CN112635078A - Traditional Chinese medicine knowledge graph construction and visualization method - Google Patents
Traditional Chinese medicine knowledge graph construction and visualization method Download PDFInfo
- Publication number
- CN112635078A CN112635078A CN202011232041.5A CN202011232041A CN112635078A CN 112635078 A CN112635078 A CN 112635078A CN 202011232041 A CN202011232041 A CN 202011232041A CN 112635078 A CN112635078 A CN 112635078A
- Authority
- CN
- China
- Prior art keywords
- traditional chinese
- chinese medicine
- knowledge
- entities
- knowledge graph
- 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
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H70/00—ICT specially adapted for the handling or processing of medical references
- G16H70/40—ICT specially adapted for the handling or processing of medical references relating to drugs, e.g. their side effects or intended usage
-
- 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
- G06F16/34—Browsing; Visualisation therefor
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Medicinal Chemistry (AREA)
- Pharmacology & Pharmacy (AREA)
- Toxicology (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a traditional Chinese medicine knowledge graph construction and visualization method, which comprises the following steps: creating a mode of the knowledge graph; converting traditional Chinese medicine information in the relational database into RDF data; the entity alignment is realized according to the attribute similarity matching of the two entities, and the two entities are combined into one entity, so that the repeated redundancy of the entities is avoided; generating an own file by the aligned triple table, and importing the own file into a visualization tool project to form a knowledge map system; calculating weight information of the decoction pieces in the prescription by using an algorithm; generating an acupoint effect knowledge map by using the svg vector diagram; and building a web project. The traditional Chinese medicine knowledge graph construction and visualization method provides a further way for a user to acquire and view traditional Chinese medicine knowledge, can visually display huge traditional Chinese medicine data to the user, is well organized, and can trace back and source according to graph content and view association among data.
Description
Technical Field
The invention belongs to the technical field of knowledge maps and visualization, and particularly relates to a traditional Chinese medicine knowledge map construction and visualization method.
Background
In recent years, more and more traditional Chinese medicine databases, electronic resources and the like can be conveniently acquired on the internet, and how to acquire more comprehensive, more accurate and more authoritative traditional Chinese medicine knowledge also becomes a research requirement in the field of traditional Chinese medicine. The knowledge map is a huge and networked knowledge system which is constructed by taking a semantic network as a framework, can capture and present semantic relations among domain concepts, enables trivial and scattered knowledge on the Internet to be connected with one another, and supports intelligent applications such as comprehensive knowledge retrieval, question answering and decision support and the like. The traditional Chinese medicine knowledge graph is a knowledge graph system which is mainly constructed by taking a traditional Chinese medicine language system as a framework, and the content is filled by taking the existing database resources as the knowledge graph. The visualized semantic graph can visually express the association between the domain concepts, and a user can browse the domain concepts in an interactive mode and select a certain concept to start constructing a query or search. The three are connected in series to form a traditional Chinese medicine knowledge map and data visualization, and the application of the method can meet the comprehensive and accurate requirements of authoritative, comprehensive and accurate traditional Chinese medicine knowledge in the field of traditional Chinese medicine. Foreign scientists have achieved outstanding achievements in the field of information visualization, the university of Dereksel and the institute of scientific information in the United states are the main sites of information visualization research, and a great number of scientists have made great progress in the research of information visualization and the development and application of related tools. Foreign scholars also generalize the research of the current knowledge graph, and the knowledge graph is mainly applied to displaying the evolution condition of the visual knowledge field, visually analyzing and retrieving results, displaying the whole structure of the knowledge in the field, integrally holding subject knowledge, holding the progress of the rapidly changing knowledge field and the like. The traditional Chinese medicine has a long history and long source, and the value of the traditional Chinese medicine is proved by clinical practice and has been paid attention by Eurasian countries such as Japan, Korea, Singapore and the like, so that the traditional Chinese medicine has a very wide development prospect.
The existing traditional Chinese medicine database has incomplete data display and insufficient contact between traditional Chinese medicine knowledge. At present, Chinese medicine shows huge effect which is not in accordance with the ethical ratio in the aspect of treating difficult and complicated diseases such as hepatitis B, hemiplegia, rheumatism, new coronary pneumonia and the like. However, there are some defects of disordered data, unclear order, and mixed medical book and book references in TCM. How to organize and clarify traditional Chinese medicine data becomes an urgent task. Nowadays, computer technology is rapidly developed, and numerous technologies such as knowledge maps, data visualization, artificial intelligence, big data and the like are derived. The computer technology can solve the disadvantages of the traditional Chinese medicine data.
Disclosure of Invention
Based on the defects of the prior art, the technical problem to be solved by the invention is to provide a traditional Chinese medicine knowledge map construction and visualization method, based on the characteristic that the knowledge map can capture semantic relations among concepts, the knowledge in the field of traditional Chinese medicine is constructed into a map for visualization, the retrieval accuracy of a traditional Chinese medicine database is improved, and the comprehensiveness and connectivity among traditional Chinese medicine knowledge are enhanced.
In order to solve the technical problems, the invention is realized by the following technical scheme:
the invention provides a traditional Chinese medicine knowledge graph construction and visualization method, which comprises the following steps:
step 1: creating a mode of the knowledge graph;
step 2: converting traditional Chinese medicine information in the relational database into RDF data;
and step 3: the entity alignment is realized according to the attribute similarity matching of the two entities, and the two entities are combined into one entity, so that the repeated redundancy of the entities is avoided;
and 4, step 4: generating an own file by the aligned triple table, importing the own file into a visualization tool project to form a knowledge graph system, and visually displaying the relation among the prescriptions, the decoction pieces, the traditional Chinese medicines, the processing of the medicinal materials and the base sources of the medicinal materials for a user;
and 5: calculating weight information of the decoction pieces in the prescription by using an algorithm;
step 6: generating an acupoint effect knowledge map by using the svg vector diagram;
and 7: and building a web project.
Optionally, in step S1, the method includes the following steps:
step 1.1: class hierarchy: the atlas adopts a two-layer class hierarchy structure;
step 1.2: class relationship definition: the classes have mutual relations, and the classes can define a one-way relation and a two-way relation;
step 1.3: class domain definition: defining multiple realms facilitates group management of classes.
Further, in step S3, the method includes the following steps:
step 3.1: performing weight evaluation on the attributes;
step 3.2: evaluating the similarity between the attribute values;
step 3.3: and (3) summing the weight calculated in the step (3.1) and the attribute similarity calculated in the step (3.2), defining the sum as the similarity of the two entities, setting a threshold value, and classifying according to the range.
Therefore, the traditional Chinese medicine knowledge graph construction and visualization method provides a further way for users to acquire and view traditional Chinese medicine knowledge, can visually display huge traditional Chinese medicine data to the users, is well organized, and can trace back to the source according to graph content and view the association among the data.
The foregoing description is only an overview of the technical solutions of the present invention, and in order to make the technical means of the present invention more clearly understood, the present invention may be implemented in accordance with the content of the description, and in order to make the above and other objects, features, and advantages of the present invention more clearly understood, the following detailed description is given in conjunction with the preferred embodiments, together with the accompanying drawings.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings of the embodiments will be briefly described below.
FIG. 1 is a functional mode diagram of a Chinese medicine knowledge base map visualization system;
FIG. 2 is a flow chart of a traditional Chinese medicine knowledge graph construction and visualization method;
FIG. 3 is a flow chart of a method of entity alignment;
FIG. 4 is a diagram of the effect of a knowledge graph of traditional Chinese medicine;
FIG. 5 is a diagram of an effect of svg vector diagram presentation;
FIG. 6 is a diagram of a login registration entry;
FIG. 7 is a diagram of a registration interface;
FIG. 8 is a top page view of a traditional Chinese medicine knowledge graph construction and visualization system;
FIG. 9 is a medical literature knowledge question and answer page diagram;
FIG. 10 is a medical document review success page diagram;
FIG. 11 is a medical literature query page diagram;
FIG. 12 is an encyclopedia page view of acupuncture points of a human body;
FIG. 13 is a medical record mining encyclopedia page view;
FIG. 14 is a diagram of an encyclopedia page of pharmacological actions;
FIG. 15 is a drug pair map page view;
FIG. 16 is a protein map.
Detailed Description
Other aspects, features and advantages of the present invention will become apparent from the following detailed description, taken in conjunction with the accompanying drawings, which form a part of this specification, and which illustrate, by way of example, the principles of the invention. In the referenced drawings, the same or similar components in different drawings are denoted by the same reference numerals.
As shown in fig. 1 to 3, the method for constructing and visualizing a traditional Chinese medicine knowledge graph of the present invention comprises the following steps:
step 1: the association with the experts in traditional Chinese medicine specifies the structure and relationship of the atlas of traditional Chinese medicine, relating the prescription, the decoction pieces contained in the prescription, the efficacy of the decoction pieces, the diseases to be treated, the famous physicians related to the decoction pieces, and what drugs the decoction pieces can form. A number of domains are defined to facilitate the management of classes in groups. For example, there are acupoints and actions in the acupoint region. The drug pair domain includes drug pairs, and the prescription domain includes prescriptions, decoction pieces, famous physicians, and the like. And forming a large unidirectional knowledge graph.
Step 1.1: class hierarchy: the atlas adopts a two-layer class hierarchy structure.
Step 1.2: class relationship definition: the classes have a mutual relationship, and a unidirectional relationship or a bidirectional relationship can be defined between the classes. If the primary inverse relationship needs to be defined, the primary relationship and the inverse relationship need to be agreed. The atlas is in one-way relationship. For example, the prescription is made into decoction pieces.
Step 1.3: class domain definition: for management convenience, a plurality of domains are defined to facilitate the grouping management of classes. For example, there are acupoints and actions in the acupoint region. The drug pair domain includes drug pairs, and the prescription domain includes prescriptions, decoction pieces, famous physicians, and the like.
Step 2: because the data set is a structured database, the structured data is converted to RDF data using the existing tool D2R tool. I.e. consisting of a triplet form "resource-attribute value" or say "subject-predicate-object".
And step 3: many data sources include writings such as compendium of materia Medica, treatise of typhoid, etc., which means that a plurality of knowledge and graph spectrums need to be fused, which needs to deal with two problems: fusing a newly obtained body into an existing subject library and fusing a new body and an old body through the fusion of the mode layer; the fusion of the data layers, including the alternative names, the affiliations and the affiliated categories of the entities, reduces unnecessary redundancy.
The fusion of the data layer refers to the fusion of entities and relation (including attribute) tuples, mainly refers to entity matching or alignment, and is superior to some entities in a knowledge base in terms of identical meaning but different identifiers, so that the entities need to be combined, and the entity alignment is performed based on attribute similarity matching in the invention.
Defining two knowledge to be matched G1And G2Entities A, B and
entity A has its set of attributes { A }1,A2,A3…Ai…, entity B has its set of attributes B1,B2,B3…Bi…}。
Step 3.1: and performing weight evaluation on the attributes. According to a common keyword evaluation technology in IDF (inverse Document frequency) information retrieval, the importance of the keywords in the whole corpus is evaluated by using a statistical method. The main idea is as follows: the importance of a keyword decreases inversely with the frequency with which it appears in the corpus. I.e. the more times a keyword appears in a document, the less information it conveys to the user, the less important it is. The frequently occurring keywords should be assigned a smaller weight. According to this IDF idea, the weight when the attribute value is equal to v is defined as:
IDFi(v)＝log(n/Fi(v)) (2)
where n represents the total number of all attributes in the knowledge-graph, Fi(v) Representing the total number of attributes in the knowledge base with attribute value v.
In practical applications, an entity may have multiple attributes, and therefore, the weighting of the multiple attributes of the entity needs to be normalizediWeight w ofiCan be calculated from the following formula:
step 3.2: and evaluating the similarity between the attribute values. Attributes of an entity generally take three forms: text type, numerical type, and list type. The list-type attribute is an attribute of an entity corresponding to a plurality of attribute values, for example, a decoction piece attribute of a prescription, and has a plurality of decoction piece compositions, and the list-type attribute is the list-type attribute. The entity similarity is calculated in three cases. For a numerical entity we calculate the similarity of attributes between two entities using the following formula:
wherein d isiAnd djIs the attribute value of the desired matching attribute of the two entities. The value range of the similarity is [0, 1] according to the formula]And when d isjAnd diThe larger the difference, D (D)i,dj) The smaller the value of (a), the greater the distance between the two attributes and the smaller the similarity.
For textual entity attributes, we first use the vector space model-based TF-IDF method. TF-IDF is a traditional statistical method that works to assess how important a word is to a document. The method counts unit position words, so that a Chinese word segmentation tool is used for segmenting words of a text book, and each text data TF-IDF vector is obtained through calculation. After vectors are obtained, cosine similarity is used for measuring similarity between the vectors. Then for the two n-dimensional TF-IDF vectors of a, B, the pre-similarity between them can be obtained by the following formula:
it can be seen from equation (5) that the range of the similarity is [0, 1], and the larger the value is, the higher the similarity is.
The data of the list type can be processed as a collection. For attribute values of the list phenotype, we used two indices to measure their similarity. One is to calculate the number of intersections, and the larger the number of intersections is, the more similar the intersections are; the other is Jaccard similarity, and the calculation formula is shown in formula (6):
for both sets a and B, the Jaccard similarity has a value range of [0, 1], with a larger value indicating a higher similarity between them.
Step 3.3: and (3) summing the weight calculated in the step (3.1) and the attribute similarity calculated in the step (3.2), setting a threshold value, and classifying according to the range of the similarity. Let us set e1，e2The similarity between the two entities is
Then:
setting 2 similarity thresholds, judging total similarity score, setting 2 similarity thresholds, and judging total similarity score
Which similarity interval is located can be expressed as:
when the two entities are matched, the two entities are merged, the attribute information of the entities is mutually supplemented, and the link between the entities is completed. When the two entities are possibly matched, the part of the entities is screened out and manually marked. And if the two entities are matched, putting the entities into a matching set, combining the entities with other entities, and completing the knowledge graph.
And 4, step 4: and after alignment, generating an owl file by the triple table, and reading the information of the entity in the triple table by using Python language, wherein the information is the information of the prescription and the decoction piece in the map. The formula (I) is as follows: storing the read data in a list form of decoction piece 1, decoction piece 2 and decoction piece 3, and corresponding the prescription and decoction piece information; reading the objects, property and value of the first three columns in the table, and saving the data in the dictionary and the dual dictionary by using a for loop.
And 5: a knowledge graph is built in the neo4j graph database. Python provides a method of linking neo4j data: from py2neo import Graph, Node, Relationship, connects to the neo4j database using the methods provided in the package.
Step 5.1: and (4) reading the data in the dictionary and the double dictionaries in the step (4), establishing the Node by using a Node () method, and filling labels such as label, name and the like in brackets to enable the Node to have uniqueness. Create a node with a graph () statement. And (5) importing the node data in the dictionary into a neo4j database to form nodes of prescriptions, decoction pieces and the like.
Step 5.2: relationships in the graph are created using the relationship (node 1, relationship name, node 2) method, which is the relationship that node 1 points to node 2. Create () statements are used to import all relationship data. Reading all relations in the triples, connecting the nodes together and establishing a knowledge graph arranged according to the hierarchy, solving the problem of knowledge island and making the relations between the knowledge clear at a glance. The effect is shown in figure 4.
Step 6: the Neo4j provides a function of changing the colors of the nodes, and clicks one type of nodes to select colors, so that the map is beautified, and the knowledge map is clearer. In order to more intuitively know the efficacy of the acupuncture points, a scalable svg vector diagram is created in HTML5 to generate an acupuncture point efficacy knowledge map, an svg xml file is created, an arrow and an origin point label are created, finally, a user mouse is suspended on the origin point of the acupuncture points, so that the arrow vector can be extended to point at the efficacy origin point of the acupuncture points, the shading color of the pointed origin point of the acupuncture points is deepened, and the memory of the user is facilitated. The E-charts are connected with a single concept in series to form a small knowledge graph according to the hierarchical relationship, so that the understanding of a user on the traditional Chinese medicine concept is deepened, and the effect is shown in the attached figure 5.
And 7: and building a web project. For the convenience of users, a traditional Chinese medicine knowledge map visualization system is set up. The system comprises functions of user registration and login, knowledge question answering, knowledge encyclopedia and the like. Firstly, a web page is designed by utilizing JavaScript and H5 technology, and then a basic traditional Chinese medicine page website system is set up by adopting an SMM framework of Java. The information of traditional Chinese medicine, prescription information, traditional Chinese medicine figures, medical books, cases and the like can be read from the database. Meanwhile, the user can check all medical contents through the system and inquire single medical information which the user wants to know.
The user enters the system and the first step requires the option of registration and login as shown in figure 6. The unregistered user clicks a registration button, inputs a user name and a password set by the user, and clicks a submit button to complete registration, as shown in fig. 7. And jumps back to the system start page. And the user selects a login button, clicks a jump login interface, and inputs a user name and a password which are registered by the user.
The first page is mainly used to introduce the composition of the system. The method mainly comprises four parts, namely a knowledge map, knowledge encyclopedia, knowledge question answering and contact us, which are displayed on a paging column. And inserts a js file of the tree graph of E-characters. The part can be moved, zoomed in and zoomed out or clicked to expand the content, which is convenient for the user to view, as shown in fig. 8.
The knowledge question-answering module comprises specific pages of medical documents, symptoms, treatment methods, traditional Chinese medicine diseases, prescriptions, traditional Chinese medicine figures, traditional Chinese medicine medical records and the like. Here take an example of a medical literature page. And clicking the Chinese book literature in the knowledge question-answer paging column to enter a medical book literature page, as shown in the attached figure 9. The pages of the medical book and literature show the detailed introduction of the four classic medical books of China, yellow emperor's internal classic, difficult classic, treatise on febrile diseases and miscellaneous diseases, Shennong's herbal classic. The right-hand button can be used for checking and inquiring medical book documents. Clicking the medical book viewing button, the system displays all relevant information of the serial numbers, names, explanations and the like of the medical book documents in the database to the user, as shown in figure 10. And clicking a medical book document query button, entering a query interface by the system, inputting the name of the medical book document to be queried, and displaying the specific query information by the system. And shows the proportion of each classification of the medical book and literature, as shown in figure 11.
The knowledge encyclopedia is divided into an encyclopedia page of human acupuncture points, an encyclopedia page of medical record mining and an encyclopedia page of pharmacological action. Clicking the human acupuncture point encyclopedia in the knowledge encyclopedia paging column to enter the human acupuncture point encyclopedia page, as shown in figure 12. The page shows all acupuncture points of the human body to a user in a grading mode through an E-charts relation graph. The first is the human body; the second level is all parts of the human body, including the head, the chest, the feet, the back, the legs, the arms, the hands and the six major parts; the third level is the name of the relevant acupoint. And the hiding of the secondary classification can be realized by clicking the left acupoint button. The right side of the page is the popular science of human body acupuncture points, which show eight functional and rich acupuncture points and functions of the human body to users, namely, Dazhui acupuncture point, Zusanli acupuncture point, Sanyinjiao acupuncture point, Zhongwan acupuncture point, Tianshu acupuncture point, Qihai acupuncture point, Shenque acupuncture point and Mingmen acupuncture point. Clicking a medical case in the knowledge encyclopedia paging column to mine encyclopedias, and entering a medical case mining encyclopedia page, as shown in figure 13. The page extracts the medicinal materials which are most applied in the prescription for treating the cold, namely almond, liquorice and the like respectively through a word cloud technology according to nearly hundred Chinese medical records related to the cold in a database and related to the cold and complications thereof. This result was confirmed by looking up the ancient book of medical books, and the main effects of almond, licorice and peppermint are shown on the right side of the page, all of which are related to the treatment of cold. Clicking the knowledge encyclopedia paging column to access the pharmacological encyclopedia page, as shown in figure 14. This page presents the user with the relevant concept of medicine, pharmacologic action. The right page shows the pharmacological action of common fruits
The knowledge map module of the system comprises pages such as a general map, a medicine pair map, an acupuncture point and function map, a project map and the like. Clicking the total map of the knowledge map paging column, and jumping the page to the total map page, as shown in the attached figure 4. The neo4j database is used for storing data in a triple mode, the data are imported into the neo4j database through a python program, and the database shows prescriptions, decoction pieces forming the prescriptions, decoction pieces treating diseases, effects, famous doctors related to the decoction pieces and other information. The user can check concepts such as prescriptions or decoction pieces more intuitively by dragging the mouse, and check related concepts by the expansion button, and the locking button is used for locking the map graph, so that the user can check conveniently. Clicking the knowledge graph paging column medicine pair graph, and jumping the page to a medicine pair graph page, as shown in the attached figure 15. And drawing a knowledge graph of sixty-multiple medicine pairs according to the E-characters relation graph, wherein a user can enlarge and reduce by rolling a mouse, and move the mouse to move pages. Clicking the knowledge map paging column to jump to the acupuncture point and efficacy map page, as shown in figure 5. Forty-eight acupuncture points and corresponding effects are drawn by using the svg vector diagram, the mouse is moved to the name of the acupuncture points, the extending vector lines point to the corresponding effects, and the color of the acupuncture points where the mouse stays is deepened, as shown in figure 16, so that the memory of a user is facilitated.
While the foregoing is directed to the preferred embodiment of the present invention, other and further embodiments of the invention may be devised without departing from the basic scope thereof, and the scope thereof is determined by the claims that follow.
Claims (3)
1. A traditional Chinese medicine knowledge graph construction and visualization method is characterized by comprising the following steps:
step 1: creating a mode of the knowledge graph;
step 2: converting traditional Chinese medicine information in the relational database into RDF data;
and step 3: the entity alignment is realized according to the attribute similarity matching of the two entities, and the two entities are combined into one entity, so that the repeated redundancy of the entities is avoided;
and 4, step 4: generating an own file by the aligned triple table, importing the own file into a visualization tool project to form a knowledge graph system, and visually displaying the relation among the prescriptions, the decoction pieces, the traditional Chinese medicines, the processing of the medicinal materials and the base sources of the medicinal materials for a user;
and 5: calculating weight information of the decoction pieces in the prescription by using an algorithm;
step 6: generating an acupoint effect knowledge map by using the svg vector diagram;
and 7: and building a web project.
2. The method for constructing and visualizing a knowledge graph of traditional Chinese medicine according to claim 1, wherein in step S1, the method comprises the steps of:
step 1.1: class hierarchy: the atlas adopts a two-layer class hierarchy structure;
step 1.2: class relationship definition: the classes have mutual relations, and the classes can define a one-way relation and a two-way relation;
step 1.3: class domain definition: defining multiple realms facilitates group management of classes.
3. The method for constructing and visualizing a knowledge graph of traditional Chinese medicine according to claim 1, wherein in step S3, the method comprises the steps of:
step 3.1: performing weight evaluation on the attributes;
step 3.2: evaluating the similarity between the attribute values;
step 3.3: and (3) summing the weight calculated in the step (3.1) and the attribute similarity calculated in the step (3.2), defining the sum as the similarity of the two entities, setting a threshold value, and classifying according to the range.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011232041.5A CN112635078A (en) | 2020-11-06 | 2020-11-06 | Traditional Chinese medicine knowledge graph construction and visualization method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011232041.5A CN112635078A (en) | 2020-11-06 | 2020-11-06 | Traditional Chinese medicine knowledge graph construction and visualization method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112635078A true CN112635078A (en) | 2021-04-09 |
Family
ID=75302933
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011232041.5A Pending CN112635078A (en) | 2020-11-06 | 2020-11-06 | Traditional Chinese medicine knowledge graph construction and visualization method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112635078A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113505242A (en) * | 2021-07-16 | 2021-10-15 | 珍岛信息技术（上海）股份有限公司 | Method and system for automatically embedding knowledge graph |
| CN114882985A (en) * | 2022-07-11 | 2022-08-09 | 北京泽桥医疗科技股份有限公司 | Medicine multimedia management system and method based on database and AI algorithm identification |
| CN117076689A (en) * | 2023-08-21 | 2023-11-17 | 浙江大学 | Intelligent design method for traditional Chinese medicine pharmaceutical process route |
| CN117408338A (en) * | 2023-12-14 | 2024-01-16 | 神州医疗科技股份有限公司 | Method and system for constructing knowledge graph of traditional Chinese medicine decoction pieces based on Chinese pharmacopoeia |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109815327A (en) * | 2018-12-07 | 2019-05-28 | 南京中新赛克科技有限责任公司 | A kind of big data knowledge mapping visibility solution based on SVG |
| CN110825721A (en) * | 2019-11-06 | 2020-02-21 | 武汉大学 | Hypertension knowledge base construction and system integration method under big data environment |
| CN111597788A (en) * | 2020-05-18 | 2020-08-28 | 腾讯科技（深圳）有限公司 | Attribute fusion method, device and equipment based on entity alignment and storage medium |
| CN111639190A (en) * | 2020-04-30 | 2020-09-08 | 南京理工大学 | Medical knowledge map construction method |
| CN111881290A (en) * | 2020-06-17 | 2020-11-03 | 国家电网有限公司 | Distribution network multi-source grid entity fusion method based on weighted semantic similarity |
-
2020
- 2020-11-06 CN CN202011232041.5A patent/CN112635078A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109815327A (en) * | 2018-12-07 | 2019-05-28 | 南京中新赛克科技有限责任公司 | A kind of big data knowledge mapping visibility solution based on SVG |
| CN110825721A (en) * | 2019-11-06 | 2020-02-21 | 武汉大学 | Hypertension knowledge base construction and system integration method under big data environment |
| CN111639190A (en) * | 2020-04-30 | 2020-09-08 | 南京理工大学 | Medical knowledge map construction method |
| CN111597788A (en) * | 2020-05-18 | 2020-08-28 | 腾讯科技（深圳）有限公司 | Attribute fusion method, device and equipment based on entity alignment and storage medium |
| CN111881290A (en) * | 2020-06-17 | 2020-11-03 | 国家电网有限公司 | Distribution network multi-source grid entity fusion method based on weighted semantic similarity |
Non-Patent Citations (4)
| Title |
|---|
| 张德政 等: "基于本体的中医知识图谱构建", 《情报工程》, vol. 3, no. 1, pages 35 - 42 * |
| 邹银凤: "知识图谱构建中的多数据源实体匹配研究", 《中国优秀硕士学位论文全文数据库经济与管理科学辑》, no. 01, pages 12 - 16 * |
| 郭文龙: "中医方剂知识图谱构建研究与实现", 《中国优秀硕士学位论文全文数据库医药卫生科技辑》, no. 08, pages 9 - 12 * |
| 阮彤 等: "中医药知识图谱构建与应用", 《医学信息学杂志》, vol. 37, no. 4, pages 8 - 13 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113505242A (en) * | 2021-07-16 | 2021-10-15 | 珍岛信息技术（上海）股份有限公司 | Method and system for automatically embedding knowledge graph |
| CN114882985A (en) * | 2022-07-11 | 2022-08-09 | 北京泽桥医疗科技股份有限公司 | Medicine multimedia management system and method based on database and AI algorithm identification |
| CN117076689A (en) * | 2023-08-21 | 2023-11-17 | 浙江大学 | Intelligent design method for traditional Chinese medicine pharmaceutical process route |
| CN117408338A (en) * | 2023-12-14 | 2024-01-16 | 神州医疗科技股份有限公司 | Method and system for constructing knowledge graph of traditional Chinese medicine decoction pieces based on Chinese pharmacopoeia |
| CN117408338B (en) * | 2023-12-14 | 2024-03-12 | 神州医疗科技股份有限公司 | Method and system for constructing knowledge graph of traditional Chinese medicine decoction pieces based on Chinese pharmacopoeia |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112635078A (en) | Traditional Chinese medicine knowledge graph construction and visualization method | |
| CN112487202B (en) | Chinese medical named entity recognition method and device fusing knowledge map and BERT | |
| WO2021213314A1 (en) | Data processing method and device, and computer readable storage medium | |
| CN112148851A (en) | Construction method of medicine knowledge question-answering system based on knowledge graph | |
| CN110189802B (en) | Bidirectional mapping queue research information system based on index storage model | |
| CN112635071B (en) | Diabetes knowledge graph construction method integrating Chinese and Western medicine knowledge | |
| Alam et al. | Latviz: A new practical tool for performing interactive exploration over concept lattices | |
| Wang | thesis | |
| Xiao et al. | Library and information science research in China—A survey based analysis of 10 lis educational institutes | |
| Gruber et al. | Semantic web technologies applied to numismatic collections | |
| Isemann et al. | Ontological access to images of fine art | |
| Miao et al. | Construction of semantic-based traditional Chinese medicine prescription knowledge graph | |
| Ding et al. | Research on question answering system for COVID-19 based on knowledge graph | |
| Burrows et al. | Medieval manuscripts and their migrations: Using SPARQL to investigate the research potential of an aggregated Knowledge Graph | |
| Cojan et al. | Filling the gaps among DBpedia multilingual chapters for question answering | |
| Leskinen et al. | Combining Faceted Search with Data-analytic Visualizations on Top of a SPARQL Endpoint. | |
| Hernandez et al. | Synchronized tag clouds for exploring semi-structured clinical trial data | |
| Badre et al. | Comparative ease of use of a diagrammatic vs. an iconic query language | |
| Cui et al. | AMFormulaS: an intelligent retrieval system for traditional Chinese medicine formulas | |
| CN114490930A (en) | Cultural relic question-answering system and question-answering method based on knowledge graph | |
| Su et al. | An interactive knowledge graph based platform for covid-19 clinical research | |
| Lin | The study of premodern chinese literature in the digital era: New methods of quantitative statistics, databases, and visualization analyses | |
| Haryadi et al. | Classification of Drug Effectiveness Based on Patient's Condition Using Text Mining With K-Nearest Neighbor | |
| Hou et al. | Research and practice on the construction of Chinese medicine knowledge intelligent question answering system based on knowledge graph | |
| Abedi et al. | Comparison on information-seeking behavior of postgraduated students in Isfahan University of Medical Sciences and University of Isfahan in writing dissertation based on Kuhlthau model of information search process |
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 |