations, segmenting words, marking parts of speech by using a word segmentation tool and a TF-IDF method, counting subject words by word frequency, and then performing entity marking, entity relation extraction, attribute extraction and construction of knowledge maps of laws and regulations;
matching the entities in the problem management type knowledge graph and the entities in the food safety regulation knowledge graph through entity similarity calculation, and connecting the two types of knowledge graphs;
and storing the obtained entities, entity relations and entity attributes of the knowledge graph into a Neo4j database to realize visualization.
2. The method for constructing a knowledge graph of a food safety regulation question-answering system according to claim 1, wherein classification results are obtained, fields are divided based on the obtained problem classification results by taking the first-level classification thereof as a reference, and specific entity names of the remaining classes are counted as a dictionary of the fields.
3. The method for constructing a knowledge graph of a question-answer system for food safety regulations according to claim 1, wherein keywords of specified regulations are extracted by a TF-IDF statistical method, a text is converted into a csv format, a word frequency matrix is constructed, weight calculation is performed by a TF-IDF algorithm, the calculated weights are ranked, and a keyword with a larger weight is selected. The TF-IDF algorithm is calculated as follows:
TF-IDF＝TF×IDF
4. the method for constructing a knowledge graph of a food safety regulation question-answering system according to claim 1, wherein a character string matching pattern is described by regular expression matching using the self characteristics of regulation documents, the character string matching pattern can be used for checking whether a certain substring is contained in a character string, and the matched regulation chapter name is used as a dictionary in the field of food safety regulations.
5. The method as claimed in claim 1, wherein the method comprises entity extraction and relationship extraction based on a dependency model, and according to the obtained part-of-speech labeled rule document, extracting a dependency path between different entities and extracting entities and relationships.
6. The method as claimed in claim 1, wherein the entity similarity between the entities in the problem management class knowledge map and the entities in the rule knowledge map is calculated, the entity similarity is calculated by using a Singe link algorithm in a hierarchical clustering method, the Singe link algorithm is a distance matrix that inputs N entity objects to be calculated and N × N, each class contains only one object, and the distance between classes is the distance between the objects contained in the classes, and then the calculated distance values are sorted by calculating the distance between each class, and the two closest classes are merged, that is, the two similar entities are merged.
7. The method for constructing the knowledge graph of the food safety regulation question-answering system according to claim 1, wherein the constructed knowledge graph is displayed by using a visualization tool, and the visualization tool comprises a Neo4j graphic database.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910898625.7A CN110674274B (en) | 2019-09-23 | 2019-09-23 | Knowledge graph construction method for food safety regulation question-answering system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910898625.7A CN110674274B (en) | 2019-09-23 | 2019-09-23 | Knowledge graph construction method for food safety regulation question-answering system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110674274A true CN110674274A (en) | 2020-01-10 |
| CN110674274B CN110674274B (en) | 2022-07-01 |
Family
ID=69077523
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910898625.7A Active CN110674274B (en) | 2019-09-23 | 2019-09-23 | Knowledge graph construction method for food safety regulation question-answering system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110674274B (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111708874A (en) * | 2020-08-24 | 2020-09-25 | 湖南大学 | Man-machine interaction question-answering method and system based on intelligent complex intention recognition |
| CN111737489A (en) * | 2020-06-17 | 2020-10-02 | 广联达科技股份有限公司 | Building information retrieval method, device, equipment and readable storage medium |
| CN111782825A (en) * | 2020-08-20 | 2020-10-16 | 支付宝(杭州)信息技术有限公司 | Knowledge base construction method and device |
| CN111930793A (en) * | 2020-06-26 | 2020-11-13 | 西安电子科技大学 | Target behavior mining and retrieval analysis method, system, computer equipment and application |
| CN112100405A (en) * | 2020-09-23 | 2020-12-18 | 中国农业大学 | Veterinary drug residue knowledge graph construction method based on weighted LDA |
| CN112487205A (en) * | 2020-12-03 | 2021-03-12 | 湖北大学 | Food safety risk knowledge graph supporting hazard identification and construction method |
| CN112559765A (en) * | 2020-12-11 | 2021-03-26 | 中电科大数据研究院有限公司 | Multi-source heterogeneous database semantic integration method |
| CN113010694A (en) * | 2021-04-19 | 2021-06-22 | 华北电力大学 | Regular expression-based relay protection defect text proper noun dictionary construction method |
| CN113254594A (en) * | 2021-06-21 | 2021-08-13 | 国能信控互联技术有限公司 | Smart power plant-oriented safety knowledge graph construction method and system |
| CN113407688A (en) * | 2021-06-15 | 2021-09-17 | 西安理工大学 | Method for establishing knowledge graph-based survey standard intelligent question-answering system |
| CN114186561A (en) * | 2021-10-20 | 2022-03-15 | 福建亿榕信息技术有限公司 | Electronic file association analysis method and system based on knowledge graph |
| CN114547324A (en) * | 2021-12-31 | 2022-05-27 | 华中农业大学 | Method, system and medium for constructing body in national standard field of food safety |
| CN115759256A (en) * | 2022-11-24 | 2023-03-07 | 中安华邦(北京)安全生产技术研究院股份有限公司 | Method, system, medium and equipment for constructing safety production digital knowledge base |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108804521A (en) * | 2018-04-27 | 2018-11-13 | 南京柯基数据科技有限公司 | A kind of answering method and agricultural encyclopaedia question answering system of knowledge based collection of illustrative plates |
| CN109408811A (en) * | 2018-09-29 | 2019-03-01 | 联想(北京)有限公司 | A kind of data processing method and server |
| CN110209787A (en) * | 2019-05-29 | 2019-09-06 | 袁琦 | A kind of intelligent answer method and system based on pet knowledge mapping |
-
2019
- 2019-09-23 CN CN201910898625.7A patent/CN110674274B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108804521A (en) * | 2018-04-27 | 2018-11-13 | 南京柯基数据科技有限公司 | A kind of answering method and agricultural encyclopaedia question answering system of knowledge based collection of illustrative plates |
| CN109408811A (en) * | 2018-09-29 | 2019-03-01 | 联想(北京)有限公司 | A kind of data processing method and server |
| CN110209787A (en) * | 2019-05-29 | 2019-09-06 | 袁琦 | A kind of intelligent answer method and system based on pet knowledge mapping |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111737489A (en) * | 2020-06-17 | 2020-10-02 | 广联达科技股份有限公司 | Building information retrieval method, device, equipment and readable storage medium |
| CN111930793A (en) * | 2020-06-26 | 2020-11-13 | 西安电子科技大学 | Target behavior mining and retrieval analysis method, system, computer equipment and application |
| CN111782825A (en) * | 2020-08-20 | 2020-10-16 | 支付宝(杭州)信息技术有限公司 | Knowledge base construction method and device |
| CN111708874A (en) * | 2020-08-24 | 2020-09-25 | 湖南大学 | Man-machine interaction question-answering method and system based on intelligent complex intention recognition |
| CN112100405A (en) * | 2020-09-23 | 2020-12-18 | 中国农业大学 | Veterinary drug residue knowledge graph construction method based on weighted LDA |
| CN112100405B (en) * | 2020-09-23 | 2024-01-30 | 中国农业大学 | Veterinary drug residue knowledge graph construction method based on weighted LDA |
| CN112487205B (en) * | 2020-12-03 | 2022-04-29 | 湖北大学 | Food safety risk knowledge graph supporting hazard identification and construction method |
| CN112487205A (en) * | 2020-12-03 | 2021-03-12 | 湖北大学 | Food safety risk knowledge graph supporting hazard identification and construction method |
| CN112559765A (en) * | 2020-12-11 | 2021-03-26 | 中电科大数据研究院有限公司 | Multi-source heterogeneous database semantic integration method |
| CN112559765B (en) * | 2020-12-11 | 2023-06-16 | 中电科大数据研究院有限公司 | Semantic integration method for multi-source heterogeneous database |
| CN113010694A (en) * | 2021-04-19 | 2021-06-22 | 华北电力大学 | Regular expression-based relay protection defect text proper noun dictionary construction method |
| CN113407688A (en) * | 2021-06-15 | 2021-09-17 | 西安理工大学 | Method for establishing knowledge graph-based survey standard intelligent question-answering system |
| CN113407688B (en) * | 2021-06-15 | 2022-09-16 | 西安理工大学 | Method for establishing knowledge graph-based survey standard intelligent question-answering system |
| CN113254594A (en) * | 2021-06-21 | 2021-08-13 | 国能信控互联技术有限公司 | Smart power plant-oriented safety knowledge graph construction method and system |
| CN113254594B (en) * | 2021-06-21 | 2022-01-14 | 国能信控互联技术有限公司 | Smart power plant-oriented safety knowledge graph construction method and system |
| CN114186561A (en) * | 2021-10-20 | 2022-03-15 | 福建亿榕信息技术有限公司 | Electronic file association analysis method and system based on knowledge graph |
| CN114547324A (en) * | 2021-12-31 | 2022-05-27 | 华中农业大学 | Method, system and medium for constructing body in national standard field of food safety |
| CN114547324B (en) * | 2021-12-31 | 2023-02-14 | 华中农业大学 | Method, system and medium for constructing body in national standard field of food safety |
| CN115759256A (en) * | 2022-11-24 | 2023-03-07 | 中安华邦(北京)安全生产技术研究院股份有限公司 | Method, system, medium and equipment for constructing safety production digital knowledge base |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110674274B (en) | 2022-07-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110674274B (en) | Knowledge graph construction method for food safety regulation question-answering system | |
| CN116628172B (en) | Dialogue method for multi-strategy fusion in government service field based on knowledge graph | |
| CN109684448B (en) | Intelligent question and answer method | |
| Inzalkar et al. | A survey on text mining-techniques and application | |
| Kaushik et al. | A comprehensive study of text mining approach | |
| Mansouri et al. | Named entity recognition approaches | |
| US6728695B1 (en) | Method and apparatus for making predictions about entities represented in documents | |
| Tiun et al. | Automatic topic identification using ontology hierarchy | |
| CN111475623A (en) | Case information semantic retrieval method and device based on knowledge graph | |
| CN110968699A (en) | Logic map construction and early warning method and device based on event recommendation | |
| CN112395395B (en) | Text keyword extraction method, device, equipment and storage medium | |
| CN108595525B (en) | Lawyer information processing method and system | |
| Daryani et al. | An automated resume screening system using natural language processing and similarity | |
| CN113239163A (en) | Intelligent question-answering method and system based on traffic big data | |
| CN110019733A (en) | A kind of intelligent answer method and device of Community-oriented | |
| CN115292450A (en) | Data classification field knowledge base construction method based on information extraction | |
| CN114138979B (en) | Cultural relic safety knowledge map creation method based on word expansion unsupervised text classification | |
| Ribeiro et al. | Discovering IMRaD structure with different classifiers | |
| Long | An agent-based approach to table recognition and interpretation | |
| Devisree et al. | A hybrid approach to relationship extraction from stories | |
| Schumann et al. | Query-Based Retrieval of German Regulatory Documents for Internal Auditing Purposes | |
| CN111767716B (en) | Method and device for determining enterprise multi-level industry information and computer equipment | |
| Rădescu | Automatic adaptive system for filtering fake news by methods based on artificial intelligence using machine learning | |
| Hartadi et al. | Punishment provision extraction from Indonesian law texts with knowledge acquisition rules | |
| Gupta et al. | Impact of Deep Learning on Semantic Sentiment Analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |