CN114238584A - Information search method, device and storage medium - Google Patents
Information search method, device and storage medium Download PDFInfo
- Publication number
- CN114238584A CN114238584A CN202111579356.1A CN202111579356A CN114238584A CN 114238584 A CN114238584 A CN 114238584A CN 202111579356 A CN202111579356 A CN 202111579356A CN 114238584 A CN114238584 A CN 114238584A
- Authority
- CN
- China
- Prior art keywords
- medical
- entity
- search
- knowledge
- text query
- 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
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- 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
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
-
- 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
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
Abstract
The embodiment of the application relates to the technical field of knowledge maps, in particular to an information searching method, equipment and a storage medium, which are used for acquiring medical searching contents submitted by a user through a client; generating at least two text query sentences according to the medical search content; inquiring in the established medical knowledge map through at least two text inquiry sentences to obtain medical knowledge corresponding to the text inquiry sentences; reasoning medical knowledge corresponding to the text query sentence to obtain a search result; and outputs the search result to the client. By the scheme, the user can simply and quickly search medical knowledge through the client, and can obtain a structured medical decision through reasoning, so that the daily medical requirements of the user can be met. In addition, based on the text query sentence, the medical knowledge is retrieved and inferred, so that the proper medical knowledge can be quickly positioned and selected, and the accuracy and the acquisition efficiency of the medical knowledge are improved.
Description
Technical Field
The present application relates to the field of knowledge graph technology, and in particular, to an information search method, device, and storage medium.
Background
With the continuous progress of medical level, higher requirements are put on the convenience and reliability of medical treatment, high-quality medical resources are seriously insufficient, the medical level is also uneven, and how to improve or standardize the diagnosis and medical level is required, so that the popularization of the high-quality medical resources at low cost is a problem which needs to be solved urgently today.
In the related art, medical knowledge may be stored by a medical aid decision-making system, thereby providing a user with a medical knowledge query entry. However, the conventional medical aid decision-making system usually stores medical knowledge by using a big data method, and many medical knowledge are stored in unstructured data sources such as books, electronic documents, papers and the like, so that it is difficult to form a structured result for users to refer to, and the accuracy and efficiency of the obtained medical knowledge are low.
Disclosure of Invention
The application provides an information searching method, equipment and a storage medium, which are used for solving the technical problems of low accuracy and low acquisition efficiency of medical knowledge obtained by a current medical aid decision-making system.
In a first aspect, the present application provides an information search method, including: acquiring medical search content submitted by a user through a client; generating at least two text query sentences according to the medical search content; inquiring in the established medical knowledge map through at least two text inquiry sentences to obtain medical knowledge corresponding to the text inquiry sentences; reasoning medical knowledge corresponding to the text query sentence to obtain a search result corresponding to the medical search content, wherein the search result comprises a reasoning result and is used for assisting a user in obtaining a medical decision related to the medical search content; and outputting the search result to the client so that the client outputs the search result to the user in an application mode.
In some embodiments, generating at least two text query statements from the medical search content includes: identifying a target entity contained in the medical search content; and generating at least two text query sentences related to the target entities, wherein the text query sentences contain at least two target entities.
In some embodiments, identifying a target entity contained in the medical search content comprises: marking medical entities and medical entity relations contained in the medical search content through a natural language model, wherein the natural language model is used for marking the medical entities and medical entity relations of the input content; and extracting a target entity contained in the medical search content according to the labeling result, wherein the target entity is structured data.
In some embodiments, identifying the target entity contained in the medical search content further comprises: and expanding the extracted target entity based on the entity application rule to obtain an expanded target entity, wherein the expanded entity comprises the extracted target entity and the expanded target entity.
In some embodiments, the information search method further includes: a method combining manual input and deep learning input is adopted to create a medical knowledge map; the manual entry is that related personnel enter the medical entity and the medical entity relationship through a graphical interface; and deep learning input is to label the unknown data set by using a natural language model, generalize to obtain a labeling result, and analyze the medical entity and the medical entity relationship in the unknown data set according to the labeling result.
In some embodiments, the information search method further includes: and representing the medical knowledge in the medical knowledge map by adopting a triple, wherein the triple is represented as follows: medical entity, medical entity relationship, medical entity.
In a second aspect, the present application provides an information search apparatus, comprising:
the acquisition module is used for acquiring medical search contents submitted by a user through a client;
the generating module is used for generating at least two text query sentences according to the medical search content;
the query module is used for querying in the established medical knowledge map through at least two text query sentences to obtain medical knowledge corresponding to the text query sentences;
the reasoning module is used for reasoning the medical knowledge corresponding to the text query sentence to obtain a search result corresponding to the medical search content, the search result comprises a reasoning result, and the search result is used for assisting a user in obtaining a medical decision related to the medical search content;
and the output module is used for outputting the search result to the client so that the client outputs the search result to the user in an application mode.
In some embodiments, the generation module is specifically configured to: identifying a target entity contained in the medical search content; and generating at least two text query sentences related to the target entities, wherein the text query sentences contain at least two target entities.
In some embodiments, the generation module is specifically configured to: marking medical entities and medical entity relations contained in the medical search content through a natural language model, wherein the natural language model is used for marking the medical entities and medical entity relations of the input content; and extracting a target entity contained in the medical search content according to the labeling result, wherein the target entity is structured data.
In some embodiments, the generation module is further to: and expanding the extracted target entity based on the entity application rule to obtain an expanded target entity, wherein the expanded entity comprises the extracted target entity and the expanded target entity.
In some embodiments, the information search apparatus further includes: the processing module is used for creating a medical knowledge map by adopting a method combining manual input and deep learning input; the manual entry is that related personnel enter the medical entity and the medical entity relationship through a graphical interface; and deep learning input is to label the unknown data set by using a natural language model, generalize to obtain a labeling result, and analyze the medical entity and the medical entity relationship in the unknown data set according to the labeling result.
In some embodiments, the processing module is further to: and representing the medical knowledge in the medical knowledge map by adopting a triple, wherein the triple is represented as follows: medical entity, medical entity relationship, medical entity.
In a third aspect, an embodiment of the present application further provides an electronic device, including: a processor, and a memory communicatively coupled to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored by the memory to implement the information search method of the first aspect.
In a fourth aspect, the present application provides a computer-readable storage medium having stored therein computer-executable instructions for implementing the information search method according to the first aspect when the computer-executable instructions are executed by a processor.
In a fifth aspect, the present application provides a computer program product comprising a computer program that, when executed by a processor, implements the information search method of the first aspect.
According to the information searching method, the information searching equipment and the information searching storage medium, medical searching contents submitted by a user through a client are obtained; generating at least two text query sentences according to the medical search content; inquiring in the established medical knowledge map through at least two text inquiry sentences to obtain medical knowledge corresponding to the text inquiry sentences; reasoning medical knowledge corresponding to the text query sentence to obtain a search result corresponding to the medical search content, wherein the search result comprises a reasoning result and is used for assisting a user in obtaining a medical decision related to the medical search content; and outputting the search result to the client so that the client outputs the search result to the user in an application mode. By the scheme, the user can simply and quickly search medical knowledge through the client, and can obtain a structured medical decision through reasoning, so that the daily medical requirements of the user can be met. In addition, based on the text query sentence, the medical knowledge is retrieved and inferred, so that the proper medical knowledge can be quickly positioned and selected, and the accuracy and the acquisition efficiency of the medical knowledge are improved.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and together with the description, serve to explain the principles of the application.
Fig. 1 is an exemplary diagram of an application scenario provided in an embodiment of the present application;
fig. 2 is a first schematic flow chart of an information search method according to an embodiment of the present application;
fig. 3 is a schematic flowchart illustrating a second information search method according to an embodiment of the present application;
fig. 4 is a schematic structural diagram of an information search apparatus according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
With the above figures, there are shown specific embodiments of the present application, which will be described in more detail below. These drawings and written description are not intended to limit the scope of the inventive concepts in any manner, but rather to illustrate the inventive concepts to those skilled in the art by reference to specific embodiments.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.
In the technical scheme of the present application, the collection, storage, use, processing, transmission, provision, disclosure and other processing of the information such as the user data and the medical data are all in accordance with the regulations of the relevant laws and regulations, and do not violate the good custom of the public order.
The terms referred to in this application are explained first:
text query statement: (text query language, tql).
Entity: also known as objects or instances, refer to something that is distinguishable and independent, e.g., drugs, diseases, etc., entities are the most basic elements of a knowledge graph, and each entity may correspond to a unique identity.
The relationship is as follows: refers to "edges" connecting different entities to describe the relationship between the entities, e.g., there is an indication relationship between cough and cefaclor, there is a contraindication relationship between cefaclor and cefaclor, etc.
The attributes are as follows: refers to the possible characteristics and parameters of an entity, such as cefaclor as an antibiotic drug property.
With the continuous progress of medical level, higher requirements are put on the convenience and reliability of medical treatment, high-quality medical resources are seriously insufficient, the medical level is uneven, and how to improve or standardize the diagnosis and medical level is to make the high-quality medical resources popularized with low cost an important link for solving the problem.
And China starts to step into the age-related era step by step, the proportion of the elderly population is increased year by year, the elderly stage is the high-incidence stage of diseases, a large amount of medical resources are needed to meet the medical health requirements of the part of people, the high-quality medical resources are seriously insufficient, and the medical level is uneven, so that the method is a basic problem in the old-age care scene, and how to improve or standardize the diagnosis and medical level of doctors, so that the popularization of the high-quality medical resources with low cost is an important link for solving the problem.
In the related art, medical knowledge may be stored by a medical aid decision-making system, thereby providing a user with a medical knowledge query entry. However, the conventional medical aid decision-making system usually stores medical knowledge by using a big data method, and much medical knowledge is stored in unstructured data sources such as books, electronic documents, papers and the like, so that it is difficult to form a structured result for users to refer to, and the efficiency is low and the accuracy is difficult to consider.
In view of this, embodiments of the present application provide an information search method, an information search device, and a storage medium, which enable a user to simply and quickly obtain medical knowledge from a medical knowledge map through a client by constructing the medical knowledge map with medical experience and professional knowledge and providing a corresponding medical search platform, and perform relevant reasoning on the medical knowledge searched by the user to provide an accurate medical decision, thereby meeting daily medical needs of the user.
In addition, when medical knowledge is queried, the query is carried out based on the text query sentence, and the appropriate medical knowledge can be quickly positioned and selected, so that the accuracy and the acquisition efficiency of the medical knowledge can be further improved.
It should be noted that medical experience and professional knowledge in the medical knowledge map can be used as knowledge extension for inquiry and decision making, so that the medical knowledge map is applied to various medical scenes and meets medical requirements of different scenes.
For example, in a first scenario, China starts to step into the age-related era, a large amount of medical resources are needed to meet medical health requirements of the part of people, and the aid decision tools can be used for helping the old to perform daily basic diagnosis, implement health management and the like.
And the second scene can be used as a medical knowledge base for assisting the daily inquiry of doctors and providing medical knowledge reference for the inquiry of the doctors, so that the inquiry quality of the doctors is improved.
Fig. 1 is an exemplary diagram of an application scenario provided in an embodiment of the present application. As shown in fig. 1, the devices involved in this scenario include: a client 101.
The client 101 may be a Personal Digital Assistant (PDA) device, a handheld device (such as a smart phone and a tablet), a computing device (such as a Personal Computer (PC)), an in-vehicle device, a wearable device (such as a smart watch and a smart band), a smart home device (such as a smart display device), and the like, and fig. 1 illustrates a desktop computer, but is not limited thereto.
In practical applications, the client 101 has a relevant application installed thereon for querying medical knowledge, and it should be understood that the application is in various forms such as an application program, a plug-in, and the like, and a user can perform a search for medical knowledge through the application on the client 101.
Specifically, the user can input the medical search content into the search interface of the application, and the client 101 can query the created medical knowledge graph by using the information search method provided in the embodiment of the application to obtain the medical knowledge corresponding to the medical search content, perform inference according to the medical knowledge to obtain the search result corresponding to the medical search content, and display the search result on the interface of the client 101.
In some embodiments, the application scenario further includes a server 102, wherein the client 101 and the server 102 are connected through a network. The server 101 may be a single server, a server cluster, a distributed server, a centralized server, a cloud server, and the like, which is not specifically limited in the embodiment of the present application.
In this embodiment, the server 102 may further perform an information search operation, specifically, a user inputs medical search content to a related application of the client 101, the client 101 sends the medical search content to the server 102, the server 102 queries in the created medical knowledge map by using the medical search content to obtain medical knowledge corresponding to the medical search content, performs inference according to the medical knowledge to obtain a search result corresponding to the medical search content, and sends the search result to the client 101, so that the search result is presented to the user through an interface of the client 101.
The following describes the technical solutions of the present application and how to solve the above technical problems with specific embodiments. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments. Embodiments of the present application will be described below with reference to the accompanying drawings.
Fig. 2 is a first schematic flow chart of an information search method according to an embodiment of the present application. It should be understood that the execution subject of the embodiment of the present application may be the terminal device or the server, as shown in fig. 2, the information search method of the embodiment of the present application includes the following steps:
s201, medical search content submitted by a user through a client is obtained.
S202, generating at least two text query sentences according to the medical search content.
It should be understood that the medical knowledge map provided by the embodiment of the present application is used for describing basic medical knowledge of symptoms, diseases, signs, auxiliary examinations, drug contraindications and the like, and a user can search any content according to needs.
In some embodiments, the medical search content may include a plurality of entities, such search content often requires a query in multiple steps, in this step, semantic analysis may be performed on the medical search content, a search intention corresponding to the medical search content is obtained according to the plurality of entities included therein, so as to determine inference logic, and a text query statement corresponding to each step is generated according to the inference logic.
Taking the medical search content as "whether the medicine a and the medicine b have the common use contraindication" as an example, the entities contained in the medical search content include the medicine a and the medicine b, and the intention is "query common use contraindication", so that the text query sentence corresponding to the medical search content includes the following two types:
text query statement 1, and query of a medicine set A with a contraindication of use with medicine a;
and (5) text query statement 2, and determining whether the medicine set A contains the medicine b.
It should be understood that, in this step, it may also be performed to query whether the drug B has the drug set B with the contraindication for use, and then query whether the drug set B contains the drug a, which is not specifically limited in this embodiment of the present application.
S203, inquiring in the established medical knowledge map through at least two text inquiry sentences to obtain medical knowledge corresponding to the text inquiry sentences.
In this step, after the corresponding query statement is generated, according to the query logic, the medical knowledge corresponding to each text query statement is obtained by querying in the created medical knowledge map through the corresponding query statement.
Still taking the above example as an example, the query content of the text query statement 1 is content related to the medical knowledge-graph, while the query content of the text query statement 2 is not content in the medical knowledge-graph. Therefore, the medical knowledge graph is queried only by using the text query statement 1, so as to obtain a drug set a, where the drug set a is medical knowledge corresponding to the text query statement 1, and the drug set a includes at least one drug that has a contraindication for use with the drug a.
For the text query statement 2, the text query statement 2 does not need to be queried in the medical knowledge map.
And S204, reasoning according to the medical knowledge corresponding to the text query statement to obtain a search result corresponding to the medical search content.
Wherein the search results are used to assist the user in obtaining medical decisions related to the medical search content.
In the embodiment of the application, after the medical knowledge corresponding to the text query statement 1 is obtained, the medical knowledge corresponding to the text query statement 1 is inferred based on the text query statement 2, specifically, the medical knowledge corresponding to the text query statement 1 is queried in the drug set a according to the text query statement 2, and whether the drug set a contains the drug b is determined, so that an inference result is obtained.
Specifically, if the inference result is that the drug set a does not include the drug b, the search result is "there is no taboo in combination between the drug a and the drug b", and if the inference result is that the drug set a includes the drug b, the search result is "there is a taboo in combination between the drug a and the drug b".
And S205, outputting the search result to the client so that the client outputs the search result to the user in an application mode.
Specifically, after the search result is obtained, the search result is displayed in an application program interface of the client, and as for the display mode, the embodiment of the present application is not particularly limited, and for example, the search result may be displayed in a graphic and text mode.
In the embodiment of the application, medical knowledge is constructed by medical experience and professional knowledge and a corresponding medical search platform is provided, so that a user can simply and quickly acquire medical knowledge from the medical knowledge map through a client, and when the medical knowledge is inquired, the inquiry is carried out based on the text inquiry sentences, the proper medical knowledge can be quickly positioned and selected, and the accuracy and the acquisition efficiency of the medical knowledge can be further improved.
In addition, in the embodiment of the application, accurate medical decision can be provided for the user by carrying out relevant reasoning on the medical knowledge searched by the user, and compared with the query strategy provided in the prior art, the scheme can obtain structured data and can better meet the daily medical requirements of the user.
In some embodiments, the medical search content may also be query content that does not contain inference results, such query content typically containing only one entity.
For example, in one aspect, a search may be performed based on a disease or symptom, for example, a medicine that should be used for a symptom may be searched, a disease name corresponding to a symptom may be searched, a symptom corresponding to a disease name may be searched, and related medicine contraindications may be searched based on a symptom or disease, or a search may be performed based on a medicine, for example, a disease/symptom that is treated by a medicine may be searched, attributes of a medicine may be searched (e.g., a medicine classification, a medicine attribute package, etc.), and a contraindication of a medicine may be used.
In this embodiment, the medical search content may correspond to only one text query statement, and for example, the medical search content is "contraindication for drug 1", and the corresponding text query statement includes: and inquiring a medicine set which has a contraindication of the combination with the medicine 1.
Correspondingly, the medical knowledge corresponding to the text query sentence is the search result, and through the embodiment of the application, the query requirement of the user can be met, and the user experience is improved.
Fig. 3 is a schematic flowchart of a second information search method according to an embodiment of the present application. On the basis of fig. 2, the information search method is described in more detail by the embodiment of the present application, and as shown in fig. 3, the information search method of the embodiment of the present application includes the following steps:
s301, medical search content submitted by a user through a client is obtained.
S302, identifying a target entity contained in the medical search content.
Specifically, step S302 includes steps S3021 to S3022 as follows:
s3021, marking medical entities and medical entity relations contained in the medical search content through the natural language model.
The natural language model is obtained by training based on the medical entity and the medical entity relationship, and is used for identifying the medical entity and the medical entity relationship.
In this step, the medical search content is input to the natural language model, the medical entity and medical entity relationship in the medical search content are identified through the natural language model, and the medical entity and medical entity relationship is marked.
For example, still take the medical search content as "whether there is a contraindication for use between drug a and drug b" as an example, where the medical entities included are "drug a" and "drug b", the medical entity relationship is "contraindication for use", the "drug a" and "drug b" are labeled as medical entities, and the "contraindication for use" is labeled as medical entity relationship.
And S3022, extracting a target entity contained in the medical search content according to the labeling result.
Wherein, the target entity is structured data, in the above example, the target entity includes the following: "drug a-medical entity", "drug b-medical entity", "contraindication for use-medical entity relation".
Optionally, after the target entity is obtained, the extracted target entity may be expanded based on the entity application rule to obtain an expanded target entity, where the expanded entity includes the extracted target entity and the expanded target entity.
It should be understood that the embodiments of the present application are not particularly limited to the entity application rule. In one embodiment, the target entities corresponding to the medical entities may be expanded according to the medical knowledge graph, and for example, the alternative names corresponding to the target entities may be queried according to "entity-alternative name-entity" in the medical knowledge graph, so as to determine that the alternative names are the target entities corresponding to the medical entities.
Illustratively, taking the drug a as the penicillin, the expanded target entities include all terms corresponding to penicillin, such as "penicillin G-medical entity, peillin G-medical entity, penicillin sodium-medical entity, and the like".
In another embodiment, the target entity corresponding to the medical entity may be expanded according to the synonym, and for example, for the target entity "combination contraindication-medical entity relationship", the combination contraindication may be expanded to "medication contraindication-medical entity relationship, drug contraindication-medical entity relationship, and the like".
In the embodiment of the application, the relation between the medical entity and the medical entity contained in the medical search content can be accurately extracted through the natural language model, so that the accuracy of medical decision is improved. In addition, after the target entity is obtained, the target entity is expanded, so that the medical knowledge in the medical knowledge map can be traversed as much as possible during query, the accuracy and the integrity of the medical knowledge are guaranteed, the accuracy of medical decision is finally improved, and the situation that the corresponding medical knowledge cannot be found due to the fact that a certain entity name does not exist in the medical knowledge map, and further a search result cannot be given is avoided.
S303, generating at least two text query sentences related to the target entity.
In a possible implementation manner, a text query statement corresponding to the relationship between a medical entity and the medical entity may be constructed based on at least one medical entity in the target entities, and the constructed text query statement still includes, for example, target entities including "drug a-medical entity", "drug b-medical entity", "contraindications for use-medical entity relationship": text query statement 1, and query of a medicine set A with a contraindication of use with medicine a;
further, other text query sentences are constructed based on the text query sentence 1 and the rest of the medical entities in the target entities.
Illustratively, the text query statement 2 is "whether drug b is contained in drug set A".
It should be understood that, when the target entity is an expanded target entity, all expanded contents in the target entity corresponding to the medical entity and the target entity corresponding to the medical entity relationship may be arranged and combined to obtain at least two text query sentences corresponding to each combination.
For example, the text query statement 1 may be "query and drug anMedicine set A with contraindication of combinationn", the text query statement 2 is" drug set AnWhether or not it contains a drug bn"wherein, the drug anIs any one of the alternative names of the drugs a, the drug bnIs an arbitrary name of the drug b.
S304, inquiring in the established medical knowledge map through at least two text inquiry sentences to obtain medical knowledge corresponding to the text inquiry sentences.
S305, reasoning the medical knowledge corresponding to the text query sentence to obtain a search result corresponding to the medical search content, wherein the search result comprises a reasoning result.
Wherein the search results are used to assist the user in obtaining medical decisions related to the medical search content.
Correspondingly, if the target entity is an expanded target entity, a plurality of inference results corresponding to the combination may be queried in the medical knowledge graph.
S306, outputting the search result to the client so that the client outputs the search result to the user in an application mode.
It should be noted that steps S304 to S306 are similar to steps S203 to S205 in the embodiment shown in fig. 2, and are not described again here.
In some embodiments, a medical knowledge map may be created using a combination of manual entry and deep learning entry.
The manual entry is that related personnel enter the medical entity and the medical entity relationship through a graphical interface;
the deep learning input is to label the unknown data set by using a natural language model, generalize to obtain a labeling result, and analyze the medical entity and the medical entity relationship in the unknown data set according to the labeling result.
Specifically, medical entities and medical entity relations are labeled in a graphical interface through related personnel, an initial natural language model is trained through a labeling result to obtain a natural language model, an unknown data set is labeled through the natural language model, and a labeling result is obtained in a generalization mode.
Further, the medical entity and the medical entity relationship in the unknown data set are analyzed according to the labeling result based on the natural language model.
Optionally, the triplets are used to represent medical knowledge in the medical knowledge-graph.
Wherein the triplets are represented as: medical entity, medical entity relationship, medical entity. For example, taking the medical data as "a subclass of drug a" as an example, first identifying the entities in the data through a natural language model includes: "drug a, subclass and subclass A".
Furthermore, labeling each entity through the natural language model, and obtaining a labeling result as follows: "drug a-medical entity", "subclass-medical entity relationship", "subclass a-medical entity".
Further, according to the labeling result, the medical knowledge is represented by a triple, wherein the medical knowledge is 'a-subclass A' of medicine.
The method of acquiring medical knowledge about other medical entities and medical entity relationships is similar to the above example and is not shown here.
In the related art, because the expression of medical knowledge faces many problems, doctors often have personalized characteristics for the summarization and description of experiences, information is lost during the summarization and description, and the medical experience needs to be converted into system language, which also faces great difficulty and has high requirements for input personnel. In the embodiment of the application, the medical entity and the medical entity relation are input by providing the graphical interface, the requirement on input personnel is low, the construction efficiency of the medical knowledge graph can be greatly improved, the error is not easy to occur, and the reliability of the medical knowledge graph can be synchronously improved.
In addition, in the embodiment of the application, a method combining manual input and deep learning input is adopted to create the medical knowledge map, only a small part of medical knowledge needs to be manually input to train the natural language model, the medical knowledge in the unknown data set can be obtained through the natural language model, manual input of all data sets is not needed, the input efficiency can be greatly improved, the structure extracted through the natural language model is more accurate, and the reliability of the medical knowledge map can be synchronously improved.
Corresponding to the information searching method shown in the above embodiment, fig. 4 is a schematic structural diagram of an information searching apparatus provided in the embodiment of the present application. For convenience of explanation, only a part related to an embodiment of the present application is shown, and as shown in fig. 4, the information search apparatus 400 includes:
an obtaining module 401, configured to obtain medical search content submitted by a user through a client;
a generating module 402, configured to generate at least two text query sentences according to the medical search content;
the query module 403 is configured to query the created medical knowledge map through at least two text query statements to obtain medical knowledge corresponding to the text query statements;
the reasoning module 404 is configured to reason the medical knowledge corresponding to the text query statement to obtain a search result corresponding to the medical search content, where the search result includes a reasoning result and is used to assist the user in obtaining a medical decision related to the medical search content;
and an output module 405, configured to output the search result to the client, so that the client outputs the search result to the user in an application manner.
In some embodiments, the generating module 402 is specifically configured to: identifying a target entity contained in the medical search content; and generating at least two text query sentences related to the target entities, wherein the text query sentences contain at least two target entities.
In some embodiments, the generating module 402 is specifically configured to: marking medical entities and medical entity relations contained in the medical search content through a natural language model, wherein the natural language model is used for marking the medical entities and medical entity relations of the input content; and extracting a target entity contained in the medical search content according to the labeling result, wherein the target entity is structured data. In some embodiments, the generation module 403 is further configured to: and expanding the extracted target entity based on the entity application rule to obtain an expanded target entity, wherein the expanded entity comprises the extracted target entity and the expanded target entity.
In some embodiments, the information search apparatus further includes: the processing module 406 is configured to create a medical knowledge map by using a method combining manual entry and deep learning entry; the manual entry is that related personnel enter the medical entity and the medical entity relationship through a graphical interface; and deep learning input is to label the unknown data set by using a natural language model, generalize to obtain a labeling result, and analyze the medical entity and the medical entity relationship in the unknown data set according to the labeling result.
In some embodiments, the processing module 406 is further configured to: and representing the medical knowledge in the medical knowledge map by adopting a triple, wherein the triple is represented as follows: medical entity, medical entity relationship, medical entity.
The information search device provided by the embodiment of the application can be used for executing the technical scheme of the information search method, the implementation principle and the technical effect are similar, and details are not repeated here.
Fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application. The electronic device may be a mobile terminal, a computer, a messaging device, a tablet device, a personal digital assistant, etc., among others.
As shown in fig. 5, electronic device 500 may include one or more of the following components: the processor 501 and the memory 502, in some embodiments, the electronic device 500 further comprises: power components 503, multimedia components 504, audio components 505, input/output (I/O) interfaces 506, sensor components 507, and communication components 508, among others.
The processor 501 generally controls, among other things, the overall operation of the electronic device 500, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. Processor 501 may include one or more processors to execute instructions to perform all or a portion of the steps of the method described above. Further, processor 501 may include one or more modules that facilitate interaction between processor 501 and other components. For example, the processor 501 may include a multimedia module to facilitate interaction between the multimedia components 504 and the processor 501.
The memory 502 is configured to store types of data to support operations at the electronic device 500. Examples of such data include instructions for any application or method operating on the electronic device 500, contact data, phonebook data, messages, pictures, videos, and so forth. The memory 502 may be implemented by any type or combination of volatile or non-volatile memory devices such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
The power supply component 503 provides power to the components of the electronic device 500. The power components 503 may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for the electronic device 500.
The multimedia component 504 includes a screen providing an output interface between the electronic device 500 and the user; the audio component 505 is configured to output and/or input audio signals; I/O interface 506 provides an interface between processor 501 and peripheral interface modules, which may be keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to: a home button, a volume button, a start button and a lock button; the sensor component 507 includes one or more sensors for providing various aspects of status assessment for the electronic device 500.
The communication component 506 is configured to facilitate wired or wireless communication between the electronic device 500 and other devices. The electronic device 500 may access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 506 receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 506 further includes a Near Field Communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.
In an exemplary embodiment, the electronic device 500 may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, micro-controllers, microprocessors or other electronic components for performing the above-described methods.
Embodiments of the present application also provide a non-transitory computer-readable storage medium, where instructions in the storage medium are executed by a processor of a terminal device, so that the terminal device can execute the above-mentioned heterogeneous data processing method.
It should be noted that the computer readable medium mentioned above in the present application may be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present application, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.
The computer-readable storage medium may be included in the electronic device; or may exist separately without being assembled into the electronic device.
The computer-readable storage medium carries one or more programs which, when executed by the electronic device, cause the electronic device to perform the video processing method shown in the above-described embodiments.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units described in the embodiments of the present application may be implemented by software or hardware. Where the name of a unit does not in some cases constitute a limitation of the unit itself, for example, the first retrieving unit may also be described as a "unit for retrieving at least two internet protocol addresses".
The above description is only a preferred embodiment of the application and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the disclosure herein is not limited to the particular combination of features described above, but also encompasses other arrangements formed by any combination of the above features or their equivalents without departing from the spirit of the disclosure. For example, the above features may be replaced with (but not limited to) features having similar functions disclosed in the present application.
Although the subject matter has been described in language specific to structural features and/or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.
In particular, according to an embodiment of the present application, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present application include a computer program product comprising a computer program embodied on a computer-readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the above-described functions defined in the heterogeneous data processing method of the embodiment of the present application are performed when the computer program is executed.
Other embodiments of the present application will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the application and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the application being indicated by the following claims.
It will be understood that the present application is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the application is limited only by the appended claims.
Claims (10)
1. An information search method, comprising:
acquiring medical search content submitted by a user through a client;
generating at least two text query sentences according to the medical search content;
inquiring in the established medical knowledge map through the at least two text inquiry sentences to obtain medical knowledge corresponding to the text inquiry sentences;
reasoning according to the medical knowledge corresponding to the text query statement to obtain a search result corresponding to the medical search content, wherein the search result comprises a reasoning result and is used for assisting the user in obtaining a medical decision related to the medical search content;
and outputting the search result to the client so that the client outputs the search result to the user in an application mode.
2. The information search method of claim 1, wherein generating at least two text query sentences according to the medical search content comprises:
identifying a target entity contained in the medical search content;
generating at least two text query statements related to the target entity.
3. The information search method of claim 2, wherein the identifying of the target entity contained in the medical search content comprises:
marking medical entities and medical entity relations contained in the medical search content through a natural language model, wherein the natural language model is obtained by training based on the medical entities and the medical entity relations;
and extracting a target entity contained in the medical search content according to the labeling result, wherein the target entity is structured data.
4. The information search method of claim 3, wherein the identifying of the target entity contained in the medical search content further comprises:
and expanding the extracted target entity based on the entity application rule to obtain an expanded target entity, wherein the expanded entity comprises the extracted target entity and the expanded target entity.
5. The information search method according to any one of claims 1 to 4, characterized by further comprising:
creating the medical knowledge map by adopting a method combining manual input and deep learning input;
the manual entry is the entry of medical entities and medical entity relations of related personnel through a graphical interface; and the deep learning input is to label an unknown data set by using a natural language model, generalize to obtain a labeling result, and analyze a medical entity and a medical entity relationship in the unknown data set according to the labeling result.
6. The information search method according to claim 5, further comprising:
representing medical knowledge in the medical knowledge-graph with triplets, the triplets appearing as: medical entity, medical entity relationship, medical entity.
7. An information search apparatus characterized by comprising:
the acquisition module is used for acquiring medical search contents submitted by a user through a client;
the generating module is used for generating at least two text query sentences according to the medical search content;
the query module is used for querying in the established medical knowledge map through the at least two text query sentences to obtain medical knowledge corresponding to the text query sentences;
the reasoning module is used for reasoning the medical knowledge corresponding to the text query statement to obtain a search result corresponding to the medical search content, the search result comprises a reasoning result, and the search result is used for assisting the user in obtaining a medical decision related to the medical search content;
and the output module is used for outputting the search result to the client so that the client outputs the search result to the user in an application mode.
8. An electronic device, comprising: a processor, and a memory communicatively coupled to the processor;
the memory stores computer-executable instructions;
the processor executes computer-executable instructions stored by the memory to implement the information search method of any one of claims 1 to 6.
9. A computer-readable storage medium having stored therein computer-executable instructions for implementing the information search method of any one of claims 1 to 6 when executed by a processor.
10. A computer program product comprising a computer program which, when executed by a processor, implements the information search method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111579356.1A CN114238584A (en) | 2021-12-22 | 2021-12-22 | Information search method, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111579356.1A CN114238584A (en) | 2021-12-22 | 2021-12-22 | Information search method, device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114238584A true CN114238584A (en) | 2022-03-25 |
Family
ID=80761230
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111579356.1A Pending CN114238584A (en) | 2021-12-22 | 2021-12-22 | Information search method, device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114238584A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116303980A (en) * | 2023-05-19 | 2023-06-23 | 无码科技（杭州）有限公司 | Large language model knowledge enhancement method, system, electronic equipment and medium |
-
2021
- 2021-12-22 CN CN202111579356.1A patent/CN114238584A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116303980A (en) * | 2023-05-19 | 2023-06-23 | 无码科技（杭州）有限公司 | Large language model knowledge enhancement method, system, electronic equipment and medium |
| CN116303980B (en) * | 2023-05-19 | 2023-08-15 | 无码科技（杭州）有限公司 | Large language model knowledge enhancement method, system, electronic equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10380505B2 (en) | Methods and systems for dynamically generating real-time recommendations | |
| Zook et al. | Ten simple rules for responsible big data research | |
| US10698956B2 (en) | Active knowledge guidance based on deep document analysis | |
| US10402408B2 (en) | Versioning of inferred data in an enriched isolated collection of resources and relationships | |
| US11640583B2 (en) | Generation of user profile from source code | |
| US9015118B2 (en) | Determining and presenting provenance and lineage for content in a content management system | |
| KR20190128116A (en) | Methods and systems for identifying, selecting, and presenting media-content items related to a common story | |
| CN111984784B (en) | Person post matching method, device, electronic equipment and storage medium | |
| US20180039657A1 (en) | Managing data across multiple systems | |
| CN111667374A (en) | Method and device for constructing user portrait, storage medium and electronic equipment | |
| Weerasinghe et al. | Big data analytics for clinical decision-making: Understanding health sector perceptions of policy and practice | |
| US20150006537A1 (en) | Aggregating Question Threads | |
| US20180060537A1 (en) | Systems and methods employing merge technology for the clinical domain | |
| CN112037905A (en) | Medical question answering method, equipment and storage medium | |
| Felix et al. | Mobile sensing for behavioral research: A component-based approach for rapid deployment of sensing campaigns | |
| CN114238584A (en) | Information search method, device and storage medium | |
| Kamoun-Chouk et al. | Towards integrated model of big data (BD), business intelligence (BI) and knowledge management (KM) | |
| CN112582073B (en) | Medical information acquisition method, device, electronic equipment and medium | |
| Jandrić et al. | Postdigital Citizen Science: Mapping the Field | |
| CN116450806A (en) | Systems and methods for context memory capture and recall | |
| Baccari et al. | Design for a context-aware and collaborative mobile learning system | |
| US20170193179A1 (en) | Graphical user interface (gui) for accessing linked communication networks and devices | |
| CN109524074B (en) | Case discussion method and device, computer-readable storage medium and electronic equipment | |
| CN116467433A (en) | Knowledge graph visualization method, device, equipment and medium for multi-source data | |
| Forbes et al. | Ontology engineering applications in healthcare and workforce management systems |
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 |