Multi-scene fusion method of medical knowledge organization system and related device

Abstract

The invention discloses a multi-scene fusion method of a medical knowledge organization system, and relates to the technical field of software, a word list fusion task including a plurality of word lists to be fused and fusion parameter configuration is received, and the fusion parameter configuration includes relation calculation configuration and field supplement configuration; determining a source vocabulary and a target vocabulary of the current fusion from the plurality of vocabulary; if the source vocabulary and the target vocabulary are the same in type, determining synonymous objects of the source vocabulary and the target vocabulary according to the relation calculation configuration, and supplementing field contents, which are under the synonymous objects and are related to the field supplement configuration, of the source vocabulary into the target vocabulary; and adding the supplemented target vocabulary as a fused vocabulary obtained by the current fusion into the plurality of vocabulary, returning to execute, and outputting a fused vocabulary obtained by the last fusion until the plurality of vocabulary are completely fused. And if the two types are different, performing type conversion on the source word list according to the type of the target word list, and returning to execute. Therefore, multiplexing, fusion and rapid construction of the multi-source heterogeneous word list are realized.

Description

Technical Field

[0001] The present application relates to the field of software technology, and in particular to a multi-scenario fusion method and related devices for a medical knowledge organization system. Background Art

[0002] Medical knowledge organization system (MedKOS) refers to semantic tools that effectively express, organize and utilize medical knowledge content, concepts, terms and their interrelationships according to the inherent logical connections of medical knowledge, such as glossaries, classification tables, subject thesaurus, ontologies, semantic networks, etc. It has the characteristics of cross-language, multi-data source, strong professionalism, large data volume and complex structure. The rich existing resources are mostly completed by professional institutions and experts in the field, and have brought together a lot of expert wisdom.

[0003] In the big data environment, the types and scale of biomedical data are increasing rapidly. A single vocabulary is difficult to cover the entire field, nor can it support the application requirements of multiple languages, multiple disciplines, multiple scenarios and information systems. Since building from scratch is time-consuming and labor-intensive, in order to quickly meet the needs of specific application scenarios, how to reuse, integrate and quickly build multi-source heterogeneous vocabulary has become an urgent problem to be solved. Summary of the invention

[0004] In view of the above problems, this application provides a multi-scenario fusion method and related devices for a medical knowledge organization system to achieve the purpose of reuse, fusion and rapid construction of multi-source heterogeneous vocabulary. The specific scheme is as follows:

[0005] In a first aspect, the present application provides a multi-scenario fusion method for a medical knowledge organization system, the multi-scenario fusion method for a medical knowledge organization system comprising:

[0006] receiving a vocabulary fusion task, wherein the vocabulary fusion task includes multiple vocabularies to be fused and a fusion parameter configuration, wherein the fusion parameter configuration includes a relationship calculation configuration and a field supplement configuration;

[0007] Determine a source vocabulary and a target vocabulary for this fusion from the multiple vocabulary lists;

[0008] If the source vocabulary and the target vocabulary are of the same type, determine the synonym objects of the source vocabulary and the target vocabulary according to the relationship calculation configuration, and add the field content of the source vocabulary under the synonym objects and related to the field supplement configuration to the target vocabulary; add the supplemented target vocabulary as the fused vocabulary obtained by this fusion to the multiple vocabularies, and return to execute the step of determining the source vocabulary and the target vocabulary for this fusion from the multiple vocabularies, until all the multiple vocabularies are fused and the fused vocabulary obtained by the last fusion is output;

[0009] If the source vocabulary and the target vocabulary are of different types, the source vocabulary is converted according to the type of the target vocabulary, and the step of determining synonyms of the source vocabulary and the target vocabulary according to the relationship calculation configuration is returned to execute.

[0010] In a possible implementation, the fusion parameter configuration further includes a vocabulary priority configuration, and the determining of the source vocabulary and the target vocabulary for this fusion from the multiple vocabulary includes:

[0011] Determining the priority of each vocabulary in the plurality of vocabulary based on the vocabulary priority configuration;

[0012] The fusion direction of the multiple vocabularies is determined according to the priority of each vocabularies, and the source vocabularies and the target vocabularies are determined according to the fusion direction.

[0013] In a possible implementation, the source vocabulary and the target vocabulary are both descriptor structure vocabulary, the synonym object is a descriptor, the fusion parameter configuration further includes a matching range configuration and a descriptor fusion range configuration, the matching range configuration is a concept mode, and the field content of the source vocabulary under the synonym object and related to the field supplement configuration is supplemented to the target vocabulary, including:

[0014] Determine a concept matching pair between the source vocabulary and the target vocabulary under the synonym object, wherein the concept matching pair includes a first concept in the source vocabulary and a second concept in the target vocabulary;

[0015] Supplement the target vocabulary with the terms of the first concept as non-preferred terms of the second concept, add the concept attributes of the first concept as the concept attributes of the second concept, add other concepts and attributes in the source vocabulary as non-preferred concepts and attributes of the target vocabulary, and add the descriptive attributes of the source vocabulary as the descriptive attributes of the target vocabulary;

[0016] constructing conceptual relationships within the synonymous objects in the supplemented target vocabulary;

[0017] Based on the configuration of the thesaurus fusion scope, the thesaurus relationship of the source thesaurus under the synonymous object is inherited in the supplemented target thesaurus.

[0018] In a possible implementation, the source vocabulary and the target vocabulary are both concept vocabulary, the synonym object is a concept, the fusion parameter configuration further includes a matching range configuration and a concept fusion range configuration, the matching range configuration is a concept mode, and the field content of the source vocabulary under the synonym object and related to the field supplement configuration is supplemented to the target vocabulary, including:

[0019] The target vocabulary is supplemented by using the preferred terms and non-preferred terms of the source vocabulary under the synonymous object as the non-preferred terms of the target vocabulary under the synonymous object, and the concept attributes of the source vocabulary under the synonymous object as the concept attributes of the target vocabulary under the synonymous object;

[0020] Based on the concept fusion scope configuration, the concept relationship of the source vocabulary under the synonymous object is inherited in the supplemented target vocabulary.

[0021] In a possible implementation, the type of the source vocabulary is a descriptive structure vocabulary, the type of the target vocabulary is a concept structure vocabulary, the fusion parameter configuration further includes a matching range configuration, the matching range configuration is a concept mode or a descriptive model, and the type conversion of the source vocabulary according to the type of the target vocabulary includes:

[0022] When the matching scope is configured as a concept mode, acquiring the descriptor relationships in the source word list and the conceptual relationships between non-preferred concepts and preferred concepts in each descriptor in the source word list; converting the type of the source word list into a concept structure word list according to the acquired descriptor relationships and conceptual relationships;

[0023] In the case where the matching scope is configured as a descriptor mode, non-preferred concepts in each descriptor in the source thesaurus are merged into its preferred concepts, so as to merge all concepts in each descriptor in the source thesaurus into one concept.

[0024] In a possible implementation, the type of the source vocabulary is a concept structure vocabulary, the type of the target vocabulary is a descriptive structure vocabulary, the fusion parameter configuration further includes a matching range configuration, the matching range configuration is a concept mode, and the type conversion of the source vocabulary according to the type of the target vocabulary includes:

[0025] Establishing the descriptors corresponding to the concepts in the source word list, taking each concept as the preferred concept of the descriptor, and taking the concept attribute of each concept as the descriptor attribute of the descriptor and the attribute of its preferred concept;

[0026] The concept relationships in the source vocabulary are obtained, and the obtained concept relationships are converted into descriptive thesaurus relationships, so as to convert the type of the source vocabulary into a descriptive structure vocabulary.

[0027] In a possible implementation, the fusion parameter configuration further includes a relationship conflict configuration, and the step of adding the supplemented target vocabulary as the fusion vocabulary obtained by this fusion to the multiple vocabulary includes:

[0028] Performing conflict processing on the supplemented target vocabulary based on the relationship conflict configuration;

[0029] The target vocabulary after conflict resolution is added to the multiple vocabulary as the fused vocabulary obtained in this fusion.

[0030] A second aspect of the present application provides a vocabulary fusion device for a medical knowledge organization system, the vocabulary fusion device for the medical knowledge organization system comprising:

[0031] A task receiving module, used for receiving a vocabulary fusion task, wherein the vocabulary fusion task includes multiple vocabularies to be fused and a fusion parameter configuration, wherein the fusion parameter configuration includes a relationship calculation configuration and a field supplement configuration;

[0032] A vocabulary fusion module is used to determine the source vocabulary and target vocabulary of this fusion from the multiple vocabularies; if the source vocabulary is of the same type as the target vocabulary, determine the synonym objects of the source vocabulary and the target vocabulary according to the relationship calculation configuration, and supplement the field content of the source vocabulary under the synonym objects and related to the field supplement configuration into the target vocabulary; add the supplemented target vocabulary as the fusion vocabulary obtained in this fusion to the multiple vocabularies, and return to execute the step of determining the source vocabulary and target vocabulary of this fusion from the multiple vocabularies, until all the multiple vocabularies are fused and the fusion vocabulary obtained by the last fusion is output; if the source vocabulary is of different types from the target vocabulary, perform type conversion on the source vocabulary according to the type of the target vocabulary, and return to execute the step of determining the synonym objects of the source vocabulary and the target vocabulary according to the relationship calculation configuration.

[0033] A third aspect of the present application provides an electronic device, comprising at least one processor and a memory connected to the processor, wherein:

[0034] The memory is used to store computer programs;

[0035] The processor is used to execute the computer program so that the electronic device can implement the multi-scenario fusion method of the medical knowledge organization system of the above-mentioned first aspect or any implementation method of the first aspect.

[0036] The fourth aspect of the present application provides a computer program product, including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements the multi-scenario fusion method of the medical knowledge organization system of the above-mentioned first aspect or any implementation method of the first aspect.

[0037] The fifth aspect of the present application provides a computer storage medium, which carries one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement the multi-scenario fusion method of the medical knowledge organization system of the above-mentioned first aspect or any implementation method of the first aspect.

[0038] By means of the above technical scheme, the present application provides a multi-scenario fusion method for a medical knowledge organization system, receiving a vocabulary fusion task, the vocabulary fusion task includes multiple vocabularies to be fused and fusion parameter configurations, the fusion parameter configuration includes a relational calculation configuration and a field supplement configuration; determining the source vocabulary and target vocabulary for this fusion from multiple vocabularies. If the source vocabulary and the target vocabulary are of the same type, the synonym objects of the source vocabulary and the target vocabulary are determined according to the relational calculation configuration, and the field content of the source vocabulary under the synonym object and related to the field supplement configuration is supplemented to the target vocabulary; the supplemented target vocabulary is added to the multiple vocabularies as the fusion vocabulary obtained in this fusion, and the step of determining the source vocabulary and the target vocabulary for this fusion from multiple vocabularies is returned until all the multiple vocabularies are fused and the fusion vocabulary obtained by the last fusion is output. If the source vocabulary and the target vocabulary are of different types, the source vocabulary is converted according to the type of the target vocabulary, and the step of determining the synonym objects of the source vocabulary and the target vocabulary according to the relational calculation configuration is returned. This application can solve the fusion and rapid construction of vocabularies in different application scenarios through fusion parameter configuration, making the reuse and fusion between vocabularies possible, which helps to improve the efficiency of vocabulary construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0040] Figure 1 It is a schematic diagram of the structure of the concept structure vocabulary and the descriptive structure vocabulary;

[0041] Figure 2 A flowchart of a multi-scenario fusion method for a medical knowledge organization system provided in an embodiment of the present application;

[0042] Figure 3 A partial flow chart of a multi-scenario fusion method for a medical knowledge organization system provided in an embodiment of the present application;

[0043] Figure 4 Another partial flow chart of a multi-scenario fusion method for a medical knowledge organization system provided in an embodiment of the present application;

[0044] Figure 5 An example of converting a descriptive structure vocabulary into a concept structure vocabulary provided in an embodiment of the present application;

[0045] Figure 6 An example of inheritance of attributes and relationships from a descriptive structure vocabulary to a concept structure vocabulary provided in an embodiment of the present application;

[0046] Figure 7 A schematic diagram of the structure of a vocabulary fusion device for a medical knowledge organization system provided in an embodiment of the present application;

[0047] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0048] The following describes the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. The terms used in the implementation method section of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application.

[0049] The embodiments of the present application are described below in conjunction with the accompanying drawings. Those skilled in the art will appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0050] The terms "first", "second" etc. in the specification of the application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable in appropriate circumstances, and this is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0051] There are two types of vocabularies in the medical knowledge organization system: concept structure vocabularies and descriptive structure vocabularies. Figure 1 , Figure 1 The following is a structural diagram of the concept structure word list and the descriptive word structure word list. The concept structure word list and the descriptive word structure word list are introduced and explained respectively:

[0052] 1) Concept structure vocabulary. Figure 1As shown in Figure 1, the concept structure vocabulary is a two-level structure vocabulary. The core elements of the vocabulary are concepts and terms. The hierarchical structure of the vocabulary is constructed through the relationship between concepts. Concepts include several synonymous terms. For terms in the same language, one is the preferred term and the others are non-preferred terms. The attributes corresponding to the concepts (hereinafter referred to as concept attributes) include data attributes (i.e. Figure 1 conceptual data attributes in ) and relational attributes (i.e. Figure 1 There are two types of data attributes: data attributes and relationship attributes. Data attributes are attributes with strings, numbers, etc. as attribute values, such as definitions, comments, etc., while relationship attributes are attributes with another concept as attribute value. It essentially establishes a relationship between two concepts.

[0053] 2) Descriptive structure vocabulary. Figure 1 As shown, it is a special type of three-level structured vocabulary, such as the Medical Subject Headings (MeSH) developed by the National Library of Medicine of the United States and the Chinese Medical Subject Headings (CMeSH) compiled and maintained by the Institute of Medical Information of the Chinese Academy of Medical Sciences. The core elements of this vocabulary are descriptors, concepts, and terms. This vocabulary establishes a hierarchical relationship (including superordinate descriptors / subordinate descriptors) between descriptors. A descriptor includes a preferred concept and several non-preferred concepts. Each descriptor has one and only one concept that is the preferred concept of the descriptor, and the others are non-preferred concepts. The concepts also include terms with different word forms. Non-preferred concepts have a certain semantic relationship with the preferred concepts, such as similar, narrow, broad, related, etc. Of course, the attributes of this vocabulary also include data attributes and relationship attributes, which can be established on concepts (hereinafter referred to as concept attributes, including Figure 1 The concept data attributes and concept relationship attributes in the text can also be built on the descriptor (hereinafter referred to as the descriptor attribute, including Figure 1 descriptor data attributes and descriptor relationship attributes in ).

[0054] At present, scholars mainly study the fusion methods between concept structure vocabularies. In the process of constructing the vocabulary, some open source vocabularies are used as sources to match concepts with the vocabulary currently being constructed. The source vocabulary is used to supplement the vocabulary currently being constructed with new concepts, new terms and attribute content, and the data in the source vocabulary is fully integrated into the vocabulary currently being constructed, so as to achieve the purpose of rapid auxiliary construction of the concept structure vocabulary. However, due to the existence of a certain scale of high-quality descriptive structure vocabularies in the rich existing resources, no fusion construction methods and practices between concept structure vocabularies and descriptive structure vocabularies have been proposed. Therefore, there are currently the following main problems:

[0055] (1) The default mode is the full fusion mode of all vocabularies. The fusion parameter configuration is not autonomous enough, resulting in unclear fusion direction. Some unnecessary concepts, relationships and attributes are prone to quality problems and relationship conflicts. The relationship conflicts caused by fusion are not automatically processed, resulting in a burden for manual review. (2) There are differences in data structure between the concept structure vocabulary and the descriptive structure vocabulary, and the fusion method between the concept structure vocabulary cannot be directly adopted. (3) The granularity of the two vocabularies is different. No effective method has been proposed for the use of the concept structure vocabulary and the descriptive structure vocabulary in various construction scenarios.

[0056] Since the data structures of the concept structure vocabulary and the descriptor structure vocabulary are very different, if it is necessary to use one of the two vocabulary as a source to assist the construction of the other vocabulary, completely copying the indifferent full inheritance and fusion method used for the fusion of concept structure vocabulary, some problems will arise due to the different data structures. For example, when the descriptor structure vocabulary is used as a source to assist the construction of the concept structure vocabulary, how to transform the three-level descriptor structure into a two-level concept structure, and how to deal with the concepts in the descriptor when the relationship is merged; conversely, when the concept structure vocabulary is used as a source to assist the construction of the descriptor structure vocabulary, how to transform the two-level concept structure into a three-level descriptor structure, and how to inherit and fuse the attributes when merging with the concepts in the descriptor. Therefore, a new implementation plan is needed to perform fusion construction between these two vocabulary types and use one of them as a source to assist the construction of the other vocabulary.

[0057] The present application solves the problem of free conversion between concept structure vocabulary and descriptive structure vocabulary under different fusion modes and scenarios, provides a method and device for vocabulary fusion and relationship construction under different structural combinations, automatically generates a new vocabulary after fusion, improves the efficiency of constructing the two structural vocabulary, and promotes the development of the field of medical knowledge organization.

[0058] See also Figure 2 , Figure 2 A schematic diagram of a multi-scenario fusion method for a medical knowledge organization system provided in an embodiment of the present application. Figure 2 As shown, a multi-scenario fusion method for a medical knowledge organization system provided in an embodiment of the present application may include steps S10 to S40, and these steps are described in detail below.

[0059] S10, receiving a vocabulary fusion task, wherein the vocabulary fusion task includes multiple vocabularies to be fused and fusion parameter configuration, and the fusion parameter configuration includes a relationship calculation configuration and a field supplement configuration.

[0060] In the embodiment of the present application, a vocabulary fusion task can be received through human-computer interaction, and the vocabulary fusion task includes multiple vocabularies to be fused and fusion parameter configuration. Among them, the fusion parameter configuration includes a relationship calculation configuration and a field supplement configuration. The relationship calculation configuration is used to indicate the relationship calculation algorithm (such as Jaccard coefficient, cosine similarity, etc.) used when executing the vocabulary fusion task, the language of the relationship calculation (such as Chinese, English, etc.), and the matching rules of the relationship calculation (such as ignoring case, comparing term prototypes, etc.), while the field supplement configuration is used to indicate the supplementary field range (such as preferred terms, non-preferred terms, hypernyms or hypernyms, hyponyms or hyponyms, related concepts, concept attributes or descriptor attributes, etc.).

[0061] S20, determining a source vocabulary and a target vocabulary for this fusion from multiple vocabulary lists.

[0062] In the embodiment of the present application, by analyzing different fusion processing scenarios, the source vocabulary and target vocabulary for each fusion are selected according to the compilation purpose, authority, system integrity, etc. to determine the fusion direction of the vocabulary, thereby determining the source vocabulary and target vocabulary for each fusion.

[0063] Since the quality of the vocabularies that need to be merged varies, it is necessary to determine the priority of the vocabularies based on the purpose of compilation, authority, system integrity, etc., so as to determine the source vocabularies and target vocabularies for each fusion. The target vocabularies are the vocabularies to be constructed, and they are also the vocabularies that need to supplement concepts, entry words, and relationships. Generally, a table with a relatively complete system structure and finer granularity is selected as the target vocabularies; the source vocabularies are the material vocabularies that provide new concepts, entry words, and relationships for the target vocabularies. After selecting the source vocabularies and the target vocabularies, it is equivalent to determining the direction of fusion of the source vocabularies to the target vocabularies. Therefore, in the process of constructing the concept structure vocabularies and the narrative structure vocabularies, based on the needs and scenarios of the vocabularies, the source vocabularies are selected from the existing vocabularies in the platform library as fusion materials to determine the following fusion directions:

[0064] Scenario 1): The source vocabulary is the concept structure vocabulary and the target vocabulary is the concept structure vocabulary; Scenario 2): The source vocabulary is the descriptive structure vocabulary and the target vocabulary is the concept structure vocabulary; Scenario 3): The source vocabulary is the concept structure vocabulary and the target vocabulary is the descriptive structure vocabulary; Scenario 4): The source vocabulary is the descriptive structure vocabulary and the target vocabulary is the descriptive structure vocabulary.

[0065] When merging, you can select all or some descriptors or concepts in the source vocabulary, or all descriptors or concepts under a certain hierarchical structure branch, and then merge other contents based on the hierarchical structure of the target vocabulary. When merging multiple vocabularies, you can split it into a combination of the above scenarios. For example, when two descriptor structure vocabularies are merged with a concept structure vocabularies, and the target vocabulary is a concept structure vocabularies, you can first execute scenario 4) and then scenario 2).

[0066] In a possible implementation, the source vocabulary and target vocabulary of each fusion are determined by configuring vocabulary priority. In this regard, the fusion parameter configuration also includes vocabulary priority configuration. In step S20, "determining the source vocabulary and target vocabulary of this fusion from multiple vocabulary" can be performed in the following steps:

[0067] Determine the priority of each vocabulary among the multiple vocabulary based on the vocabulary priority configuration;

[0068] The fusion direction of multiple vocabularies is determined according to the priority of each vocabularies, and the source vocabularies and the target vocabularies are determined according to the fusion direction.

[0069] Specifically, in the embodiment of the present application, the vocabulary priority configuration includes multiple methods such as using vocabulary weight as vocabulary priority, using the number of selected concepts in the vocabulary as vocabulary priority, and using fusion direction as priority. The following are respectively described:

[0070] 1) The weight of the vocabulary is used as the priority of the vocabulary. The processing order is determined by the weight of the vocabulary. After manual evaluation according to the authority of the vocabulary, data quality and other dimensions, a weight is set for each vocabulary, and then the fusion processing is carried out in order of the vocabulary weight from high to low, so that a better fusion result may be obtained. For example, after manual evaluation, it is determined that the weight of vocabulary A is 100, the weight of vocabulary B is 90, and the weight of vocabulary C is 85. At this time, it can be determined that the priority of vocabulary A>the priority of vocabulary B>the priority of vocabulary C. Based on this, the fusion direction can be determined as follows: the first fusion fuses vocabulary B into vocabulary A to obtain vocabulary AB, and the second fusion fuses vocabulary C into vocabulary AB to obtain vocabulary ABC, that is, the source vocabulary in the first fusion is vocabulary B and the target vocabulary is vocabulary A, and the source vocabulary in the second fusion is vocabulary C and the target vocabulary is vocabulary AB.

[0071] 2) The number of selected concepts in the vocabulary is used as the vocabulary priority. The vocabulary is processed in descending order according to the number of selected concepts. For example, the number of concepts in vocabulary A is 10,000, the number of concepts in vocabulary B is 5,000, and the number of concepts in vocabulary C is 3,000. At this time, it can be determined that the priority of vocabulary A is greater than the priority of vocabulary B and greater than the priority of vocabulary C. Based on this, the fusion direction can be determined as follows: the first fusion merges vocabulary B into vocabulary A to obtain vocabulary AB, and the second fusion merges vocabulary C into vocabulary AB to obtain vocabulary ABC, that is, the source vocabulary in the first fusion is vocabulary B and the target vocabulary is vocabulary A, and the source vocabulary in the second fusion is vocabulary C and the target vocabulary is vocabulary AB. This strategy is suitable for situations where the number of concepts of the vocabularies to be fused is relatively large, and it is best to merge the vocabulary with a small number of concepts into the vocabulary with a large number of concepts.

[0072] 3) Take the fusion direction as the priority. If there are both concept structure vocabularies and descriptive structure vocabularies in the vocabularies to be fused, the fusion direction can be determined according to the type of the required vocabularies. For example, vocabulary A is a descriptive structure vocabularies, vocabularies B and vocabularies C are both concept structure vocabularies, and the weight of vocabularies B is higher than that of vocabularies C. If the type of the required vocabularies is a descriptive structure vocabularies, then the priority of vocabularies A> the priority of vocabularies B> the priority of vocabularies C. Based on this, the fusion direction can be determined as follows: the first fusion fuses vocabularies B into vocabularies A to obtain vocabularies AB, and the second fusion fuses vocabularies C into vocabularies AB to obtain vocabularies ABC (vocabulary ABC is the descriptive structure vocabularies), that is, the source vocabularies in the first fusion are vocabularies B and the target vocabularies are vocabularies A, and the source vocabularies in the second fusion are vocabularies C and the target vocabularies are AB.

[0073] S30, if the source vocabulary and the target vocabulary are of the same type, determine the synonym objects of the source vocabulary and the target vocabulary according to the relationship calculation configuration, and add the field content of the source vocabulary under the synonym objects and related to the field supplement configuration to the target vocabulary; add the supplemented target vocabulary as the fused vocabulary obtained in this fusion to the multiple vocabularies, and return to execute step S20 until all the multiple vocabularies are fused and the fused vocabulary obtained by the last fusion is output.

[0074] In an embodiment of the present application, if the source vocabulary and the target vocabulary are of the same type, synonymous objects with the same semantics as the target vocabulary can be determined based on the relational calculation configuration. Specifically, when the source vocabulary and the target vocabulary are both descriptive structure vocabulary, the synonymous objects are descriptive words; and when the source vocabulary and the target vocabulary are both concept structure vocabulary, the synonymous objects are concepts. Furthermore, the relevant field contents of the source vocabulary under the synonymous objects can be supplemented to the target vocabulary according to the field supplement configuration.

[0075] In practical applications, the mapping from the data model of the source vocabulary to the data model of the target vocabulary can be set as needed. The data model describes the data structure of the relevant objects (such as descriptors, concepts, and terms) in the vocabulary, that is, what attributes are included. Since the data models of different vocabularies may be different, a one-to-one mapping relationship is established between the data model of the source vocabulary and the data model of the target vocabulary through a combination of automatic mapping and manual settings. In this way, when the data of the source vocabulary is merged into the target vocabulary, the attributes in the source vocabulary can be automatically converted into the corresponding attributes of the target vocabulary.

[0076] In a possible implementation, the matching scope and fusion scope can be configured based on the requirements of the matching scope and fusion content in different scenarios. In this regard, the types of the source vocabulary and the target vocabulary are both descriptor structure vocabulary, the synonym object is the descriptor, and the fusion parameter configuration also includes the matching scope configuration and the descriptor fusion scope configuration, and the matching scope configuration is the concept mode.

[0077] It should be noted that, in the conceptual mode, thesaurus of descriptive structure is matched and integrated from the perspective of concepts.

[0078] It should also be noted that the configuration of the fusion scope of thesaurus includes full mode and selective inheritance mode. In full mode, the target vocabulary will fuse all thesaurus, concepts, terms, and relationships and attributes of the source vocabulary to establish data model mapping; in selective inheritance mode, the optional fusion scope of the thesaurus includes: repeated thesaurus, repeated concepts, and all thesaurus; the optional fusion information items include: preferred concepts, non-preferred concepts, preferred terms, non-preferred terms, hierarchical relationships, semantic relationships, and data attributes. Since the hierarchical structure of the thesaurus is established between thesaurus, the same hierarchical structure also exists between the preferred concepts of the thesaurus; however, since the concepts within the thesaurus are not completely synonymous, this hierarchical structure is not completely valid between non-preferred concepts. In order to ensure the accuracy of the fusion content, sometimes only the necessary concepts, attributes, and relationship content need to be selected for fusion. Therefore, in this mode, the fusion process will inherit all the contents of the target vocabulary, but will filter the contents of the source vocabulary. For example, when the source vocabulary is a descriptive structure vocabulary and the fusion scope is "repeated descriptors" and the fusion information items are "preferred concepts", "hierarchical relationships", "semantic relationships" and "data attributes", only the preferred concepts of the descriptors in the source vocabulary will be fused. The fusion process will ignore the non-preferred concepts in the descriptors and only copy the preferred concepts of the descriptors to the target vocabulary, inheriting the hierarchical relationships, semantic relationships and data attributes of the descriptors to which the preferred concepts belong.

[0079] See also Figure 3 , Figure 3 A partial flow chart of a multi-scenario fusion method for a medical knowledge organization system provided in an embodiment of the present application. Figure 3 As shown, an embodiment of the present application provides a multi-scenario fusion method for a medical knowledge organization system, wherein step S30 of "supplementing the field contents in the source vocabulary under synonymous objects and related to the field supplement configuration into the target vocabulary" may include steps S3011 to S3014, and these steps are described in detail below.

[0080] S3011, determining concept matching pairs under synonymous objects in a source vocabulary and a target vocabulary, wherein the concept matching pairs include a first concept in the source vocabulary and a second concept in the target vocabulary.

[0081] S3012, supplement the target vocabulary with the terms of the first concept as non-preferred terms of the second concept, add the concept attributes of the first concept as concept attributes of the second concept, add other concepts and attributes in the source vocabulary as non-preferred concepts and attributes of the target vocabulary, and add the descriptive attributes of the source vocabulary as descriptive attributes of the target vocabulary.

[0082] S3013, constructing conceptual relationships within synonymous objects in the supplemented target vocabulary.

[0083] S3014, based on the configuration of the thesaurus fusion scope, the thesaurus relationship of the source thesaurus under the synonym object is inherited in the supplemented target thesaurus.

[0084] Specifically, in the embodiment of the present application, both the source vocabulary and the target vocabulary are descriptive structure vocabulary, and correspondingly, the synonym object is the descriptive word. Since the matching scope is configured as a concept mode, the concepts in the source vocabulary and the target vocabulary can be matched separately in the descriptive word, and the process is as follows:

[0085] After data model mapping, the source vocabulary and the target vocabulary are matched for similarity, and the thesaurus in the source vocabulary is matched to find the thesaurus that is repeated in the target vocabulary, that is, the synonym objects. The concept matching pairs that match the source vocabulary and the target vocabulary under the synonym objects are determined, and the concept matching pairs contain the first concept in the source vocabulary and the second concept in the target vocabulary.

[0086] Add the terms of the first concept as non-preferred terms of the second concept to the target vocabulary, add the concept attributes of the first concept as concept attributes of the second concept, add other concepts and attributes in the source vocabulary as non-preferred concepts and attributes of the target vocabulary, and add all thesaurus attributes of the source vocabulary as thesaurus attributes of the target vocabulary.

[0087] Furthermore, for the target vocabulary after supplementation, the concept relationship within its synonymous objects can be constructed. In order to retain the concept relationship within the synonymous objects of the source vocabulary to the greatest extent during the fusion process, as a reference for the evaluation of automatic fusion results, all concepts, terms, relationships, attributes (including concept attributes and descriptor attributes) supplemented from the source vocabulary in the target vocabulary are marked with their sources, and the concept relationship within the synonymous objects is automatically constructed, including: the inverse relationship between isPreferredConceptOf (is the preferred concept of...) and hasPreferredConcept (has a preferred concept) is established between the non-preferred concepts and preferred concepts in the source vocabulary, and the hasSameDescriptorWith (has the same descriptor) relationship is established between the concepts belonging to the same descriptor in the source vocabulary.

[0088] Assume that thesaurus A in the source vocabulary matches thesaurus B in the target vocabulary, and that thesaurus A includes preferred concept A1, non-preferred concept A2, and non-preferred concept A3, and that thesaurus B includes preferred concept B1 and non-preferred concept B2. Non-preferred concept A2 in thesaurus A and preferred concept B1 in thesaurus B form a concept matching pair. Before the source vocabulary is merged with the target vocabulary, the isPreferredConceptOf, hasPreferredConcept, and hasSameDescriptorWith relationships are automatically established between the preferred concept and the non-preferred concept in thesaurus A in the source vocabulary, including: (A1, isPreferredConceptOf, A2), (A1, isPreferredConceptOf, A3), (A2, hasPreferredConcept, A1), (A3, hasPreferredConcept, A1), and (A2, hasSameDescriptorWith, A3). During fusion, the preferred terms of the preferred concept B1 remain unchanged after merging the non-preferred concept A2. All terms of the non-preferred concept A2 are used as non-preferred terms of the preferred concept B1. The preferred concept A1 and non-preferred concept A3 in the descriptor A are merged into the non-preferred concepts of the descriptor B. Therefore, the relationship between concepts in the source vocabulary descriptor A is transformed into: (A1, isPreferredConceptOf, B1), (A1, isPreferredConceptOf, A3), (B1, hasPreferredConcept, A1), (A3, hasPreferredConcept, A1), (B1, hasSameDescriptorWith, A3).

[0089] Furthermore, for the target vocabulary after the concept relationship is supplemented, the relationship between the terms under the synonymous object in the inherited source vocabulary can also be configured based on the scope of the fusion of terms. Specifically, in the embodiment of the present application, in the full mode, the hierarchical relationship and semantic relationship of the source vocabulary and the target vocabulary will be inherited respectively, and the superordinate terms, subordinate terms, and semantically related terms that do not have duplicate terms in the target vocabulary will be added to the target vocabulary as newly added terms. In the selective inheritance mode, if you choose to supplement the target terms to express the entry words, relationships, and attributes of the concepts, then for the terms in the target vocabulary that are repeated in the source vocabulary, the concepts, terms, and data attributes will be supplemented based on the terms in the source vocabulary. In addition to duplicate terms, terms that are in the source vocabulary but not in the target vocabulary will no longer be supplemented, so there will be no newly added terms in the formed fusion vocabulary.

[0090] In a possible implementation, the matching range and fusion range can be configured based on different matching ranges and fusion content requirements. In this regard, the source vocabulary and the target vocabulary are both concept vocabulary, the synonym object is the concept, and the fusion parameter configuration also includes the matching range configuration and the concept fusion range configuration, and the matching range configuration is the concept mode.

[0091] It should be noted that, in the conceptual mode, the conceptual structure vocabulary is matched and integrated from the perspective of concepts.

[0092] It should also be noted that the concept fusion scope configuration includes full mode and selective inheritance mode. In the full mode, the target vocabulary will integrate all concepts, terms, and relationships and attributes of the source vocabulary to establish data model mapping; in the selective inheritance mode, the optional fusion scope of the concept structure vocabulary includes: repeated concepts, all concepts; fusion information items include: preferred terms, non-preferred terms, hierarchical relationships, semantic relationships, and data attributes.

[0093] See also Figure 4 , Figure 4 Another partial flow chart of a multi-scenario fusion method of a medical knowledge organization system provided in an embodiment of the present application. Figure 4 As shown, an embodiment of the present application provides a multi-scenario fusion method for a medical knowledge organization system, wherein step S30 of "supplementing the field contents of the source vocabulary under synonymous objects and related to the field supplement configuration into the target vocabulary" may include steps S3021 to S3022, and these steps are described in detail below.

[0094] S3021, the target vocabulary is supplemented by using the preferred terms and non-preferred terms of the source vocabulary under the synonymous objects as the non-preferred terms of the target vocabulary under the synonymous objects, and the concept attributes of the source vocabulary under the synonymous objects as the concept attributes of the target vocabulary under the synonymous objects.

[0095] S3022, based on the concept fusion scope configuration, inherit the concept relationship of the source vocabulary under the synonym object in the supplemented target vocabulary.

[0096] Specifically, in the embodiment of the present application, both the source vocabulary and the target vocabulary are concept structure vocabulary, and the corresponding synonym objects are concepts. Since the matching scope is configured as a concept mode, the concepts of the source vocabulary and the target vocabulary can be matched separately, and the process is as follows:

[0097] After data model mapping, the concepts in the source vocabulary and the target vocabulary are matched for similarity, and the concepts in the source vocabulary are matched to find the concepts that are repeated in the target vocabulary, that is, synonyms. The preferred terms and non-preferred terms in the source vocabulary under the synonyms are added to the target vocabulary as non-preferred terms in the target vocabulary under the synonyms, and the concept attributes in the source vocabulary under the synonyms are added as the concept attributes in the target vocabulary under the synonyms.

[0098] Furthermore, for the supplemented target vocabulary, the conceptual relationship of the source vocabulary under the synonymous object can also be inherited based on the scope of concept fusion. Specifically, in the embodiment of the present application, in the full mode, the hierarchical relationship and semantic relationship of the source vocabulary and the target vocabulary will be inherited respectively, and the superordinate concepts, subordinate concepts, and semantically related concepts that do not have repeated concepts in the target vocabulary will be supplemented to the target vocabulary through newly added concepts. In the selective inheritance mode, if only for the purpose of supplementing the entry words, the terms of the repeated concepts can be selectively fused, and the data attributes, hierarchical relationships, and semantic relationships of the repeated concepts will no longer be fused. In this case, there will be no newly added concepts in the fused vocabulary formed.

[0099] In addition, in the embodiment of the present application, the supplemented target vocabulary is added to multiple vocabulary as the fused vocabulary obtained by this fusion, and the process returns to step S20 to continue the next fusion until all vocabulary are fused and the fused vocabulary obtained by the last fusion is output.

[0100] The final output fusion vocabulary is a descriptive structure vocabulary or a concept structure vocabulary. If it is a descriptive structure vocabulary, the descriptive table, concept table, term table, relationship attribute table, and data attribute table can be stored separately. If it is a concept structure vocabulary, the concept table, term table, relationship attribute table, and data attribute table can be stored separately.

[0101] Due to the different granularity of the word lists, many types of relationship conflicts are prone to occur during the fusion process. In order to reduce the burden of manual review caused by relationship conflicts in the fusion results, the fusion parameter configuration also includes relationship conflict configuration. In this regard, the embodiment of the present application provides a multi-scenario fusion method for a medical knowledge organization system, wherein in step S30, "adding the supplemented target word list as the fusion word list obtained by this fusion to multiple word lists" can adopt the following steps:

[0102] Based on the relationship conflict configuration, the conflict of the supplemented target vocabulary is handled.

[0103] The target vocabulary after conflict resolution is added to multiple vocabulary as the fusion vocabulary obtained in this fusion.

[0104] Specifically, in the embodiments of the present application, for different types of relationship conflicts, the relationship priority order is also followed: synonymous relationship > hierarchical relationship > semantic relationship. The target vocabulary is used for attribute conflicts, and relationships with higher priorities are retained first, while relationships with lower priorities are deleted. An automatic processing method for common relationship conflicts has been developed, including: when a synonymous relationship conflicts with a hierarchical relationship, the synonymous relationship is retained and the hierarchical relationship is deleted; when a synonymous relationship conflicts with a sibling concept, the synonymous relationship is retained and the sibling concept is deleted; when a hierarchical relationship conflicts with a sibling concept, the hierarchical relationship is retained and the sibling concept is deleted; when a hierarchical relationship is redundant, the coarse-grained hierarchical relationship is deleted and the fine-grained hierarchical relationship is retained, etc., to reduce the burden of manual review of the fused vocabulary.

[0105] In addition, for other conflicts besides those mentioned above, based on the principle of minimum conflict, all source vocabulary descriptors, concepts, terms or relationships that cause relationship conflicts will not be automatically merged, and the data causing the conflicts will be trimmed before merging.

[0106] S40, if the source vocabulary and the target vocabulary are of different types, the source vocabulary is converted according to the type of the target vocabulary, and the process returns to the step of "determining synonyms of the source vocabulary and the target vocabulary according to the relationship calculation configuration" in step S30.

[0107] Specifically, in an embodiment of the present application, when the source vocabulary is a descriptive structure vocabulary and the target vocabulary is a conceptual structure vocabulary, or when the source vocabulary is a conceptual structure vocabulary and the target vocabulary is a descriptive structure vocabulary, the source vocabulary is converted according to the type of the target vocabulary so that the type of the converted source vocabulary is the same as the type of the target vocabulary.

[0108] In a possible implementation, the type of the source vocabulary is a descriptive structure vocabulary, the type of the target vocabulary is a concept structure vocabulary, and the fusion parameter configuration also includes a matching range configuration, and the matching range configuration is a concept pattern or a descriptive pattern. It should be noted that when the descriptive structure vocabulary is converted to a concept structure vocabulary, a coarse-grained concept structure vocabulary can be established in a descriptive pattern, and a fine-grained concept structure vocabulary can be established in a concept pattern. In this regard, an embodiment of the present application provides a multi-scenario fusion method for a medical knowledge organization system, wherein "converting the source vocabulary according to the type of the target vocabulary" in step S40 may include the following steps:

[0109] When the matching scope is configured as a concept mode, the descriptor relationships in the source word list and the conceptual relationships between non-preferred concepts and preferred concepts in each descriptor in the source word list are obtained; according to the obtained descriptor relationships and conceptual relationships, the type of the source word list is converted into a concept structure word list;

[0110] When the matching scope is configured as the descriptor mode, the non-preferred concepts in each descriptor in the source thesaurus are merged into the preferred concepts thereof, so as to merge all the concepts in each descriptor in the source thesaurus into one concept.

[0111] Specifically, in an embodiment of the present application, if the matching scope is configured as a conceptual mode, on the one hand, the relationship between the descriptors in the source vocabulary (i.e., the descriptor relationship) is obtained, and on the other hand, the conceptual relationship between the non-preferred concepts and the preferred concepts in each descriptor is obtained, and the non-preferred concepts in the same descriptor are processed according to the conceptual relationship, including: merging with the preferred concept (such as "accidentally discovered" and "accidentally discovered"), or converting to a subordinate concept of the preferred concept, or converting to a sibling concept of the preferred concept, and further, combining the descriptor relationship to convert into a concept structure vocabulary.

[0112] Specifically, when a non-preferred concept has a similar relationship with a preferred concept, the non-preferred concept is merged with the preferred concept; when a non-preferred concept has a narrow relationship with a preferred concept, the non-preferred concept is transformed into a subordinate concept of the preferred concept, such as "antemortem diagnosis" and "diagnosis", and establishes a hasPreferredConcept relationship with the preferred concept, and establishes a hasSameDescriptorWith relationship with other split non-preferred concepts; when a non-preferred concept has a broad relationship or a related relationship with a preferred concept, the non-preferred concept is transformed into a sibling concept of the preferred concept.

[0113] See also Figure 5 , Figure 5 This is an example of converting a descriptive structure vocabulary into a concept structure vocabulary provided in an embodiment of the present application. Figure 5 As shown, the descriptor structure word list includes descriptors A, B and C. Descriptor B is a subordinate descriptor of descriptor A, and descriptor C is a subordinate descriptor of descriptor B. Descriptor A contains preferred concept A1 and non-preferred concept A2, descriptor B contains preferred concept B1, non-preferred concept B2, non-preferred concept B3, non-preferred concept B4 and non-preferred concept B5, and descriptor C contains preferred concept C1 and non-preferred concept C2. Among them, non-preferred concept A2 has a narrower relationship with preferred concept A1, non-preferred concept B2 has a similar relationship with preferred concept B1, non-preferred concept B3 has a narrower relationship with preferred concept B1, non-preferred concept B4 has a broader relationship with preferred concept B1, non-preferred concept B5 has a related relationship with preferred concept B1, and non-preferred concept C2 has a narrower relationship with preferred concept C1.

[0114] The structure word list of thesaurus is automatically converted into the concept structure word list based on rules. First, the concepts in the thesaurus are converted into independent concepts, that is, the preferred concept A1 is converted into concept A1, the non-preferred concept A2 is converted into concept A2, the preferred concept B1 is converted into concept B1, the non-preferred concept B2 is converted into concept B2, the non-preferred concept B3 is converted into concept B3, the non-preferred concept B4 is converted into concept B4, the non-preferred concept B5 is converted into concept B5, the preferred concept C1 is converted into concept C1, and the non-preferred concept C2 is converted into concept C2.

[0115] The concepts converted from the preferred concepts of descriptors inherit the hierarchical relationship between descriptors, that is, concept B1 inherits the hierarchical relationship of descriptor B and serves as a subordinate concept of concept A1; concept C1 inherits the hierarchical relationship of descriptor C and serves as a subordinate concept of concept B1.

[0116] Concepts converted from non-preferred concepts of descriptors are handled according to their relationship with preferred concepts. If a non-preferred concept has a narrower relationship with a preferred concept, it is treated as a subordinate concept of its preferred concept, such as concept A2 as a subordinate concept of concept A1, concept B3 as a subordinate concept of concept B1, and concept C2 as a subordinate concept of concept C1. If a non-preferred concept has a broader or related relationship with a preferred concept, it is treated as a sibling concept of its preferred concept, such as concept B4 and concept B5 as sibling concepts of concept B1, that is, as a subordinate concept of concept A1. If a non-preferred concept has a similarity relationship with a preferred concept, the non-preferred concept is merged into its preferred concept to become one concept. If concept B2 has a similarity relationship with concept B1, concept B2 is merged into concept B1. After the merger, the preferred terms of concept B1 remain unchanged, and all terms of concept B2 (including preferred terms and non-preferred terms) are converted into non-preferred terms of concept B1 to form a Figure 5 Concept B1 (including B2) in.

[0117] In practical applications, after the preferred concepts and non-preferred concepts in the same thesaurus are split, they may become subordinate concepts or sibling concepts. Establishing relationships between these concepts and the original preferred concepts can help preserve the original information of the thesaurus and facilitate the interaction of application data based on the thesaurus. The isPreferredConceptOf / hasPreferredConcept relationship pairs are established between the preferred concepts and non-preferred concepts in the thesaurus in the source thesaurus, including (preferred concept, isPreferredConceptOf, non-preferred concept), (non-preferred concept, hasPreferredConcept, preferred concept), which refer to each other and identify the preferred concepts and non-preferred concepts of the thesaurus in the thesaurus as the source thesaurus; the hasSameDescriptorWith relationship pair (non-preferred concept, hasSameDescriptorWith, non-preferred concept) is established between non-preferred concepts, indicating non-preferred concepts that belong to the same thesaurus in the thesaurus as the source thesaurus.

[0118] Continue with Figure 5 Taking as an example, the relationship between the preferred concepts and non-preferred concepts of the descriptors in the source vocabulary is constructed in the concept structure vocabulary, including (Concept A1, isPreferredConceptOf, Concept A2), (Concept A2, hasPreferredConcept, Concept A1), (Concept C1, isPreferredConceptOf, Concept C2), (Concept C2, hasPreferredConcept, Concept C1), etc.

[0119] In addition, in the converted concept structure vocabulary, the preferred concepts in the thesaurus in the source vocabulary not only inherit their own terms, concept attributes and relationships, but also convert the thesaurus attributes into their concept attributes. In addition, the non-preferred concepts in the thesaurus in the source vocabulary not only inherit their own terms, concept attributes and relationships, but also can inherit some thesaurus attributes, such as classification number, subject term-sub-subject term combination relationship (AllowableQualifiersList), related relationship (Seealso), semantic type (Semantic Type), etc.

[0120] See also Figure 6 , Figure 6 An example of inheritance of attributes and relationships from a descriptive structure vocabulary to a concept structure vocabulary is provided in an embodiment of the present application.

[0121] like Figure 6As shown, the descriptor A in the descriptor structure thesaurus contains preferred concept A1 and non-preferred concept A2. Preferred concept A1 includes preferred concept term T1, preferred concept data attribute D1 and preferred concept relationship attribute R1. Non-preferred concept A2 includes non-preferred concept term T2, non-preferred concept data attribute D2 and non-preferred concept relationship attribute R2. In addition, the descriptor structure thesaurus also includes descriptor data attribute D3, descriptor data attribute D4, descriptor relationship attribute R3 and descriptor relationship attribute R4.

[0122] In the converted concept structure vocabulary, the preferred concept A1 is concept A1, and the non-preferred concept A2 is concept A2. Concept A1 converted from the preferred concept in the thesaurus inherits not only the terms and attributes of the preferred concept A1 itself, but also all the attributes of the thesaurus to which it belongs. For example, term T1, data attribute D1, and relationship attribute R1 are inherited from the terms and attributes of the preferred concept A1; data attribute D3, data attribute D4, relationship attribute D3, and relationship attribute D4 are inherited from the attributes of thesaurus A. Concept A2 converted from the non-preferred concept of the thesaurus inherits not only the terms and attributes of the non-preferred concept A2 itself, but also selectively inherits some attributes of the thesaurus to which it belongs. For example, term T2, data attribute D2, and relationship attribute R2 are inherited from the terms and attributes of the non-preferred concept A2; data attribute D3 and relationship attribute R3 are inherited from the attributes of thesaurus A; thesaurus data attribute D4 and thesaurus relationship attribute R4 of thesaurus A are not inherited. In addition, if the matching scope is configured as thesaurus mode, all concepts in each thesaurus in the source vocabulary are merged into one concept, that is, the non-preferred concepts are merged into their preferred concepts, and the preferred terms of the merged concepts are the preferred terms of the preferred concepts before the merger. The merged concepts inherit the data attributes of the thesaurus to which they belong and all the concepts in the thesaurus, as well as the relationship attributes of the thesaurus. The merged concepts inherit the hierarchical relationship and semantic relationship of the thesaurus to which they belong. This mode is suitable for the construction of a coarse-grained vocabulary.

[0123] In a possible implementation, the type of the source vocabulary is a concept structure vocabulary, the type of the target vocabulary is a descriptive structure vocabulary, and the fusion parameter configuration also includes a matching range configuration, and the matching range configuration is a concept pattern. In this regard, the embodiment of the present application provides a multi-scenario fusion method for a medical knowledge organization system, wherein in step S40, "converting the source vocabulary according to the type of the target vocabulary" may include the following steps:

[0124] Establishing the descriptors corresponding to the concepts in the source word list, taking each concept as the preferred concept of the descriptor, and taking the concept attribute of each concept as the descriptor attribute of the descriptor and the attribute of its preferred concept;

[0125] The concept relationships in the source vocabulary are obtained, and the obtained concept relationships are converted into descriptive thesaurus relationships, so as to convert the type of the source vocabulary into a descriptive structure vocabulary.

[0126] Specifically, a thesaurus corresponding to each concept is established, each concept is converted into a preferred concept of the corresponding thesaurus, and the conceptual attributes of each concept are used as the thesaurus attributes of the corresponding thesaurus and the conceptual attributes of the preferred concept, thus forming a thesaurus containing only one preferred concept.

[0127] Furthermore, conceptual relationships representing hierarchical relationships between concepts in the source vocabulary are obtained and converted into descriptor relationships representing hierarchical relationships between descriptors. For example, conceptual relationships in the source vocabulary include (concept A subordinate concept Concept A1) (concept A subordinate concept Concept A2) (concept A1 superordinate concept Concept A) (concept A2 superordinate concept Concept A), which can be converted into descriptor relationships (descriptor A subordinate descriptor Descriptor A1) (descriptor A subordinate descriptor Descriptor A2) (descriptor A1 superordinate descriptor Descriptor A) (descriptor A2 superordinate descriptor Descriptor A).

[0128] From the above description, vocabulary fusion is not just a simple merger of terms, concepts and relationships. It is necessary to achieve consistency in synonymous relationships, hierarchical relationships, relationship attributes and data attributes during the fusion process, eliminate semantic conflicts caused by the fusion process, and form a structurally consistent fusion vocabulary. In order to avoid a large number of relationships or structural conflicts caused by ambiguity, the order of fusion is: synonymous relationships > hierarchical relationships > relationship attributes > data attributes.

[0129] There are mainly two types of vocabulary structures: descriptive structure vocabulary and concept structure vocabulary. In the process of constructing any vocabulary, a large number of existing vocabulary results of the two structures need to be reused and integrated. The embodiment of the present application provides a multi-scenario fusion method of a medical knowledge organization system, which solves the problem of selective fusion and relationship construction in multiple scenarios of concept structure vocabulary and descriptive structure vocabulary. The present application has the following advantages:

[0130] 1) It provides the function of autonomous configuration of fusion parameters. Through autonomous configuration, the fusion content and process can be strictly constrained, which can reduce the relationship conflicts caused by the integration of worthless content;

[0131] 2) Through comparative analysis of descriptive structure vocabulary and concept structure vocabulary, a method for automatic conversion of the two vocabulary data structures and automatic relationship construction was proposed, solving the problem of data model differences between the two structure vocabulary;

[0132] 3) For the four scenarios of concept structure vocabulary-concept structure vocabulary, concept structure vocabulary-descriptive structure vocabulary, descriptive structure vocabulary-concept structure vocabulary, and descriptive structure vocabulary-descriptive structure vocabulary, relationship fusion and automatic construction methods are proposed respectively, making it possible to reuse and fuse the two structures of vocabularies in multiple scenarios, which helps to greatly improve the efficiency of vocabulary construction and reduce the manual burden.

[0133] The above introduces a multi-scenario fusion method for a medical knowledge organization system provided in an embodiment of the present application. The following will introduce a device for executing the multi-scenario fusion method for the medical knowledge organization system.

[0134] See also Figure 7 , Figure 7 A schematic diagram of the structure of a vocabulary fusion device for a medical knowledge organization system provided in an embodiment of the present application. Figure 7 As shown, the vocabulary fusion device of the medical knowledge organization system provided in the embodiment of the present application includes:

[0135] The task receiving module 10 is used to receive a vocabulary fusion task, wherein the vocabulary fusion task includes multiple vocabularies to be fused and fusion parameter configuration, and the fusion parameter configuration includes a relationship calculation configuration and a field supplement configuration;

[0136] The vocabulary fusion module 20 is used to determine the source vocabulary and target vocabulary of this fusion from multiple vocabularies; if the source vocabulary and the target vocabulary are of the same type, the synonym objects of the source vocabulary and the target vocabulary are determined according to the relationship calculation configuration, and the field content of the source vocabulary under the synonym objects and related to the field supplement configuration is supplemented into the target vocabulary; the supplemented target vocabulary is added to the multiple vocabularies as the fused vocabulary obtained in this fusion, and the step of determining the source vocabulary and target vocabulary of this fusion from the multiple vocabularies is returned until all the multiple vocabularies are fused and the fused vocabulary obtained by the last fusion is output.

[0137] In a possible implementation, the fusion parameter configuration further includes a vocabulary priority configuration, which is used to determine the vocabulary fusion module 20 of the source vocabulary and the target vocabulary of this fusion from multiple vocabulary tables, and is specifically used to:

[0138] Based on the vocabulary priority configuration, the priority of each vocabulary among the multiple vocabulary is determined; the fusion direction of the multiple vocabulary is determined according to the priority of each vocabulary, and the source vocabulary and the target vocabulary are determined according to the fusion direction.

[0139] In a possible implementation, the source vocabulary and the target vocabulary are both descriptor structure vocabulary, the synonym object is a descriptor, and the fusion parameter configuration also includes a matching range configuration and a descriptor fusion range configuration. The matching range configuration is a concept mode.

[0140] The vocabulary fusion module 20 is used to supplement the field contents of the source vocabulary under the synonymous object and related to the field supplement configuration into the target vocabulary, specifically used to:

[0141] Determine concept matching pairs under synonymous objects between a source vocabulary and a target vocabulary, wherein the concept matching pairs include a first concept in the source vocabulary and a second concept in the target vocabulary; supplement the target vocabulary with terms of the first concept as non-preferred terms of the second concept, add concept attributes of the first concept as concept attributes of the second concept, add other concepts and attributes in the source vocabulary as non-preferred concepts and attributes of the target vocabulary, and add thesaurus attributes of the source vocabulary as thesaurus attributes of the target vocabulary; construct concept relationships within synonymous objects in the supplemented target vocabulary; and inherit thesaurus relationships of the source vocabulary under synonymous objects in the supplemented target vocabulary based on thesaurus fusion scope configuration.

[0142] In a possible implementation, the source vocabulary and the target vocabulary are both conceptual vocabulary, the synonym objects are concepts, and the fusion parameter configuration further includes a matching range configuration and a concept fusion range configuration. The matching range configuration is a concept mode, which is used to supplement the field content of the source vocabulary under the synonym object and related to the field supplement configuration to the vocabulary fusion module 20 in the target vocabulary, specifically for:

[0143] The target vocabulary is supplemented by taking the preferred terms and non-preferred terms of the source vocabulary under synonymous objects as the non-preferred terms of the target vocabulary under synonymous objects, and taking the conceptual attributes of the source vocabulary under synonymous objects as the conceptual attributes of the target vocabulary under synonymous objects; based on the configuration of the concept fusion scope, the conceptual relationship of the source vocabulary under synonymous objects is inherited in the supplemented target vocabulary.

[0144] In a possible implementation, the type of the source vocabulary is a descriptive structure vocabulary, the type of the target vocabulary is a concept structure vocabulary, the fusion parameter configuration further includes a matching range configuration, the matching range configuration is a concept mode or a descriptive model, and the vocabulary fusion module 20 for converting the source vocabulary according to the type of the target vocabulary is specifically used to:

[0145] When the matching scope is configured as the concept mode, the descriptive relationships in the source vocabulary and the conceptual relationships between non-preferred concepts and preferred concepts in each descriptor in the source vocabulary are obtained; the type of the source vocabulary is converted into a concept structure vocabulary based on the acquired descriptive relationships and conceptual relationships; when the matching scope is configured as the descriptive mode, the non-preferred concepts in each descriptor in the source vocabulary are merged into their preferred concepts, so as to merge all concepts in each descriptor in the source vocabulary into one concept.

[0146] In a possible implementation, the type of the source vocabulary is a concept structure vocabulary, the type of the target vocabulary is a descriptive structure vocabulary, the fusion parameter configuration further includes a matching range configuration, the matching range configuration is a concept mode, and the vocabulary fusion module 20 for converting the source vocabulary according to the type of the target vocabulary is specifically used to:

[0147] Establish the descriptors corresponding to each concept in the source thesaurus, take each concept as the preferred concept of the descriptor, and take the conceptual attributes of each concept as the descriptor attributes of the descriptor and the attributes of its preferred concept; obtain the concept relationship in the source thesaurus, and convert the obtained concept relationship into a descriptor relationship, so as to convert the type of the source thesaurus into a descriptor structure thesaurus.

[0148] In a possible implementation, the fusion parameter configuration further includes a relationship conflict configuration, which is used to add the supplemented target vocabulary as the fusion vocabulary obtained by this fusion to the vocabulary fusion module 20 in the multiple vocabulary, specifically for:

[0149] The target vocabulary after supplementation is conflict-processed based on the relationship conflict configuration; the target vocabulary after conflict processing is added to the multiple vocabulary as the fusion vocabulary obtained in this fusion.

[0150] It should be noted that the detailed functions of each module in the embodiment of the present application can be found in the corresponding public part of the multi-scenario fusion method embodiment of the above-mentioned medical knowledge organization system, and will not be repeated here.

[0151] The present application also provides an electronic device in an embodiment. Figure 8 , Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device in the embodiment of the present application may include but is not limited to fixed terminals such as mobile phones, laptops, PDAs (personal digital assistants), PADs (tablet computers), desktop computers, etc. Figure 8 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0152] like Figure 8 As shown, the electronic device may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 801, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 802 or a program loaded from a storage device 808 to a random access memory (RAM) 803. When the electronic device is powered on, various programs and data required for the operation of the electronic device are also stored in the RAM 803. The processing device 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.

[0153] Typically, the following devices may be connected to the I / O interface 805: input devices 806 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 807 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 808 including, for example, a memory card, a hard disk, etc.; and communication devices 809. The communication device 809 may allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although Figure 8 An electronic device having various devices is shown, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead.

[0154] Also provided in an embodiment of the present application is a computer program product including computer-readable instructions. When the computer-readable instructions are executed on an electronic device, the electronic device implements a multi-scenario fusion method of any medical knowledge organization system provided in an embodiment of the present application.

[0155] A computer-readable storage medium is also provided in an embodiment of the present application. The storage medium carries one or more computer programs. When one or more computer programs are executed by an electronic device, the electronic device can implement any multi-scenario fusion method of a medical knowledge organization system provided in an embodiment of the present application.

[0156] It should also be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present application, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines.

[0157] Through the description of the above implementation mode, the technicians in the field can clearly understand that the present application can be implemented by means of software plus necessary general hardware, and of course, it can also be implemented by special hardware including special integrated circuits, special CPUs, special memories, special components, etc. In general, all functions completed by computer programs can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be various, such as analog circuits, digital circuits or special circuits. However, for the present application, software program implementation is a better implementation mode in more cases. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a readable storage medium, such as a computer floppy disk, a U disk, a mobile hard disk, a ROM, a RAM, a disk or an optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, a training device, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0158] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0159] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a training device, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, training device, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device, a data center, etc. that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

Claims

1. A multi-scenario fusion method for a medical knowledge organization system, characterized in that: The multi-scenario fusion method of the medical knowledge organization system includes: receiving a vocabulary fusion task, wherein the vocabulary fusion task includes multiple vocabularies to be fused and a fusion parameter configuration, wherein the fusion parameter configuration includes a relationship calculation configuration and a field supplement configuration; Determine a source vocabulary and a target vocabulary for this fusion from the multiple vocabulary lists; If the source vocabulary and the target vocabulary are of the same type, determine the synonym objects of the source vocabulary and the target vocabulary according to the relationship calculation configuration, and add the field content of the source vocabulary under the synonym objects and related to the field supplement configuration to the target vocabulary; add the supplemented target vocabulary as the fused vocabulary obtained by this fusion to the multiple vocabularies, and return to execute the step of determining the source vocabulary and the target vocabulary for this fusion from the multiple vocabularies, until all the multiple vocabularies are fused and the fused vocabulary obtained by the last fusion is output; If the source vocabulary and the target vocabulary are of different types, the source vocabulary is converted according to the type of the target vocabulary, and the step of determining synonyms of the source vocabulary and the target vocabulary according to the relationship calculation configuration is returned to execute.

2. The multi-scenario fusion method of the medical knowledge organization system according to claim 1 is characterized in that: The fusion parameter configuration also includes a vocabulary priority configuration, and the determining of the source vocabulary and the target vocabulary for this fusion from the multiple vocabulary includes: Determining the priority of each vocabulary in the plurality of vocabulary based on the vocabulary priority configuration; The fusion direction of the multiple vocabularies is determined according to the priority of each vocabularies, and the source vocabularies and the target vocabularies are determined according to the fusion direction.

3. The multi-scenario fusion method of the medical knowledge organization system according to claim 1 is characterized in that: The source vocabulary and the target vocabulary are both descriptive structure vocabulary, the synonym object is a descriptive word, the fusion parameter configuration further includes a matching range configuration and a descriptive fusion range configuration, the matching range configuration is a concept mode, and the field content of the source vocabulary under the synonym object and related to the field supplement configuration is supplemented to the target vocabulary, including: Determine a concept matching pair between the source vocabulary and the target vocabulary under the synonym object, wherein the concept matching pair includes a first concept in the source vocabulary and a second concept in the target vocabulary; Supplement the target vocabulary with the terms of the first concept as non-preferred terms of the second concept, add the concept attributes of the first concept as the concept attributes of the second concept, add other concepts and attributes in the source vocabulary as non-preferred concepts and attributes of the target vocabulary, and add the descriptive attributes of the source vocabulary as the descriptive attributes of the target vocabulary; constructing conceptual relationships within the synonymous objects in the supplemented target vocabulary; Based on the configuration of the thesaurus fusion scope, the thesaurus relationship of the source thesaurus under the synonymous object is inherited in the supplemented target thesaurus.

4. The multi-scenario fusion method of the medical knowledge organization system according to claim 1 is characterized in that: The source vocabulary and the target vocabulary are both concept vocabulary, the synonym object is a concept, the fusion parameter configuration further includes a matching range configuration and a concept fusion range configuration, the matching range configuration is a concept mode, and the field content of the source vocabulary under the synonym object and related to the field supplement configuration is supplemented to the target vocabulary, including: The target vocabulary is supplemented by using the preferred terms and non-preferred terms of the source vocabulary under the synonymous object as the non-preferred terms of the target vocabulary under the synonymous object, and the concept attributes of the source vocabulary under the synonymous object as the concept attributes of the target vocabulary under the synonymous object; Based on the concept fusion scope configuration, the concept relationship of the source vocabulary under the synonymous object is inherited in the supplemented target vocabulary.

5. The multi-scenario fusion method of the medical knowledge organization system according to claim 1 is characterized in that: The type of the source vocabulary is a descriptive structure vocabulary, the type of the target vocabulary is a concept structure vocabulary, the fusion parameter configuration further includes a matching range configuration, the matching range configuration is a concept mode or a descriptive mode, and the type conversion of the source vocabulary according to the type of the target vocabulary includes: When the matching scope is configured as a concept mode, acquiring the descriptor relationships in the source word list and the conceptual relationships between non-preferred concepts and preferred concepts in each descriptor in the source word list; converting the type of the source word list into a concept structure word list according to the acquired descriptor relationships and conceptual relationships; In the case where the matching scope is configured as a descriptor mode, non-preferred concepts in each descriptor in the source thesaurus are merged into its preferred concepts, so as to merge all concepts in each descriptor in the source thesaurus into one concept.

6. The multi-scenario fusion method of the medical knowledge organization system according to claim 1 is characterized in that: The type of the source vocabulary is a concept structure vocabulary, the type of the target vocabulary is a descriptive structure vocabulary, the fusion parameter configuration further includes a matching range configuration, the matching range configuration is a concept mode, and the type conversion of the source vocabulary according to the type of the target vocabulary includes: Establishing the descriptors corresponding to the concepts in the source word list, taking each concept as the preferred concept of the descriptor, and taking the concept attribute of each concept as the descriptor attribute of the descriptor and the attribute of its preferred concept; The concept relationships in the source vocabulary are obtained, and the obtained concept relationships are converted into descriptive thesaurus relationships, so as to convert the type of the source vocabulary into a descriptive structure vocabulary.

7. The multi-scenario fusion method of the medical knowledge organization system according to claim 1 is characterized in that: The fusion parameter configuration also includes a relationship conflict configuration, and the supplemented target vocabulary is added to the multiple vocabulary as the fusion vocabulary obtained by this fusion, including: Performing conflict processing on the supplemented target vocabulary based on the relationship conflict configuration; The target vocabulary after conflict resolution is added to the multiple vocabulary as the fused vocabulary obtained in this fusion.

8. A vocabulary fusion device for a medical knowledge organization system, characterized in that: The vocabulary fusion device of the medical knowledge organization system includes: A task receiving module, used for receiving a vocabulary fusion task, wherein the vocabulary fusion task includes multiple vocabularies to be fused and a fusion parameter configuration, wherein the fusion parameter configuration includes a relationship calculation configuration and a field supplement configuration; A vocabulary fusion module is used to determine the source vocabulary and target vocabulary of this fusion from the multiple vocabularies; if the source vocabulary is of the same type as the target vocabulary, determine the synonym objects of the source vocabulary and the target vocabulary according to the relationship calculation configuration, and supplement the field content of the source vocabulary under the synonym objects and related to the field supplement configuration into the target vocabulary; add the supplemented target vocabulary as the fusion vocabulary obtained in this fusion to the multiple vocabularies, and return to execute the step of determining the source vocabulary and target vocabulary of this fusion from the multiple vocabularies, until all the multiple vocabularies are fused and the fusion vocabulary obtained by the last fusion is output; if the source vocabulary is of different types from the target vocabulary, perform type conversion on the source vocabulary according to the type of the target vocabulary, and return to execute the step of determining the synonym objects of the source vocabulary and the target vocabulary according to the relationship calculation configuration.

9. An electronic device, characterized in that: The method comprises at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs; The processor is used to execute the computer program so that the electronic device can implement the multi-scenario fusion method of the medical knowledge organization system as described in any one of claims 1 to 7.

10. A computer program product, characterized in that It includes computer-readable instructions, which, when executed on an electronic device, enable the electronic device to implement the multi-scenario fusion method of the medical knowledge organization system as described in any one of claims 1 to 7.

Images