KR101792332B1 - Method and Device for Providing Multi Path Relation Information between Entities - Google Patents
Method and Device for Providing Multi Path Relation Information between Entities Download PDFInfo
- Publication number
- KR101792332B1 KR101792332B1 KR1020150100006A KR20150100006A KR101792332B1 KR 101792332 B1 KR101792332 B1 KR 101792332B1 KR 1020150100006 A KR1020150100006 A KR 1020150100006A KR 20150100006 A KR20150100006 A KR 20150100006A KR 101792332 B1 KR101792332 B1 KR 101792332B1
- Authority
- KR
- South Korea
- Prior art keywords
- information
- entity
- multipath
- entities
- relation
- Prior art date
Links
Images
Classifications
-
- G06F19/24—
-
- G06F17/30604—
-
- G06F19/28—
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
An apparatus and method for providing multipath relationship information between entities is disclosed. The disclosed apparatus includes: an entity relation storage unit for storing an individual relationship type and direction information of the relation type; A user interface unit for receiving object information from a user; And a multipath entity information output unit for generating multipath information on the entity information inputted on the basis of the entity information input from the user, wherein the multipath information is obtained by connecting the direction information between the entities stored in the entity relation storage unit This information is generated. According to the disclosed apparatus and method, it is possible to provide multi-path information between entities, and scholars can provide more efficient information in constructing a hypothesis.
Description
The present invention relates to an apparatus and method for providing multi-path relationship information between entities, and more particularly, to an apparatus and method for providing multi-path information between entities in medical and bio-fields.
In the biotechnology field, a vast amount of research documents are stored in database form to obtain information on relationships among individuals. Since the amount of documents is vast, it is difficult to efficiently access the desired information. In order to establish a new hypothesis in the biotechnology field, it is not easy to learn relationship between entities from a large amount of research documents.
Text mining is used to obtain inter-entity relationship information from vast amounts of research documents. Text mining is a task of analyzing a document in a natural language, selecting the information that the user wants, and presenting the result in a refined and processed form.
In biotechnology, there is a fairly heterogeneous aspect compared to plain text because words that rarely appear in plain text form the text.
Text mining in biotechnology is done by recognizing pairs of entities with entity name and relationship type in biotext. Here, the object name refers to protein or lane electron, and the relation type means the interaction between the two objects such as inhibition.
The biotechnology document is updated in vast quantities, for example, about 500,000 new documents are added annually to the Medline database. Due to such a massive update, biologists were limited in reading the latest information about the relationships among individuals, and text mining has the advantage of effectively providing up-to-date and meaningful information to bio-researchers .
In the method of providing entity information using existing text mining, only relation type information about entity pair could be obtained. However, it could not provide relation information between plural entities although it could accurately express relationship information between both entities.
An aspect of the present invention is to propose a method and apparatus for providing multipath information for a plurality of entities.
According to an aspect of the present invention, there is provided an information processing apparatus comprising: an entity relation storage unit for storing a relationship type between entities and direction information of the relationship type; A user interface unit for receiving object information from a user; And a multipath entity information output unit for generating multipath information on the entity information inputted on the basis of the entity information input from the user, wherein the multipath information is obtained by connecting the direction information between the entities stored in the entity relation storage unit An apparatus for providing multi-path relationship information between entities, which is generated information, is provided.
Wherein the inter-entity direction information is information set based on the appearance order of the objects and the form of the sentence in the document in which the entity is recorded.
Wherein the form of the sentence includes a passive state and an active state.
The user interface unit receives the two pieces of entity information and provides the multipath information starting from the first entity and ending with the second entity.
The entity relationship storage further stores relationship classification information, and the relationship classification information includes at least whether the relationship is a negative relationship or an affirmative relationship.
The inter-entity multi-path relation information providing apparatus further includes a priority setting unit for setting a priority order of a plurality of multipath entity information output from the multi-path entity information output unit.
The priority setting unit sets a priority based on the reliability of the individual paths constituting the multipath.
The reliability of the individual paths constituting the multipath is based on the probability information that the entities constituting the individual paths simultaneously appear in one document.
According to another aspect of the present invention, there is provided a method comprising: (a) storing inter-entity relationship type and direction type information of the relationship type; (B) receiving entity information from a user; (C) generating multipath information on the inputted entity information based on the entity information input from the user, wherein the multipath information is generated by connecting directionality information between the entities stored in the step (a) A method of providing multipath relationship information between entities is provided.
According to still another aspect of the present invention, there is provided a recording medium on which a program for performing the above-described method is recorded and an executable program by the digital data processing apparatus is recorded.
According to the present invention, it is possible to provide multi-path information between entities, and scholars can provide more efficient information in constructing a hypothesis.
BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram illustrating a structure of a relationship setting machine for an entity according to an embodiment of the present invention; FIG.
Figure 2 illustrates a type of relationship between entities in accordance with an embodiment of the present invention.
Figure 3 illustrates a method for establishing a relationship between entities in accordance with an embodiment of the present invention.
4 illustrates a method for establishing relationships between entities in accordance with another embodiment of the present invention.
FIG. 5 is a block diagram illustrating a structure of an inter-entity multipath relationship information providing apparatus according to an embodiment of the present invention; FIG.
6 illustrates a method for estimating the reliability of path information according to an embodiment of the present invention.
7 is a flow diagram illustrating a method by which an entity sets a relationship from documents according to an embodiment of the present invention.
FIG. 8 is a flowchart showing a flow of a method for providing relation information between multipath entities according to an embodiment of the present invention; FIG.
While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It should be understood, however, that the invention is not intended to be limited to the particular embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention. Like reference numerals are used for like elements in describing each drawing.
Hereinafter, embodiments according to the present invention will be described in detail with reference to the accompanying drawings.
An apparatus for providing multi-path relationship information between entities according to an embodiment of the present invention stores relationship information between entities and provides relationship information in response to a request for association information of a specific entity of the user. For example, when a user requests biological relationships that exist between entity A and entity B, the apparatus and method for providing inter-entity multipath information according to an embodiment of the present invention may include not only a direct relationship between entity A and entity B, Provides multipath information between A and B
For example, an apparatus for providing multi-path information between entities according to an embodiment of the present invention includes not only A? B information but also multipath information between A and B, such as A? C? B and A? C? .
Here, an individual may be a substance such as a protein, and may be a disease or symptom such as diabetes manifesting in a person. The present invention provides a direct relationship between entity A and entity B as well as multiple paths via other entities. By providing such a multi-path, scholars can provide various information for hypothesizing.
The multipath provided by the present invention can provide multipath information between entities that scholars have not thought of, and it becomes possible to establish a new hypothesis for the development of new drugs and the like.
In order to provide multi-path information according to an embodiment of the present invention, it is necessary to set and store inter-entity relationship information suitable for multi-path provisioning. In this embodiment, Describe the task of providing the user with multipath information between entities after the setting of the relationship information is completed.
FIG. 1 is a block diagram illustrating a structure of an inter-entity relationship information setting machine according to an exemplary embodiment of the present invention. Referring to FIG. 1, It is a block diagram.
Referring to FIG. 1, an inter-entity relationship information setting machine according to an embodiment of the present invention includes a
The
For example, a database of academic documents that records interrelationship information is provided in advance, and academic documents can be searched and collected from the academic database.
In addition, it may search and collect academic documents based on address information of web documents that can search for academic documents or algorithms that can search web documents. It will be obvious to those skilled in the art that, of course, it is possible to navigate and collect academic documents in a variety of other known ways.
The
The relationships between the entities can be predefined, and Figure 2 is a diagram illustrating the types of relationships between entities in accordance with an embodiment of the present invention.
Referring to FIG. 2, the relationships among entities can be classified into four relations of "Positive" indicating affirmative, "Negative" indicating negative, "Neutral" indicating "Neutral", and "Plain" Relationship types corresponding to "Positive" may include "Increase", "Transmit" and "Substitute". Relationship types corresponding to "Negative" may include "Decrease" and "Remove". Relationship types that correspond to "Neutral" may include "Contain", "Modify", "Method", and "Report".
The verbs corresponding to each relationship type are illustrated in FIG. 1, and the types of relationships among individuals can be defined through verbs appearing in academic documents. Of course, the terms that define inter-entity relationships may be described by other parts of the verb than the verb.
The entity
FIG. 3 is a diagram illustrating a method of setting a relationship between entities according to an embodiment of the present invention.
Referring to FIG. 3, there is shown an example in which entity information and relationship information are extracted from a sentence "The results suggests that folic acid supplementation reduced arterial stiffness in smokers ", and the entity
Two entities are extracted from the sentence, the first being Folic Acid and the second being Arterial Stiffness. The type of relationship between entities can be verified by verbs. The verb "reduce" is used and the entity
The present invention additionally defines inter-entity direction information as well as inter-entity relationship type, and inter-entity direction information is information defined to provide multipath information between entities.
According to an embodiment of the present invention, the inter-entity direction information deduces from the form of the sentence extracting the entity. The sentence illustrated in FIG. 3 is an active sentence. The first entity, Folic Acid, appears before the verb, and the second entity, Arterial Stiffness, appears after the verb, thus defining directional information such as Arterial Stiffness in the Folic Acid.
As shown in FIG. 3, the entity
FIG. 4 is a diagram illustrating a method for setting a relationship between entities according to another embodiment of the present invention.
Referring to FIG. 4, there is shown an example of establishing an entity relationship from the sentence "The update of phosphate was enhanced by incorporation of lactosylceramide ".
Two individuals are extracted from the sentence, the first being the phosphate and the second being the lactosylceramide. The verb that defines the type of relationship between entities is enhanced, so that it can be inferred that the type of interaction between entities is an increase.
It can be confirmed that the sentence shown in FIG. 4 is a passive sentence, and when the sentence form is a passive sentence, the direction is set opposite to the word appearing order. In Fig. 4, Phosphate appears before the verb, and Lactosylceramide appears after the verb but since the sentence is passive, the direction is set to Lactosylceramide to Phosphate.
The entity information set by the machine is stored in the database and used for providing multipath information between entities according to an embodiment of the present invention.
On the other hand, the entity relationship setting section may additionally store the entity relationship major classification information, which includes "positive" and "negative" as described above.
FIG. 5 is a block diagram illustrating a structure of an apparatus for providing multi-path information between entities according to an embodiment of the present invention. Referring to FIG.
5 may be implemented in the form of a server, and may be implemented in the form of a program installed in a digital data processing apparatus such as a computer.
The apparatus for providing multi-path relationship information between entities according to an embodiment of the present invention includes an entity
The entity
According to a preferred embodiment of the present invention, a graph database capable of storing inter-entity directional information may be used as the entity
The
A single entity may be input to the
The multipath entity
For example, suppose the user inputs A and D. When the relationship information storage unit stores the relationship information of A? B and B? D, the multipath entity
As another example, when A → F, F → G, and G → D are stored as relation information between entities, the multipath entity
The multi-path object
The multi-path entity
The
According to an embodiment of the present invention, reliability of each path information output based on average semantic similarity is estimated.
FIG. 6 is a diagram illustrating a method for estimating reliability of path information according to an embodiment of the present invention. Referring to FIG.
Referring to FIG. 6, the path of Lactosylceramide → Breast Cancer → APTBA2 is illustrated. In order to estimate the reliability of the path information, the reliability of each path of the multipath is estimated and the final multipath reliability is estimated.
The reliability of the individual paths is estimated based on the average semantic similarity. For example, the reliability can be set based on the probability that words corresponding to two objects appear at the same time in one document. For example, if two words appear in multiple academic documents at the same time, the reliability of the individual path is estimated to be high. However, if the number of simultaneous appearances of two words in multiple academic documents is low, the reliability of individual paths is estimated to be low. Of course, it will be apparent to those skilled in the art that the reliability of the individual paths may be estimated by other means than the mean semantic similarity described above.
Referring to FIG. 6, the reliability of Lactosylceramide → Breast Cancer is -0.004293, and the reliability of Breast Cancer → APTBA2 is estimated to be 0.040319. If the relationship classification between two entities is irrelevant, the reliability can have a negative value.
The final reliability is estimated by dividing the reliability of the individual paths by the number of paths. FIG. 6 shows a case where there are two individual paths and a value obtained by adding 0.004293 and 0.040319, which are the reliability of individual paths, is divided by two.
When the priorities for each route are set by the
The multi-path entity relationship information providing apparatus as described above allows the user to receive multi-path information on the inputted entity keyword, and the user can generate a new hypothesis based on the new route information that the user has not recognized will be.
7 is a flowchart illustrating a method for establishing an entity from documents according to an embodiment of the present invention.
Referring to FIG. 7, first, documents to acquire entity information are collected (step 700). Document collection can be based on various information, and collects documents based on database information where documents are stored, address information of the World Wide Web where documents are stored, and the like.
When document collection is performed, text mining is performed to analyze the collected documents (step 702). In document work, extract information that can infer entity information and relationship between entities from sentences described in academic documents. As described above, the object information mainly has a noun form, and the information that can infer the relation between the objects has the form of a verb.
The relationship information between entities may additionally include relationship classification information, wherein the classification information is information for defining whether the relationship is an irrelevant relationship or an affirmative relationship. The relationship classification information can be used for the reliability calculation described above.
When the document analysis is performed, an object relationship setting operation is performed to set the relationship type between the two objects and the direction of the relationship (step 704). The orientation of the relationship between two entities can be determined based on the location of the entity in the sentence and the form of the sentence. If object A appears before the verb, object B appears after the verb, and the sentence is active, the direction to A B is set. On the other hand, when object A appears before the verb, object B appears after the verb, and the sentence is passive, the direction to B A is set.
FIG. 8 is a flowchart illustrating a method of providing relationship information between multipath entities according to an embodiment of the present invention. Referring to FIG.
The inter-entity relationship set through the operation of Fig. 7 is stored in the database (step 800). The database storing the relationships among the entities is preferably a graph database, but is not limited thereto.
And receives object information from the user to know the relationship (step 802). The user will be able to enter two objects and input one object. In the case of inputting only one object, the user may be provided with a plurality of path information starting from the corresponding object.
When receiving the entity information from the user, the multi-path information corresponding to the inputted entity information is output using the relation information between the two entities stored in step 800 (step 804).
Multipath information can be inferred using direction information between entities, and all inferable multipath information can be output.
When the multipath information is output, a priority is set for a plurality of multipath information (step 806). The priorities for a plurality of multipaths can be determined based on the reliability of the individual paths, and as described above, the reliability of the individual paths can be estimated based on the average semantic similarity, There will be.
As described above, the present invention has been described with reference to particular embodiments, such as specific elements, and specific embodiments and drawings. However, it should be understood that the present invention is not limited to the above- And various modifications and changes may be made thereto by those skilled in the art to which the present invention pertains. Accordingly, the spirit of the present invention should not be construed as being limited to the embodiments described, and all of the equivalents or equivalents of the claims, as well as the following claims, belong to the scope of the present invention .
Claims (17)
A user interface unit for receiving object information from a user;
And a multipath entity information output unit for generating multipath information on the inputted entity information based on the entity information inputted from the user,
Wherein the multipath information is information generated by connecting directional information between entities stored in the entity relation storage unit,
Wherein the entity relation storage further stores relationship classification information,
The information further includes a priority setting unit for setting a priority for a plurality of multipath entity information output from the multipath entity information output unit, including at least a relation of whether the relation is an irregular relation or an affirmative relation, Wherein the priority setting unit sets the priority based on the reliability of the individual paths constituting the multipath, and applies the negative reliability when the relationships between the entities are irrelevant.
Wherein the inter-entity direction information is information set based on the appearance order of the objects and the form of the sentence in the document in which the entity is recorded.
Wherein the form of the sentence includes a passive state and an active state.
Wherein the user interface unit receives the two pieces of entity information and provides the multi-path information starting from the first entity and ending with the second entity.
Wherein the priority setting unit sets a priority based on the reliability of the individual paths constituting the multi-path.
Wherein the reliability of the individual paths constituting the multipath is based on probability information that the entities constituting the individual path appear simultaneously in one document.
(B) receiving entity information from a user;
(C) generating multipath information on the inputted entity information based on the entity information input from the user,
Wherein the multipath information is information generated by connecting directionality information between entities stored in the step (a)
In the step (a), the relation classification information is further stored, and the relation classification information includes at least a relation of whether the relation is an irrelevant relation or an affirmative relation, and the plurality of multipath entity information generated in the step (c) Wherein the step of setting the priority sets a priority based on the reliability of the individual paths constituting the multi-path, and sets a negative reliability when the relationships between the entities are in an irrelevant relationship To-multipoint relationship information.
Wherein the inter-entity direction information is information set based on the appearance order of the objects and the form of the sentence in the document in which the entity is recorded.
Wherein the form of the sentence includes a passive state and an active state.
Wherein the second entity information is input in step (b), and the multipath information starting from the first entity and ending with the second entity is provided in step (c).
Wherein the reliability of the individual paths constituting the multipath is based on probability information that the entities constituting the individual path appear simultaneously in one document.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR1020150100006A KR101792332B1 (en) | 2015-07-14 | 2015-07-14 | Method and Device for Providing Multi Path Relation Information between Entities |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR1020150100006A KR101792332B1 (en) | 2015-07-14 | 2015-07-14 | Method and Device for Providing Multi Path Relation Information between Entities |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| KR20170008616A KR20170008616A (en) | 2017-01-24 |
| KR101792332B1 true KR101792332B1 (en) | 2017-11-20 |
Family
ID=57993215
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| KR1020150100006A KR101792332B1 (en) | 2015-07-14 | 2015-07-14 | Method and Device for Providing Multi Path Relation Information between Entities |
Country Status (1)
| Country | Link |
|---|---|
| KR (1) | KR101792332B1 (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102382914B1 (en) * | 2020-06-17 | 2022-04-06 | 주식회사 엔씨소프트 | Method and apparatus for relation extraction between entities |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101488338B1 (en) * | 2014-10-20 | 2015-01-30 | 한국과학기술정보연구원 | method for combining biopathways, apparatus for combining biopathways and storage medium for storing a program combining biopathways |
-
2015
- 2015-07-14 KR KR1020150100006A patent/KR101792332B1/en active IP Right Grant
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101488338B1 (en) * | 2014-10-20 | 2015-01-30 | 한국과학기술정보연구원 | method for combining biopathways, apparatus for combining biopathways and storage medium for storing a program combining biopathways |
Also Published As
| Publication number | Publication date |
|---|---|
| KR20170008616A (en) | 2017-01-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110765275B (en) | Search method, search device, computer equipment and storage medium | |
| CN111143479B (en) | Knowledge graph relation extraction and REST service visualization fusion method based on DBSCAN clustering algorithm | |
| US9009134B2 (en) | Named entity recognition in query | |
| US20150074112A1 (en) | Multimedia Question Answering System and Method | |
| CN106815307A (en) | Public Culture knowledge mapping platform and its use method | |
| CN109325201A (en) | Generation method, device, equipment and the storage medium of entity relationship data | |
| CN108305180B (en) | Friend recommendation method and device | |
| JP5012078B2 (en) | Category creation method, category creation device, and program | |
| KR102334236B1 (en) | Method and application of meaningful keyword extraction from speech-converted text data | |
| CN101739407A (en) | Method and system for automatically constructing information organization structure used for related information browse | |
| JP7103496B2 (en) | Related score calculation system, method and program | |
| KR102334255B1 (en) | Text data collection platform construction and integrated management method for AI-based voice service | |
| KR101638535B1 (en) | Method of detecting issue patten associated with user search word, server performing the same and storage medium storing the same | |
| Bykau et al. | Fine-grained controversy detection in Wikipedia | |
| Sabri et al. | Web data extraction approach for deep web using WEIDJ | |
| JP2014146218A (en) | Information providing device | |
| CN107506349A (en) | A kind of user's negative emotions Forecasting Methodology and system based on network log | |
| KR101375221B1 (en) | A clinical process modeling and verification method | |
| KR101792332B1 (en) | Method and Device for Providing Multi Path Relation Information between Entities | |
| KR101741249B1 (en) | System and method for generating category | |
| JP5112027B2 (en) | Document group presentation device and document group presentation program | |
| KR101846347B1 (en) | Method and apparatus for managing massive documents | |
| KR20160136014A (en) | Method and system for topic clustering of big data | |
| KR20110062896A (en) | Apparatus and method for searching local information | |
| KR101526312B1 (en) | System and method for packaging research and development information services on pending national issues |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| A201 | Request for examination | ||
| E902 | Notification of reason for refusal | ||
| E701 | Decision to grant or registration of patent right | ||
| GRNT | Written decision to grant |