KR101686764B1 - Apparatus for substituting predicate and ranking method for verb predicate thereof - Google Patents
Apparatus for substituting predicate and ranking method for verb predicate thereof Download PDFInfo
- Publication number
- KR101686764B1 KR101686764B1 KR1020160002546A KR20160002546A KR101686764B1 KR 101686764 B1 KR101686764 B1 KR 101686764B1 KR 1020160002546 A KR1020160002546 A KR 1020160002546A KR 20160002546 A KR20160002546 A KR 20160002546A KR 101686764 B1 KR101686764 B1 KR 101686764B1
- Authority
- KR
- South Korea
- Prior art keywords
- verb
- predicate
- noun
- predicates
- domain
- Prior art date
Links
Images
Classifications
-
- G06F17/3053—
-
- G06F17/274—
-
- G06F17/277—
-
- G06F17/30389—
-
- G06F17/30442—
Landscapes
- Machine Translation (AREA)
Abstract
A verb predicate ranking method for verb predicate replacement of a predicate replacement apparatus and a noun predicate thereof according to the present invention includes: obtaining domain information corresponding to a field of application of a noun predicate and a noun predicate; Extracting a plurality of verb predicates corresponding to the noun predicates; Generating a plurality of query terms corresponding to each verb predicate by combining each of the extracted verb predicates with the domain information; Measuring a frequency of use of each verb predicate by searching the plurality of query terms through the set search engine; And assigning a ranking of each verb predicate based on the frequency of use of each verb predicate.
Description
The present invention relates to a verb substitution device for selecting and replacing verb predicates in consideration of the application fields of noun predicates when substituting verb predicates of noun predicates for knowledge search and the like, and verbs for verb substitution of noun predicates To a predicate ranking method.
Wikipedia is an open-source encyclopedia based on Wiki Syntax that can be freely used by anyone on the Web.
A Wikipedia page is called an article, and an article is a plain text part that describes an independent subject or entity, an image that provides additional information, a hyperlink, a category, a template, etc. It consists of various structured texts.
Templates are one of the useful functions provided by wikis. They define specific parts that are repeatedly used in several pages in advance as functions, and repeatedly use the same template by adding new instances to each page on multiple pages. It says.
This is the same as calling a foreign function in a programming language. The most popular template provided by Wikipedia is InfoBox.
An infobox is a type of metadata that is located at the top of an article and provides a summary of a single article. It is written in the form of describing key attributes and their values for each object. However, the level of information provided by InfoBox has different disadvantages depending on language and subject. That is, as shown in FIG. 1, the amount of information provided through Info Box differs depending on the subject (see English Wikipedia of pasta and English Wikipedia of dumpling), and the amount of information provided through InfoBox There are also disadvantages (see Pasta's English Wikipedia and Korean Wikipedia).
This means that even users of the same Wikipedia will differ in the amount of information that can be provided according to the topic and the language to be searched.
Thus, when a source language and a target language, which are two different languages, are given, Korean Patent No. 10-1497806 discloses a method for inputting topics written in a source language, We propose a knowledge convergence system and method that can generate new knowledge by automatically generating technology, adding instances, and merging it with infobox of the target language.
However, when retrieving content from the Web or a document in a sentence unit (simple sentence in the form of subject-predicate-object), especially when the vocabulary of the predicate of a given search term and the target sentence are different from each other, Lt; / RTI >
For example, if you want to search Wikipedia for the ingredients of the Acorn Mum, you can create the following query.
Ingredient (Dotorimuk ,?)
On the other hand, there is a sentence on the Dotorimuk page of English Wikipedia.
"Dotorimuk (also spelled tot'orimuk) or acorn jelly is a Korean food which is a jelly made from acorn starch."
However, since the predicate of the search term does not exist in the target sentence, the sentence will not be searched. Even if the subject is matched and the searched word exists, there are many predicates ('is' or 'made from'), You will not know if the object you want is 'Korean food' or 'acorn starch'.
In existing language resources (dictionaries, homologous dictionaries, etc.), there is a disadvantage that the verb form can not be presented if the noun predicate does not include the verb root.
Thus, in Korean Patent No. 10-1497806, when the predicate of the search word and the root of the search target are different from each other (ingredient vs. made from), the Predicate Substitution of the knowledge fusion process using DBpedia As mentioned above, the noun predicate performs the function of replacing the noun predicate with the verb predicate even if the noun predicate does not include the verb root.
However, the Predicate Substitution of the knowledge fusion process using the conventional DBpedia provides multiple verb predicates for a single noun predicate, but does not show the ranking of each verb predicate, Has a limitation that it does not show whether it improves the result of the search further.
The present invention proposes a verb substitution method which gives a ranking of verb predicates corresponding to a noun predicate in each application field of a noun predicate and performs a substitution operation with a verb predicate on the basis thereof to greatly improve the accuracy and reliability of a substitution operation And a method of ranking verb predicates for verb predicate replacement of a device and its noun predicate.
According to an embodiment of the present invention, there is provided a method for generating a noun predicate, the method comprising: obtaining domain information corresponding to an application field of a noun predicate and a noun predicate; Extracting a plurality of verb predicates corresponding to the noun predicates; Generating a plurality of query terms corresponding to each verb predicate by combining each of the extracted verb predicates with the domain information; Measuring a frequency of use of each verb predicate by searching the plurality of query terms through the set search engine; And predicting each of the verb predicates based on the frequency of use of each of the verb predicates. The present invention also provides a verb predicate ranking method for verb predicate replacement of a noun predicate.
The domain information is a domain name or a set of domain instances.
The step of generating the query term may include generating at least one of a verb form of the verb predicate, a third-person singular form, a past form, a current injection form, and a past form, and then generating a query having the domain information as a subject or object of the verb predicate .
The step of ranking each of the verb predicates comprises:
According to another embodiment of the present invention, there is provided a domain name and predicate predicate acquisition unit for obtaining domain information corresponding to a field of application of a noun predicate and a noun predicate. A verb predicate extractor for extracting a plurality of verb predicates corresponding to the noun predicates; Combining the extracted verb predicates with the domain information to generate a query term corresponding to each verb predicate and then searching the web through a predefined search engine to determine the frequency of use of each verb predicate And a verb predicate rank calculation unit for assigning the rank of each verb predicate by analyzing; And a verb predicate replacement unit replacing the noun predicate using a verb predicate having a highest ranking among the plurality of verb predicates.
The technique disclosed in the present invention can have the following effects. It is to be understood, however, that the scope of the disclosed technology is not to be construed as limited thereby, as it is not meant to imply that a particular embodiment should include all of the following effects or only the following effects.
The verb substitution apparatus of the present invention and the verb predicate ranking method for the verb predicate substitution of the noun predicate of the present invention are classified into the verb predicate corresponding to the same noun predicate and the use frequency of each verb predicate according to the domain to which the noun predicate is applied In consideration of the occurrence of the difference, it is possible to give a ranking to the substituted verb predicates in a single noun predicate considering the domain in which the verb predicates are used. As a result, based on the domain to which the noun predicate is applied, it is possible to grasp and replace the preferred verb predicate in the field, thereby improving the accuracy and reliability of the replacement operation.
1 is a diagram for explaining an amount of information provided for different languages and themes.
FIG. 2 is a diagram for explaining the principle of substitution of a verb predicate with a verb predicate according to an embodiment of the present invention.
3 is a diagram for explaining a predicate replacement apparatus according to an embodiment of the present invention.
4 is a diagram for explaining a verb predicate ranking method for verb predicate replacement of a noun predicate according to an embodiment of the present invention.
FIG. 5 is a diagram for explaining a verb predicate ranking method for verb predicate replacement of a noun predicate according to another embodiment of the present invention.
For the embodiments of the invention disclosed herein, specific structural and functional descriptions are set forth for the purpose of describing an embodiment of the invention only, and it is to be understood that the embodiments of the invention may be practiced in various forms, The present invention should not be construed as limited to the embodiments described in Figs.
The present invention is capable of various modifications and various forms, and specific embodiments are illustrated in the drawings and described in detail in the text. It is to be understood, however, that the invention is not intended to be limited to the particular forms disclosed, but on the contrary, is intended to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.
The terms first, second, etc. may be used to describe various components, but the components should not be limited by the terms. The terms may be used for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component.
It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, . On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between. Other expressions that describe the relationship between components, such as "between" and "between" or "neighboring to" and "directly adjacent to" should be interpreted as well.
The terminology used in this application is used only to describe a specific embodiment and is not intended to limit the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In the present application, the terms "comprise", "having", and the like are intended to specify the presence of stated features, integers, steps, operations, elements, components, or combinations thereof, , Steps, operations, components, parts, or combinations thereof, as a matter of principle.
Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Terms such as those defined in commonly used dictionaries should be construed as meaning consistent with meaning in the context of the relevant art and are not to be construed as ideal or overly formal in meaning unless expressly defined in the present application .
Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The same reference numerals are used for the same constituent elements in the drawings and redundant explanations for the same constituent elements are omitted.
In the substitution of a verb predicate with a verb predicate according to the present invention, when a verb predicate is replaced with a plurality of verb predicates having different roots, more preferred predicates exist depending on the domain, An algorithm that assigns a ranking to each verb predicate. The user has the advantage of improving the accuracy of a job by using a more preferred verb predicate depending on the ranking.
FIG. 2 is a diagram for explaining the principle of substitution of a verb predicate with a verb predicate according to an embodiment of the present invention.
"add", "enter", "list", "list", "list", " quot; by "and" occupy with ".
However, as shown in FIG. 2, the use frequency of each verb predicate differs depending on the use domain of the noun predicate, even if the verb surname corresponds to the same noun predicate.
That is, "make from" is the most representative use in the food domain, while "contain" is the most representative use in the drug domain.
As a result, it can be seen that the "ingredient" used in the food domain is replaced with a verb predicate "make from", and the "ingredient" used in the drug domain is most preferably replaced with the verb predicate "contain".
Accordingly, the present invention proposes a technique for identifying the use frequency of each verb predicate in a domain in which "ingredient" is used and performing a noun predicate replacement operation based on the verb predicate having the highest use frequency.
3 is a diagram for explaining a predicate replacement apparatus according to an embodiment of the present invention.
3, the predicting apparatus of the present invention includes a domain and noun
Domain and noun
The verb
The verb predicate
For the sake of reference, the query language of the present invention can be implemented in a form having a verb predicate as a verb, a domain instance as a subject or an object, and the verb predicate has not only a form of a prototype of a verb but also a verb form, a past form, Or the like.
The verb predicate
At this time,
Finally, the verb
Hereinafter, referring to FIG. 4, a verb predicate ranking method for verb predicate replacement of a noun predicate according to an embodiment of the present invention will be described.
In FIG. 4, for convenience of explanation, description will be made only in the case where pizza and pasta are input as the instance D representing the domain to which the ingredient and the noun predicate P are to be applied as the noun predicate P.
First, the predicate replacement apparatus 100 obtains information about a noun predicate P, a term representing a food domain, and a food domain instance D (S1).
Then, using the Machine Readable Dictionary, knowledge base, and encyclopedia, such as Wikipedia, WordNet, Google Translator, Naver Dictionary, etc., Quot ;, " list of ", " make in ", "quot; and "occupy with" (S2).
Then, a plurality of query terms corresponding to each of a plurality of verb predicates is generated by combining the plurality of verb verbs, the present verb, the past verb, and the past verb form with each of the domain instances D (S3), the frequency of use of each of a plurality of verb predicates is measured based on a plurality of query terms (S4).
For example, in the case of a verb predicate of "make from", a large number of query terms (d i || v f ) can be acquired in the following manner, and "make from" (Hits) can be measured.
D: pizza, pastry
v: make
v f : made, makes, making
d i || v f: "pizza make", "pizza made", "pizza makes", "pizza making", "pasta make", "pasta made", "pasta makes", "pasta making"
hits (search of make): 7,452,928 cases
It is also possible to measure hits in the same way in the case of the verb predicate of "consist of".
D: pizza, pastry
v: contain
v f : contains, contained, containing
d i || v f: "pizza contain", "pizza contains", "pizza contained", "pizza containing", "pasta contain", "pasta contains", "pasta contained", "pasta containing"
hits (contain search): 101,411 cases
Then, the use frequency of each of a plurality of verb predicates is normalized to the use frequency of the verb verbs whose verifications are checked the most frequently among a plurality of verb verbs by using
That is, since make is used most frequently in the case of the food domain, make becomes normalizer (denominator in Equation 1), so that DVA of make becomes 7,452,928 / 7,452,928 = 1 and DVA of contain is 101,411 / 7,452,928 = 0.14. Of course, the DVA values of the remaining verb predicates are also obtained in this manner. The thus obtained DVA value has a value between 0 and 1.
Then, the respective verb predicates are ranked based on the DVA values corresponding to each of the plurality of verb predicates (S6). As a result, the noun predicate applied to the food domain is replaced with a verb predicate (ie, make from) with the highest DVA value among the many verb predicates.
In addition, FIG. 5 shows DVA values of each verb predicate corresponding to a noun predicate used in the drug domain. Referring to FIG. 5, DVA values of a plurality of verb predicates corresponding to the same noun predicate ingredient As shown in Fig.
In other words, by inputting "aspirin" and "advil" entities representing the drug domain instead of "pizza" and "pasta" as entities representing the food domain, quot; contain ". < / RTI >
As described above, the method according to one embodiment of the present invention can be implemented as program instructions for implementing the above. The computer-readable recording medium on which the program instructions are recorded includes, for example, a ROM, a RAM, CD-ROMs, magnetic tapes, floppy disks, optical media storage devices, and the like.
The computer-readable recording medium on which the above-described program is recorded may be distributed to a computer apparatus connected via a network so that computer-readable codes can be stored and executed in a distributed manner. In this case, one or more of the plurality of distributed computers may execute some of the functions presented above and send the results of the execution to one or more of the other distributed computers, The computer may also perform some of the functions described above and provide the results to other distributed computers as well.
A computer capable of reading a recording medium on which an application as a program for driving the apparatus and method according to each embodiment of the present invention is read can be used not only in a general PC such as a general desktop or a notebook computer but also in a smart phone, Personal digital assistants (PDAs), mobile communication terminals, and the like, and it should be interpreted as all devices capable of computing.
While the present invention has been described in connection with what is presently considered to be the most practical and preferred embodiments, it is to be understood that the invention is not limited to the disclosed embodiments. That is, within the scope of the present invention, all of the components may be selectively coupled to one or more of them. In addition, although all of the components may be implemented as one independent hardware, some or all of the components may be selectively combined to perform a part or all of the functions in one or a plurality of hardware. As shown in FIG. The codes and code segments constituting the computer program may be easily deduced by those skilled in the art. Such a computer program is stored in a computer readable storage medium, readable and executed by a computer, thereby realizing an embodiment of the present invention. As a storage medium of the computer program, a magnetic recording medium, an optical recording medium, or the like can be included.
The foregoing description is merely illustrative of the technical idea of the present invention, and various changes and modifications may be made by those skilled in the art without departing from the essential characteristics of the present invention. Therefore, the embodiments disclosed in the present invention are intended to illustrate rather than limit the scope of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. The scope of protection of the present invention should be construed according to the following claims, and all technical ideas within the scope of equivalents should be construed as falling within the scope of the present invention.
Claims (5)
Extracting a plurality of verb predicates corresponding to the noun predicates;
Generating a plurality of query terms corresponding to each verb predicate by combining each of the extracted verb predicates with the domain information;
Searching the web by a plurality of query terms through a predetermined search engine and measuring frequency of use of each verb predicate; And
And assigning a ranking of each verb predicate based on a frequency of use of each verb predicate,
The step of generating the query term
Wherein a query term having at least one of a verb form of the verb predicate, a third person singular form, a past form, a current form of injection, and a past form of injection is obtained and the domain information is used as a subject or object of the verb predicate A method of predicting verb predicates for verb substitution.
A domain name or a set of domain instances. A method of ranking verb predicates for verb predicate substitution of a noun predicate.
"
remind
A verb predicate extractor for extracting a plurality of verb predicates corresponding to the noun predicates;
Combining the extracted verb predicates with the domain information to generate a query term corresponding to each verb predicate and then searching the web through a predefined search engine to determine the frequency of use of each verb predicate And a verb predicate rank calculation unit for assigning the rank of each verb predicate by analyzing; And
And a verb predicate replacing unit replacing the verb predicate using a verb predicate having a highest rank among the plurality of verb predicates,
The verb predicate rank calculation unit may calculate,
"
remind
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR20150010757 | 2015-01-22 | ||
| KR1020150010757 | 2015-01-22 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| KR20160090744A KR20160090744A (en) | 2016-08-01 |
| KR101686764B1 true KR101686764B1 (en) | 2016-12-16 |
Family
ID=56706941
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| KR1020160002546A KR101686764B1 (en) | 2015-01-22 | 2016-01-08 | Apparatus for substituting predicate and ranking method for verb predicate thereof |
Country Status (1)
| Country | Link |
|---|---|
| KR (1) | KR101686764B1 (en) |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101233423B1 (en) * | 2012-05-17 | 2013-02-15 | 한국과학기술정보연구원 | Device and method for extacting descriptive expressions denoting terminological concepts from documents |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101497806B1 (en) * | 2013-06-12 | 2015-03-04 | 한국과학기술원 | System and method for knowledge fusion |
-
2016
- 2016-01-08 KR KR1020160002546A patent/KR101686764B1/en active IP Right Grant
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101233423B1 (en) * | 2012-05-17 | 2013-02-15 | 한국과학기술정보연구원 | Device and method for extacting descriptive expressions denoting terminological concepts from documents |
Also Published As
| Publication number | Publication date |
|---|---|
| KR20160090744A (en) | 2016-08-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Stamatatos et al. | Overview of the PAN/CLEF 2015 evaluation lab | |
| US20170177180A1 (en) | Dynamic Highlighting of Text in Electronic Documents | |
| Forstall et al. | Modeling the scholars: Detecting intertextuality through enhanced word-level n-gram matching | |
| Dawdy-Hesterberg et al. | Learnability and generalisation of Arabic broken plural nouns | |
| US10242033B2 (en) | Extrapolative search techniques | |
| RU2674331C2 (en) | System and process for analysis, qualification and acquisition of sources of unstructured data by means of empirical attribution | |
| JP6404511B2 (en) | Translation support system, translation support method, and translation support program | |
| KR101491628B1 (en) | Method, apparatus and system for extracting keyword affecting for mood change of the public using blog | |
| JP6529133B2 (en) | Apparatus, program and method for analyzing the evaluation of topics in multiple regions | |
| Rodrigues et al. | Advanced applications of natural language processing for performing information extraction | |
| US9378272B1 (en) | Determining correction of queries with potentially inaccurate terms | |
| WO2015004006A1 (en) | Method and computer server system for receiving and presenting information to a user in a computer network | |
| JP5718405B2 (en) | Utterance selection apparatus, method and program, dialogue apparatus and method | |
| US20190303437A1 (en) | Status reporting with natural language processing risk assessment | |
| RU2563148C2 (en) | System and method for semantic search | |
| JP6942759B2 (en) | Information processing equipment, programs and information processing methods | |
| JP6106489B2 (en) | Semantic analyzer and program | |
| Noah et al. | Evaluation of lexical-based approaches to the semantic similarity of Malay sentences | |
| KR101686764B1 (en) | Apparatus for substituting predicate and ranking method for verb predicate thereof | |
| KR101308821B1 (en) | Keyword extraction system for search engines and extracting method thereof | |
| KR101781597B1 (en) | Apparatus and method for creating information on electronic publication | |
| KR101238927B1 (en) | Electronic book contents searching service system and electronic book contents searching service method | |
| JP6488399B2 (en) | Information presentation system and information presentation method | |
| JP2009086772A (en) | Retrieval service device | |
| Pentoney et al. | Does query expansion limit our learning? A comparison of social-based expansion to content-based expansion for medical queries on the internet |
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 | ||
| FPAY | Annual fee payment |
Payment date: 20191126 Year of fee payment: 4 |