KR20170135331A - A data exchange semantic interoperability processing system for health measuring equipments and the method thereof - Google Patents

Abstract

The present invention relates to a system and method for processing semantic interoperability for data exchange between heterogeneous health measurement instruments, which builds one scalable common terminology system and connects terms in a plurality of medical terminology systems to terms in the common terminology system, thereby making all the medical terminology systems interoperable. The system comprises: a reference terminology constructing unit receiving a plurality of reference terminologies and relationships thereof to build a reference terminology system; a common terminology constructing unit receiving common terminology and relationships thereof to build a common terminology system based on ontology; a term mapping unit mapping terms in the reference terminology system to terms in the common terminology system and setting a term relationship for mapping between the terms; and a term deriving unit receiving one query term in the reference terminology system and searching for terms in another reference terminology system corresponding to the query term, wherein the term deriving unit traces terms having a direct or indirect relationship with the query term to discover a final search target term. According to the system and method for processing semantic interoperability for data exchange between heterogeneous health measurement instruments, different terminology systems are linked with each other by mapping each of the reference terminology systems or standard terminology systems to one common terminology system. Accordingly, it is possible to reuse terms in the different terminology systems and guarantee interoperability for medical information exchange.

Description

Technical Field [0001] The present invention relates to a data exchange semantic interoperability processing system,

The present invention relates to a heterogeneous health measurement device which constitutes one scalable common term system and connects each term of a plurality of medical terminology systems to terms of the common term system so that all medical terminology systems can be interoperable Data exchange semantic interoperability processing system and method.

In particular, the present invention establishes a common terminology system on the basis of an ontology, establishes an ontology relationship among concepts of the terms, defines a property value by a plurality of qualifiers of one term concept, The present invention relates to a heterogeneous health measurement device, a data exchange semantic interoperability processing system and method, which infer a relationship between terms using an ontology-based reasoning, even though they do not have a direct relationship.

Clinical documents used in the medical field generally include patient's medical records, patient's symptoms, doctor's diagnosis, prescription, etc. At this time, the medical terminology or medical data used for expressing the medical record in the clinical document is expressed differently for each medical staff or medical institution.

Such medical terminology or medical data is data that requires very fast and accurate handling in its use and should be able to ensure semantic and functional interoperability with multiple health care providers and support areas beyond the use of a single medical staff. In order to maintain these characteristics, a medical terminology should be constructed so that the information in the clinical document can be acquired more easily, collected, transmitted and processed.

As the terminology system based on the existing data model is not enough to accomplish the above-mentioned purpose, a supporting tool for structured data input is needed, and furthermore, in order to support the sharing of clinical decision support system and electronic health record The development of a terminology system is essential.

First, the concept of medical terminology to be described in a clinical document must be clarified. This is because the use of various terms for the same concept among health professionals causes problems in smooth information communication.

To this end, a terminology system has been developed that includes a set of medical concepts used in a specific medical field, definitions of concepts, terms used to refer to concepts, and relationships between concepts. According to their purpose and characteristics, Reference terminology, and interface terminology.

Reference terminology is a set of formalized representations of medical concepts and inter-relationships, such as SNOMED CT, RxNorm, NCI Thesaurus, and ICNP.

Even though there is a reference term system, which is a set of formal expressions for medical concepts, each local term used by various medical staff in hospitals is different, so mapping work is needed to make the difference. In particular, interface terminology should be based on reference terminology, since it must interface with medical staff and reference terminology.

In order to solve the above problems, there has been proposed a reference term-based term system authoring support system for constructing a medical terminology used in a medical institution using a reference term system in a structured term system in relation to an individual [Patent Literatures 1 and 2 , 3]. In this way, it is possible to construct interface terms in the medical field to communicate more smoothly between doctors or medical staffs, but the terminology of the prior arts only supports the systematic structuring of the concepts of terms more accurately.

However, in order to clearly understand medical data (or medical records), it is important to understand how the concepts are being used, even if the concepts of medical terminology are systematized. For example, if the blood pressure of a clinical document is determined to be arterial blood pressure, systolic blood pressure, or diastolic blood pressure, it is necessary to describe in what state the blood pressure is the measured data. Further, in order to more clearly describe the contents of the patient's symptoms or diagnosis, the severity and location of the patient should be described in more detail.

That is, one medical data must know not only the medical term (or concept) but also the situation information in which the data was created. However, even in the same medical terminology, the situation information can vary greatly depending on the medical staff or the medical institution. In the primary hospital, only the systolic blood pressure and diastolic blood pressure should be described by measuring the blood pressure. However, the tertiary hospital should be able to describe in detail the measurement means, the measurement site and the measurement posture as well as the systolic blood pressure and diastolic blood pressure have. This is because there may be slight differences in the measurement results depending on the measurement situation.

Therefore, Interface Terminology proposed by medical experts and clinical support experts to acquire, collect, transmit and process information in clinical documents more easily should be able to handle such situation information, In order to be able to easily input this situation information, support tools for structured data entry should be provided.

In addition, there are a number of reference term systems mixed together, and the term systems used in each local medical institution are also different and mixed. Therefore, they should be compatible and interoperable in different medical terminology systems as a whole.

Korean Patent Laid-Open No. 10-2011-0110683 (published on October 10, 2011) Korean Patent Laid-open No. 10-2011-0119230 (published on November 2, 2011) Korean Published Patent Application No. 10-2011-0119231 (published on November 2, 2011)

It is an object of the present invention to solve one or more of the problems as described above, and to provide a system and method for constructing a scalable common term system, connecting each term of a plurality of medical terminology systems to terms of the common term system, And to provide a heterogeneous health measuring instrument data exchange semantic interoperability processing system and method that are configured to enable interoperability of the interoperability.

The object of the present invention is to construct a common term system on the basis of an ontology, to establish an ontology relation between concepts of the terms, to define an attribute value with a plurality of qualifiers, And to provide a heterogeneous health measurement device data exchange semantic interoperability processing system and method that deduce values and define inferences about terms using ontology-based reasoning even if they do not have a direct relationship.

In order to achieve the above object, the present invention relates to a heterogeneous health measuring instrument data exchange semantic interoperability processing system, comprising: a reference term constituent unit configured to receive a plurality of reference terms and their relations and construct a reference term system; A common term construction unit for receiving a common term and its relations and constructing a common term system based on the ontology; A term mapping unit for mapping the terms in the reference term system to terms in the common term system and setting a term relation for the mapping between the terms; And a search unit for searching for terms in another reference term system corresponding to the query term by receiving one query term in the reference term system and tracing terms having a direct or indirect relationship with the query term to search for a final search term And a terminology reasoning unit.

In the data interchange semantic interoperability processing system according to the present invention, the common term constituent unit searches for a data object defined by a concept related to the corresponding concept (hereinafter referred to as a related concept) , The related concept is characterized by having a term relationship with the corresponding concept.

As described above, according to the heterogeneous health measurement device data exchange semantic interoperability processing system and method according to the present invention, it is possible to map each reference term system or standard term system into one common term system, It is possible to reuse the terms of other term systems with each other or to ensure interoperability for exchanging medical information.

In addition, according to the heterogeneous health measurement device data exchange semantic interoperability processing system and method according to the present invention, by recommending the qualifiers of the data entity to be newly defined by utilizing the structure information of the already defined data entity, An effect that can be more conveniently defined can be obtained.

1 is a diagram showing a configuration of an overall system for carrying out the present invention.
2 is a block diagram of a configuration of a heterogeneous health measuring instrument data exchange semantic interoperability processing system according to an embodiment of the present invention.
FIGS. 3 and 4 are views showing the structure and structure of a terminology system according to an embodiment of the present invention.
5 is a diagram illustrating the structure and structure of a data entity system according to an embodiment of the present invention.
FIG. 6 is a diagram illustrating an example of defining a concept and a qualifier of a data entity according to an embodiment of the present invention.
FIG. 7 is an exemplary diagram illustrating a relationship between a prebuilt term system and a previously created and stored data entity according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail with reference to the drawings.

In the description of the present invention, the same parts are denoted by the same reference numerals, and repetitive description thereof will be omitted.

First, examples of the configuration of the entire system for carrying out the present invention will be described with reference to Fig.

As shown in FIG. 1A or 1B, the data entity definition support system according to the present invention can be implemented as a server system on a network or a program system on a computer terminal.

As shown in FIG. 1A, an example of an overall system for implementing the present invention comprises a user terminal 10 and a data entity definition supporting system 30 and is connected to a network 20 with each other. It is also possible to further include a database 40 for storing necessary data.

The user terminal 10 is a general computing terminal such as a PC, a notebook, a netbook, a PDA, a mobile, etc., which is used by a user such as a medical person or a person in charge of a medical institution. The user uses the user terminal 10 to transmit the medical term or medical data used in his medical institution to the definition support system 30 or requests the definition support system 30 to define the medical data.

The data object definition support system 30 can be connected to the network 20 as a normal server to provide a definition tool for defining medical data or to display a medical term concept for defining medical data, It provides a service that allows the data qualifiers to be displayed and referenced.

On the other hand, the definition support system 30 can be implemented as a web server or a web application server that provides the above services as web pages on the Internet.

The database 40 is a conventional storage medium for storing the necessary data in the definition support system 30, and stores terminology of medical terminology, medical data entities created by the user.

As shown in FIG. 1B, another example of the entire system for implementing the present invention comprises a data entity definition support system 30 in the form of a program installed in the computer terminal 13. That is, each function of the definition support system 30 is implemented as a computer program and installed in the computer terminal 13, so that the concept (or term concept) and qualifier for defining a new data entity by the user 14, The medical term terminology system and the already defined data entity information through the output device of the computer terminal 13 or inputs it through the input device of the terminal 13 or outputs it. Meanwhile, necessary data in the definition support system 30 is stored in a storage space such as a hard disk of the computer terminal 13 and used.

Next, a configuration of a heterogeneous health measurement device data exchange semantic interoperability processing system according to an embodiment of the present invention will be described with reference to FIG.

2, the heterogeneous health measurement device data exchange semantic interoperability processing system 30 according to an embodiment of the present invention includes a terminology management unit 31, a data object storage unit 32, (33), and a recommender recommendation section (34). In addition, it may further comprise a database (or storage space) 40 for storing data.

The term system management unit 31 stores a term system structured in terms of entities and relationships.

Narrow terminology refers to a system of terminology concepts that establish medical terminology as a concept, include the same terms in a concept, establish relationships between concepts, and organize them. Terminology in a broad sense is a systematic organization of medical terminology used in the medical field. In addition to systematic medical terminology (medical concept), terminology includes medical contexts that include medical contexts, data element). In the following, the term system is used as a narrow term system, but when it is necessary to distinguish it, a narrow term system is described as a term concept system and a broad term system is described as a data object or a data entity system do.

The term system (or term concept system) uses a reference term system or a local term system (or interface term system) organized by a medical staff or a medical institution. The reference term system is based on the medical concept or the medical term, and expresses the inter-relationships and attributes in a formal way. For example, SNOMED CT, RxNorm, NCI Thesaurus, and ICNP.

Terminology is a repository of relevant information about concepts, such as terms or terms based on concepts. A display term may be a "preferred term" or it may be a synonym for an already registered representative term, or an abbreviation.

Terminology can be represented by concepts and associations, each of which can be structured as an entity and a relation. Since an object and a concept, an association and a relation are corresponding concepts, they are used in the following. Further, the entity (or concept) includes at least one display term attribute, wherein one of the display term properties may be set as a representative term. Representative terms may also be used as names of objects.

On the other hand, the terminology system can be schematized as entity 61 and relationship 62 as shown in FIG. In FIG. 3, the entities 61 are C1, C2, ..., C6, and the display terms T11, T21, ..., T61. The display term of object C4 has T41, T42, T43, T44. Among them, the display term T42 is a representative term.

The relationship 62 is represented by R12, R23, ..., R56 connecting the entities. The relationship R12 is the relationship between the object C1 and the object C2, and the relationship R35 is the relationship between the object C3 and the object C5.

Relationships are largely divided into term relationships and mapping relations. The term relationship includes a hierarchical relationship that defines the relationship between the parent and the child for two individuals, and a domain relationship that defines the inclusion relationship between the two individuals. A mapping relationship (or reference relationship) refers to a relationship in which two objects correspond to each other in the same concept. In FIG. 3, the objects C1 and C2 have a vertical relationship (or a term relationship) with each other. Objects C3 and C5 have a mapping relationship with each other. The mapping relationship is also described in a single terminology. That is, we use the broader term relation as a meaning that includes both the mapping relation and the narrower term relation.

For example, as shown in FIG. 4, the concept of blood pressure can be expressed in relation to an individual.

That is, the blood pressure is classified into arterial blood pressure, capillary blood pressure, and venous blood pressure according to the name of a blood vessel. Normally, blood pressure refers to arterial blood pressure. Arterial blood pressure varies by heartbeat.

Arterial blood pressure is classified into systolic pressure, diastolic pressure, pulse pressure, mean arterial pressure, and the like. Systolic blood pressure refers to the maximum pressure in the blood vessel when the blood in the deep room is released into the aorta by contraction of the myocardium. The diastolic blood pressure refers to the pressure in the blood vessel when the heart contracts and then expands.

The pulse pressure means the difference between the systolic blood pressure and the diastolic blood pressure, and the mean arterial pressure means the mean arterial blood pressure and the lowest arterial blood pressure. In particular, mean arterial pressure can be obtained from systolic blood pressure and diastolic blood pressure by a formula, but it must be obtained through a computerized blood pressure measuring device strictly.

Next, the data object storage unit 32 stores a data element for the concept.

As shown in FIG. 5, a data element includes a concept, a term, a qualifier, a modifier, and an assertion. In addition, a management element (not shown) for managing permission information, version information, history information, and the like may be further included to manage the term system.

In this case, a qualifier, a modifier, and an assertion are elements representing context information of a concept.

Qualifiers are generally used to limit the content of diagnosis / surgery. That is, a status post a procedure, a history of a condition, and the like may be qualifiers.

A modifier is generally used to describe the content of a medical diagnosis or the like, and to provide a detailed representation of its weight, location, and the like. Values indicating severity include mild, moderate, and severe. Stage I, Stage II, etc. can be used to describe the stage of the illness.

Assertion (Assertion or Assertional Knowledge) is a concept proposed for clarification of terminology. Information provided in addition to the definition of the term concept provides information that is helpful for the clarification of concepts such as concepts, synonyms, and common modifiers related by the relevant clinical users. For example, Chest Pain is not commonly seen in healthy people, and severity is expressed as 'mild', 'moderate', or 'severe'. In addition, it can be expressed as assertional knowledge as the difference between Thorax pain and Chest Pain. Both can be defined as the same concept through the conceptual is-a relationship, but thorax pain means that the position of the pain is more intense to the medical staff, and chest pain can mean the possibility of cardiac / pulmonary disease.

The qualifiers, modifiers, and assertions described above are all contexts that define a concept. Hereinafter, all of the context information elements will be described as qualifiers.

As shown in FIG. 6, the data element of the blood pressure can be configured with a concept (or a term concept) and a qualifier.

Generally, in a primary hospital (a private hospital), doctors use a sphygmomanometer to measure the blood pressure by pressing the upper abdomen of the patient, and only the systolic blood pressure and diastolic blood pressure values are described in the clinical document. Thus, the data entity for this hospital's blood pressure can be represented as data entity "Hospital A. arterial blood pressure" in FIG. That is, it is expressed by the measurement method, the systolic blood pressure value, and the diastolic blood pressure value.

However, when measuring the patient's blood pressure, there may be slight differences depending on whether the measurement site is on the right side or on the left side. There is a difference in the measurement result depending on whether the measurement state is an upright state, a sitting state, or a lying state.

In the first hospital, such minute difference may not be so important, but the above-mentioned minute difference may be very important depending on the state of the patient in the third hospital. In such a case, the data object for the blood pressure of this hospital can be represented as the data entity "Hospital B. arterial blood pressure (1)" in FIG. That is, it comprises the qualification of the measurement posture and the measurement position as compared with the data entity "Hospital A. arterial blood pressure ".

A value set is a set of values that a qualifier can take. The value set of the measurement posture can have values of standing (standing), sitting (sitting), and lying (lying) measurements. In addition, the measurement site may have values such as upper left, upper right, left wrist, and right wrist.

On the other hand, in the tertiary hospital, the pressure in the arterial blood vessel directly is measured rather than the pressure method described above. This direct method is useful in cases where it is difficult to measure arterial blood pressure by indirect methods due to shock or hemorrhage, when the cardiovascular system is very unstable and continuous monitoring of arterial blood pressure is required, or blood gas analysis and other blood tests It is often used in cases where it is necessary to collect arterial blood frequently. In this case, mean arterial pressure may be measured to observe sustained blood pressure.

Therefore, the data entity for this purpose can be represented as the data entity "Hospital B. arterial blood pressure (2)" in FIG. In other words, compared to the data entity "Hospital B. arterial blood pressure (1)", it has modifiers such as measuring device, insertion position, mean arterial pressure instead of the qualification of the measurement posture and measurement position. The insertion position is the position of the artery into which the catheter is inserted and may have values such as radial, brachial, axillary, dorsalis pedis, femoral, and the like.

In the above example, arterial blood pressure includes systolic blood pressure and diastolic blood pressure as its qualifiers. Therefore, it can be seen that the qualifier is also a concept. That is, the measurement site used in the qualifier, the mean arterial pressure, and the like can all be defined as concepts and can be concepts included in the term system.

Next, the data entity generation unit 33 defines a data entity for a selected concept (hereinafter referred to as a concept), receives a selection of the corresponding concept, and receives and defines a qualifier for defining the concept.

As we have seen, a data object consists of a concept and a qualifier. Preferably, the concept of the data entity is selected from the terminology. That is, the medical staff selects the concept of the blood pressure desired by the medical staff from the terminology system in order to generate the data object for the blood pressure.

 For example, "arterial blood pressure" is selected in the terminology system as shown in FIG. Since blood pressure generally refers to arterial blood pressure, most physicians in the past have described only "blood pressure" in the clinical document. That is, "blood pressure 80 to 120" is described. Therefore, medical terminology can be entered more accurately into clinical documentation by choosing concepts in the terminology framework.

The data entity generating unit 33 defines a data element by adding a qualifier to the selected concept. At this time, the qualifier recommending unit 34 recommends the appropriate qualifier to the selected concept.

That is, the qualifier recommendation unit 34 searches for a stored data entity and recommends a qualifier of the retrieved data entity in the same concept as the selected concept (hereinafter referred to as the same concept). In particular, the qualifier recommendation unit 34 searches for a data entity defined by a concept related to the concept (hereinafter referred to as a related concept), and recommends a qualifier of the retrieved data entity, and the related concept has a term relationship with the corresponding concept .

The recommender recommender 34 recommends qualifiers of data entities defined by the same concept or the related concept by priorities, and recommends priorities based on the class weights assigned to the data entities.

The class weight of the data entity is determined by the class of the defining subject that defines the data entity. The class weight is assigned according to the influence of the subject in the medical field.

Data entities defined in a 3-D hospital are more influential in a medical field than data entities defined in a primary hospital. Also, even if it is the same primary hospital, the hospital has more influence than the hospital. Also, even within the same hospital, the influence is different depending on the medical staff (doctors, nurses, medical personnel, administrators, etc.) who define data objects. For example, doctors, chief nurses, general nurses, etc. have different influences. Therefore, we classify each user or hospital according to the influence as above, and set the weight differently.

In one embodiment, the class weights are determined by various variables such as the number of beds, the degree of the hospital (1, 2, 3), the position of the subject, and specialty.

Preferably, if the rating weight of the defining subject is 3, the rating weight of the data object defined by the defining subject is all 3. As another embodiment, it is preferable that the class weights of the data entity can be set separately even if the same subject is defined. For example, an administrator may specify a particular weighting value for a particular data object.

A method for recommending a recommendation by the recommender 34 based on the weight according to the first embodiment of the present invention will be described first.

The qualifier recommendation unit 34 recommends qualifiers of the data entities defined by the same concept or the related concept in order of priority and increases the priority in descending order of the number of the concepts. In particular, the qualifier recommending unit 34 preferentially recommends a qualifier having a larger sum of weights among the qualifiers having the same priority by the same degree, but the sum of the weights of the respective qualifiers (hereinafter, the qualifiers) Is calculated as the sum of the class weights.

That is, the class weight sum q _i (C _k ) of the qualifiers in the concept C _k is obtained by the following equation (1).

[Equation 1]

However, q _i (C _k) is a concept and rating a weighted sum of the modifier in the ^{_{C k, W (R k M}} ) is the sum of the same concepts or related concepts the relationship of terms related to the C _k weights in the concept C _M . Also, R (C _M ) is a set of term relationships between the concept C _M and the corresponding concept C _M or the same concept or related concept.

Referring to Fig. 7, an example in which the qualifier recommendation unit 34 recommends a qualifier according to an embodiment of the present invention will be described. FIG. 7 is a diagram showing already constructed terminology and data entities already defined and generated.

In FIG. 7, the data object is displayed on the left side and the terminology system is displayed on the right side. D1, D2, ..., D7 indicate data entities, and the numbers in parentheses (()) next to data entity identification information D1, D2, ..., D7 mean class weights. Q1, Q2, ..., Q6 are the names of the qualifiers. Also, C1, C2, ..., C7 mean the concept in the terminology system. A straight line connected between concepts in the terminology system indicates a relationship, and a dotted line connected between the terminology system and the data object indicates the concept of the concept object corresponding to the concept of the data object.

The data entity D1 defines a data entity for the concept C1 and includes Q1 and Q2 as its qualifiers. The data entity D2 defines a data entity for the concept C1 and includes Q1, Q2 and Q3 as its qualifiers. In addition, it can be seen that there are three data entities for concept C2: D2, D3, and D4. It is also possible to deduce that the concept of the data entity D2 has a lower relation with the concept of the data entity D1.

In the state shown in Fig. 7, a case of recommending a qualifier will be described separately.

In the first case, a data object corresponding to concept C2 is generated. The data objects defined and stored in the same concept as concept C2 are D2, D3, and D4. The qualifiers of these entities are Q1, Q2, Q3, Q4. We therefore recommend these qualifiers. Also, since the related concepts of concept C2 correspond to all of C1, C3 to C7, the data entities of the related concept are D1, D5, D6, D7. The qualifiers of these entities are Q1, Q2, Q4, Q5, Q6.

The qualifiers with the highest priority are the qualifiers Q1, Q2, Q3 and Q4 of the data objects D2, D3 and D4 defined with the same concept. Since the data entities including Q1 are D2, D3 and D4, the weight sum of Q1 is 4 (D2) + 3 (D3) + 1 (D4) = 8.

In other words, when expressed by the formula (1), the class weight q ₁ (C ₂ ) of the modifier Q 1 in the concept C ₂ is q ₁ (C ₂ ) = 4 × 1 + 3 × 1 + 1 × 1 = Q2, Q3, the q _i (D _j) are all 1 because both Q4 are all included in the data object.

In the same way, the sum of the weights of the qualifiers Q2, Q3 and Q4 is as follows.

Q2: 4x0 (D2) + 3x1 (D3) + 1x1 (D4) = 4

(Since D2 does not include Q2, it is 0.)

Q3: 4x1 (D2) + 3x0 (D3) + 1x0 (D4) = 4

Q4: 4x0 (D2) + 3x0 (D3) + 1x1 (D4) = 1

Thus, Q1 is the highest priority, followed by Q2 and Q3, and then Q4. Although the qualifier Q3 appears once and Q2 appears twice, the grade weights of D2 are high, so Q2 and Q3 are recommended in the same order.

Next, except Q1 to Q4 among the qualifiers of the related concepts, Q5 and Q6, all of which form a three-dimensional relationship. Therefore, Q5 and Q6 are recommended in the following order, but Q5 has a weighted sum of 5 and Q6 has a weight of 1, so we recommend Q5 first.

The number of syllables indicates how many levels of relationships are involved when they have a direct or indirect relationship between concepts in the terminology. That is, C2 has a one-village relationship with C1, a two-village relationship with C3 to C5, and a three-village relationship with C6 and C7.

On the other hand, if the number of siblings with the concept exceeds the minimum number of siblings (minimum recommended siblings), it may not be recommended. For example, if you are limited to 8 villages, it is not recommended anymore if you go beyond 8 villages.

Next, the second case is a case of generating a data object corresponding to concept C5. There are no data objects defined and stored in the same concept as concept C5. The concept of a one-village relationship with concept C5 is C1, and the data object of C1 is D1. Therefore, the weighted sum of the qualifier Q1 of D1 is 1, but the highest priority is recommended.

Next, the qualifiers Q2, Q3 and Q4 of the data objects D2 to D5 of the concepts C2 and C3 having the two-dimensional relationship are recommended (Q1 is not already recommended). The weighted sum of the qualifiers Q2, Q3, and Q4 is as follows.

Q2: 3 (D3) + 1 (D4) = 4

Q3: 4 (D2) = 4

Q4: 1 (D4) + 3 (D5) = 4

Therefore, the priorities of the qualifiers Q2, Q3, and Q4 are all the same. Although modifier Q4 appears twice in data object D4 in concept C2 and in data object D5 in concept C3, it has the same rank as Q3 once it appears. Q3 is because the data object has a high class weight.

Next, except Q1 to Q4 among the qualifiers of the related concepts, Q5 and Q6, all of which form a three-dimensional relationship. Therefore, Q5 and Q6 are recommended in the following order, but since Q5 has a weighted sum of 6 and Q6 has a weighted weight of 3, we prefer Q5 first.

 Next, a description will be given of a method in which the recommender recommendation section 34 recommends priorities based on weights according to the second embodiment of the present invention. The second embodiment of the present invention is the same as the first embodiment, and differs in the following respects.

That is, if at least two data objects having the same concept (hereinafter, the same concept data object) including the corresponding qualifiers are included, the modifier recommendation unit 34 includes only the largest class weight of the same concept data object class, The sum of weights of the qualifiers is obtained.

That is, the class weight sum q _i (C _k ) of the qualifiers in the concept C _k is obtained by the following equation (2).

&Quot; (2) "

The first case of the first embodiment will be described. The weighted sum of the qualifiers Q2, Q3, and Q4 is as follows.

Q1: Max {4x1 (D2), 3x1 (D3), 1x1 (D4)} = 4

Q2: Max {4x0 (D2), 3x1 (D3), 1x1 (D4)} = 3

Q3: Max {4x1 (D2), 3x0 (D3), 1x0 (D4)} = 4

Q4: Max {4x0 (D2), 3x0 (D3), 1x1 (D4)} = 1

Therefore, Q1 and Q2 have the highest priority, followed by Q3 and then Q4. Although qualifier Q3 appears once and Q1 appears three times, Q1 and Q3 are recommended in equal rank because they are all included in the highest weighted D2. The remainder is the same as the first embodiment.

The second case of the first embodiment will be described. That is, in the case of generating the data object corresponding to the concept C5, after the qualifier Q1 is recommended, the qualifiers Q2, Q3 and Q4 of the data objects D2 to D5 of the concepts C2 and C3 having the two- ) Is recommended. The weighted sum of the qualifiers Q2, Q3, and Q4 is as follows.

Q2: Max {3 (D3), 1 (D4)} = 3

Q3: Max {4 (D2)} = 4

Q4: Max {1 (D4)} + Max {3 (D5)} = 4

Therefore, Q3 and Q4 are the highest priority and the next is Q2. Although the class weights of D2 including Q3 are higher than D5, Q4 included in D5 are all included in the data objects D4 and D5 corresponding to concepts C2 and C3, respectively, so that the priorities are the same. The remainder is the same as the first embodiment.

Next, according to the third embodiment of the present invention, a description will be given of a method in which the recommender recommendation unit 34 includes both the class weights and the relation weights, and recommends priorities.

The recommender recommendation section 34 recommends priorities by the relation weights given to the term relationships in addition to the class weights given to the data entity. Preferably, the relationship weights of the term relationships are determined by the type of term relationship.

As we have seen, since the mapping relation is also a term relation, the mapping relation is virtually the same concept, and therefore, the related concept having the mapping relation with the corresponding concept is substantially the same as the same concept. Therefore, it is desirable to recommend the qualifiers of the data entity by the same concept of the data entity and the mapping relation with the same priority.

For example, in FIG. 3, when a qualifier of a data object for the concept C1 is to be recommended, concepts C4 and C5 are both 2 for the concept C1. However, since concepts C3 and C5 are mapping relations, C1 and C5 are actually one-village relations.

Thus, the priority of recommendation depends on the relationship between the concepts. This is given as a relationship weight. Preferably, the higher the relationship weights, the lower the priority. For example, if we set the relationship weight of the parent relationship or the inclusion relation to 1, the relationship weights of the mapping relation should be close to 0, such as 0 or 0.1, and the term relation of the opposite relation is a value much larger than 1, .

Also, the manager can set the relationship weights separately, even if they are of the same kind as the parent relationship. For example, in Fig. 3, the relationship weight of R12 may be set to 1.2, the relation weight of R23 may be set to 0.8, and the relation weight of R35 may be set to 0.3.

The qualifier recommendation unit 34 recommends qualifiers of data entities defined by the same concept or the related concept by priority, and it is proportional to the class weight of the data entity including the same concept or related concept, The priority of which is in inverse proportion to the sum of

The priority of the modifier Qi in the concept C _M is determined by the weighting sum Q _i (C _M ) according to (3). That is, when a data entity corresponding to the concept C _M is to be generated, a weight sum of the qualifiers Q1 to Q6 to be recommended is obtained by the following formula (3), and a recommendation is made from a qualifier having a weighted weight.

&Quot; (3) "

However, q _i (C _k) is a concept and rating a weighted sum of the modifier in the ^{_{C k, W (R k M}} ) is the sum of the same concepts or related concepts the relationship of terms related to the C _k weights in the concept C _M . Also, R (C _M ) is a set of term relationships between the concept C _M and the corresponding concept C _M or the same concept or related concept.

On the other hand, the class weight sum q _i (C _k ) of the modifiers in the concept C _k is obtained by the above-mentioned equation (1) or (2).

In Equation (2), in the first embodiment or the second embodiment, if there are at least two data objects having the same concept (hereinafter, the same concept data object) including the corresponding qualifiers, It is equivalent to obtaining the weighted sum of the qualifiers including only the largest class weight. [Equation 1] is the sum of the weights of all the qualifiers appearing regardless of the same concept data entity.

For example, in FIG. 7, it is assumed that the relationship weight between concept C1 and C2 is 3, and all of the other relational weights are 1s. It is assumed that a qualifier is recommended to generate a data object corresponding to concept C5.

First, the process of obtaining by the equations (1) and (3) will be described.

The sum of the weights of Q1 is obtained as follows.

And equation C _M soon C5, R (C _M) is _{^{R 5 1 (C5-C1)}} , R 5 2 (C5-C1-C2), R 5 3 (C5-C1-C3), R 5 6 ( a set of C5-C1-C4-C6) , R 5 7 (C5-C1-C4-C7). Q1 is included except for C4 among the concepts C1 to C7.

For the relationship _{^{R 5 1 (C5-C1)}} , because it is a relationship of C1-C5, the sum W (R ⁵ ₁₎ of the relationship between the weight is the weight of one of the C1-C5. And the weighted sum q ₁ (C ₁ ) of Q1 in concept C1 is one. Therefore, the total weight of Q1 in concept C1 is 1/1 = 1.

For _{^{R 5 2 (C5-C1-}} C2), the relationship is C5-C1 because the C1-C2, the sum W (R ⁵ ₂₎ of the relationship between the weight is C5-C1 and C1-C2 in relation weighted sum of 1 + 3 = 4. And C2 is the concept Q1 Rating weight of the sum of q ₁ (C ₂₎ is 4 (D2) + 3 (D3 ) + 1 (D4) = 8. Therefore, the total weight of Q1 in concept C2 is 8/4 = 2.

By a process as described _{^{above, R 5 3 (C5-C1}} -C3), R 5 6 (C5-C1-C4-C6), R 5 7 total weight of Q1 for (C5-C1-C4-C7 ) are 3/2, 5/3, and 1/3, respectively. And the sum of weights of Q1 is 1 + 2 + 3/2 + 6/3 + 3/3 = 7.5 by adding all the weights of all relations.

If Q2 through Q6 are obtained by the same procedure, the following is obtained.

Q2: (3 + 1) / 4 = 1

Q3: 4/4 = 1

Q4: 1/4 + 3/2 = 1.75

Q5: 6/3 = 2

Q6: 3/3 = 1

Therefore, the recommended priorities are Q1, Q5, and Q4, respectively. Finally, Q2, Q3, and Q6 are the same. That is, the data object Q5 is the second highest since the number of teeth is the longest but the class weight is high. According to the first and second embodiments, the priorities are first determined by the number of sights, while the third embodiment obtains a priority for recommending the same considering the class weight and the relation weights reflecting the sums.

Next, the sum of weights when [Equation 2] is used is obtained as follows.

Q1: 1/1 + Max (4,3,1) / 4 + 3/2 + 6/3 + 3/3 = 6.5

Q2: Max (3,1) / 4 = 0.75

Q3: 4/4 = 1

Q4: 1/4 + 3/2 = 1.75

Q5: 6/3 = 2

Q6: 3/3 = 1

Compared to the previous case, the priority of Q2 is pushed to the end.

The invention made by the present inventors has been described concretely with reference to the embodiments. However, it is needless to say that the present invention is not limited to the embodiments, and that various changes can be made without departing from the gist of the present invention.

10: using subject terminal 20: network
30: definition support system 31: terminology system management unit
32: Data object storage unit 33: Data object definition unit
34: Qualifier recommendation section
40: Database 41: Terminology DB
42: Data object DB

Claims (2)

Heterogeneous health measuring instrument data exchange In semantic interoperability processing systems,
A reference term constituent unit configured to receive a plurality of reference terms and their relations and construct a reference term system;
A common term construction unit for receiving a common term and its relations and constructing a common term system based on the ontology;
A term mapping unit for mapping the terms in the reference term system to terms in the common term system and setting a term relation for the mapping between the terms; And
A term derivation unit that receives one query term in the reference term system and searches for terms in another reference term system corresponding to the query term, and searches for a term having a direct or indirect relationship with the query term, Wherein the data object definition support system comprises:
The method according to claim 1,
Wherein the common term construction unit searches for a data entity defined by a concept related to the concept (hereinafter referred to as a related concept) and recommends a qualifier of the retrieved data entity, and the related concept has a term relationship with the corresponding concept Heterogeneous health measuring instrument data exchange semantic interoperability processing system.

Images