KR20190011184A - Method and apparatus for providing information using degree of association between reserved word and attribute language - Google Patents

Abstract

An embodiment of the present invention relates to an information providing method using a degree of association between a reserve word and an attribute word, and an apparatus thereof. According to an embodiment of the present invention, the information providing method comprises the following steps: storing a degree of association of an object-keyword corresponding to an object item and a pair of representative attribute keywords; storing a degree of association of a basic reserve word-keyword corresponding to a reserve word and the pair of representative attribute keywords; obtaining a receiving reserve word; obtaining the degree of association of a reserve word-object corresponding to the receiving reserve word and each of a pair of object items, by using the degree of association of the object-keyword and the degree of association of the basic reserve word-keyword; and providing the object item according to the degree of association of the reserve word-object corresponding to each of the pair of the object items. According to an embodiment of the present invention, an efficient information providing method using a reserve word and an apparatus thereof can be provided.

Description

TECHNICAL FIELD [0001] The present invention relates to a method and apparatus for providing information using an association degree between a reserved word and an attribute language,

At least some embodiments of the present disclosure relate to a method and apparatus for providing information using associations between reserved words and attribute languages.

According to a conventional search method, a user can search a desired web document or the like by inputting a search keyword into a search window. For example, a user may enter the movie title " interstellar " into the search box to retrieve information about the movie " interstellar. &Quot; However, if the user can not remember the movie title he / she wants to find, he / she needs to provide information in a different way. For example, a user can try to search for an actor who is looking for a movie, a director, a producer of the movie, or the like as a keyword. Movie information sites and movie review sites often provide movie information as well as cast information, so if the luck is not bad, the user may be able to find the desired movie by using the actor, director, producer, etc. as keywords.

However, if the information is provided based on an atypical language, for example, an emotional language, rather than the stereotyped information, the conventional retrieval method can not be used. For example, the response provided by conventional search engines for search terms such as "funny movies" or "movies when they are sad" is that someone searches for documents that have been created, including keywords such as "funny movies" or " It will be nothing more. However, atypical language requires a different approach than formalized information such as starring actor, (genre) movie genre, and year of release. Even if you did not include someone with a "funny movie" or "sad movie" keyword, there could be a lot of movies that many people might feel like "fun" or "when you're sad." Furthermore, for other areas of the film, a different approach may be required for requesting information using an unstructured language.

KR 10-2009-0080822 A

An embodiment of the present invention aims to provide a method and apparatus for efficiently providing information using reserved words.

An information providing method according to an embodiment of the present invention includes extracting a representative attribute keyword candidate set including a representative attribute keyword, setting a reserved word set including a reserved word, setting an object corresponding to an object item and a representative attribute keyword pair - storing a keyword association, associating weight values corresponding to the representative attribute keyword and the lower keyword pair, and a basic keyword corresponding to the reserved word and the lower keyword pair, Storing a reserved word-keyword association, acquiring a receiving reserved word, acquiring a reserved word-object association diagram corresponding to the receiving reserved word and each object item pair using the object-keyword association diagram and the basic reserved word- And a step of obtaining the reservation word and the reservation - it may include the steps of providing an object item in accordance with the objects associated degrees.

The information providing apparatus according to an embodiment of the present invention extracts a representative attribute keyword candidate set including a representative attribute keyword, sets a reserved word set including a reserved word, and sets an object keyword corresponding to the object item and the representative attribute keyword pair The association degree is stored in the storage unit, and the association weight corresponding to the representative attribute keyword and the lower keyword pair, and the basic word corresponding to the reserved word and the lower keyword pair corresponding to the reserved word and the lower keyword pair, A storage unit for storing the reserved word-keyword association in the storage unit; a storage unit for storing the object-keyword association degree and the basic reserved word-keyword association degree; And a communication unit for acquiring a reception reservation word. The control unit may obtain the reserved word-object association map corresponding to the reception reservation word and each object item pair using the object-keyword association map and the basic reserved word-keyword association map. The control unit may provide an object item according to the reserved word and the reserved word-object association diagram corresponding to each object item pair.

According to an embodiment of the present invention, a method and apparatus for efficiently providing information using reserved words can be provided.

1 is a network diagram of an information providing system using an attribution language according to an embodiment of the present invention.
2 is a block diagram of a terminal 200 according to an embodiment of the present invention.
3 is a block diagram of an information providing apparatus 300 according to an embodiment of the present invention.
4 is a flowchart of an information providing process through an information providing interface according to an embodiment of the present invention.
5 is a detailed flowchart of step 910 in accordance with one embodiment of the present disclosure.
6 is a detailed flowchart of step 510 in accordance with one embodiment of the present disclosure.
7 is a detailed flowchart of step 530 according to one embodiment of the present disclosure.
8 is a flowchart of an information providing process according to another embodiment of the present invention.
9 is a flowchart of an information providing process according to an embodiment of the present invention.
10 is an illustration of a stored object-keyword association diagram in accordance with one embodiment of the present disclosure.
11 is an illustration of a basic reserved word-keyword association according to one embodiment of the present disclosure.
12 is a detailed flowchart of step 940 in accordance with one embodiment of the present disclosure.
13 is a flowchart of an information providing process according to another embodiment of the present invention.
14 is a flowchart of an information providing process according to another embodiment of the present invention.
15 is a detailed flowchart of step 1340 according to the first embodiment of the present invention.
16 is a detailed flowchart of step 1340 according to another embodiment of the present disclosure.
17 is a detailed flowchart of step 1340 in accordance with another embodiment of the present disclosure.
18 is a detailed flowchart of step 1320 according to a modification of the present disclosure.
FIG. 19 is an illustration of interface 1900 generated in accordance with the interface information provided in step 1840.
Figure 20 is a diagram illustrating a hierarchy of terms in accordance with one embodiment of the present disclosure.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

In describing the embodiments, descriptions of techniques which are well known in the technical field to which this specification belongs and which are not directly related to this specification are not described. This is for the sake of clarity without omitting the unnecessary explanation and without giving the gist of the present invention.

For the same reason, some of the components in the drawings are exaggerated, omitted, or schematically illustrated. Also, the size of each component does not entirely reflect the actual size. In the drawings, the same or corresponding components are denoted by the same reference numerals.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

1 is a network diagram of an information providing system using an attribution language according to an embodiment of the present invention.

Referring to FIG. 1, an information providing system according to an embodiment of the present invention may include a terminal 200, an information providing apparatus 300, and a communication network 150.

The terminal 200 may be implemented as an electronic or similar device capable of performing communication, receiving input from a user, and outputting a screen, for example, a smart phone, a PDA, a tablet PC, a notebook computer, a laptop computer, .

The information providing device 300 may be implemented with, for example, a workstation, a server, a general purpose computer, an electronic device capable of performing other communications, or the like.

The terminal 200 and the information providing apparatus 300 are connected through the communication network 150 and communicate with each other through the communication network 150.

The communication network 150 may be, for example, an LTE (Long Term Evolution), an LTE-Advanced (LTE-A), a WI-FI, a LAN (Local Area Network), a WAN (Wide Area Network), a CDMA (TDMA), a wireless broadband (WiBro), a global system for mobile communications (GSM), and other communication methods that have been developed in the past, presently available, or become available in the future. Hereinafter, it is assumed that the terminal 200 and the information providing apparatus 300 directly communicate without referring to the communication network 150 for convenience.

Specific operations and configurations of the terminal 200 and the information providing apparatus 300 will be described later with reference to Figs. 2 to 8. Fig.

2 is a block diagram of a terminal 200 according to an embodiment of the present invention.

Referring to FIG. 2, the terminal 200 according to an embodiment of the present invention may include an input unit 210, a display unit 220, a communication unit 230, a storage unit 240, and a controller 250.

The input unit 210 converts an input operation of a user into an input signal and transmits the input signal to the controller 250. The input unit 210 may be implemented as, for example, a keyboard, a mouse, a touch sensor on a touch screen, a touch pad, a keypad, a voice input, For example, the input unit 210 may receive a user input request and transmit the input information to the control unit 250.

The display unit 220 displays a screen under the control of the control unit 250. [ The display unit 220 may be implemented as a display device which may be, for example, an LCD (Liquid Crystal Display), an LED (Light Emitting Diode), an OLED (Organic Light Emitting Diode), a projector, The display unit 220 can display, for example, an interface page for providing information or an information providing result page. In an embodiment, instead of the display unit 220, a configuration unit using other modes such as voice output, vibration, and the like may be used instead of the display unit 220. [

The communication unit 230 exchanges data with the information providing device 300 and / or other external devices. The communication unit 230 transmits the data received from the information providing device 300 to the control unit 250. Also, the communication unit 230 transmits data to the information providing apparatus 300 under the control of the controller 250. The communication technology used by the communication unit 230 may vary depending on the type of the communication network 150 or other circumstances.

The storage unit 240 stores data according to the control of the controller 250 and transmits the requested data to the controller 250. [

The control unit 250 controls the overall operation of the terminal 200 and each component. The control unit 250 transmits an information providing request or other data to the information providing apparatus 300 according to the information input from the input unit 210 as described later, And displays the result page and / or the interface page through the display unit 220.

The operations performed by the control unit 250 may be distributed and processed by a plurality of physically separated computing devices. Some of the operations performed by the control unit 250 may be performed by the first server, and the other operations performed by the second server may be performed. In this case, the control unit 250 may be implemented as a sum of physically separated computing devices.

The storage unit 240 may be implemented as a sum of physically separated storage devices.

When the control unit 250 or the storage unit 240 is implemented as a total of various devices physically separated from each other, communication between various devices may be required. Here, for the sake of simplicity of explanation, it is assumed that the storage unit 240 and the control unit 250 are implemented as one object.

The communication unit 230 may transmit or receive data under the control of the control unit 250 according to the viewpoint when the terminal 200 transmits and receives data or the control unit 250 may control the communication unit 230 to transmit and receive data It can also be expressed.

The specific operation of each component of the terminal 200 will be described later with reference to FIG. 4 to FIG.

3 is a block diagram of an information providing apparatus 300 according to an embodiment of the present invention.

Referring to FIG. 3, an information providing apparatus 300 according to an embodiment of the present invention may include a communication unit 310, a control unit 320, and a storage unit 330.

The communication unit 310 exchanges data with the terminal 200 and / or other external devices. The communication unit 310 transmits the data received from the terminal 200 to the control unit 320. In addition, the communication unit 310 transmits data to the terminal 200 under the control of the control unit 320. The communication technology used by the communication unit 310 may vary depending on the type of the communication network 150 or other circumstances.

The storage unit 330 stores data according to the control of the control unit 320 and transmits the requested data to the control unit 320 from the control unit 320.

The control unit 320 controls the overall operation of the information providing device 300 and each component. When receiving the interface page request, the information providing result page request, and other data through the communication unit 310 as described later, the control unit 320 loads necessary data from the storage unit 330 and generates page information And transmits the page information to the terminal 200 through the communication unit 310. [

The control unit 320 may transmit data to the communication unit 310 under the control of the control unit 320 in accordance with a viewpoint when the information providing apparatus 300 transmits and receives data, It can also be expressed as sending and receiving.

Specific operations of the respective components of the information providing apparatus 300 will be described later with reference to Figs. 4 to 8. Fig.

According to another embodiment, data may be transmitted and received to provide information in a voice or other manner instead of a page for visually providing information.

4 is a flowchart of an information providing process through an information providing interface according to an embodiment of the present invention.

In step 410, the control unit 320 of the information providing apparatus 300 generates interface page information. The interface page information is the information necessary to generate the interface page. The interface page is a page for guiding an input of a user, receiving a user's input, and transmitting the input to the information providing apparatus 300. For example, the interface page information may be in the form of an HTML document or other markup language document. In another embodiment, the terminal 200 has the format information of the interface page in advance, and only information corresponding to the content may be transmitted from the information providing apparatus 300 to the terminal 200. Hereinafter, it is assumed that the interface page information or other page information is transmitted in the HTML document format for convenience. However, the scope of the present invention is not limited thereto.

In operation 420, the communication unit 310 of the information providing apparatus 300 transmits the interface page information to the terminal 200.

In operation 430, the controller 250 of the terminal 200 configures the interface page using the interface page information. For example, the control unit 250 may operate a web browser to interpret an HTML document to form an interface page in the form of a web page. A separate application may be used instead of a web browser.

In step 440, the display unit 220 of the terminal 200 displays an interface page to the user 400. The interface page may include, for example, an interface through which the user 400 requests information, inputs and / or selects keywords for providing information, and makes settings for providing other information.

In operation 450, the input unit 210 of the terminal 200 receives the selection input of the user 400 input through the interface page and transmits the selection input to the controller 250.

In step 460, the communication unit 230 of the terminal 200 transmits input information for identifying the user's selection input to the information providing apparatus 300 under the control of the controller 250.

In step 470, the control unit 320 of the information providing apparatus 300 generates the result page information using the input of the user (for example, a keyword and / or other information providing setting). The preparation process for generating the result page information and the process of generating the result page information will be described later with reference to FIG. 5 to FIG. The result page information can be configured, for example, as an HTML document format and / or an image format.

In step 480, the communication unit 310 of the information providing apparatus 300 transmits the result page information to the terminal 200.

In step 490, the control unit 250 of the terminal 200 forms a result page using the result page information received by the communication unit 230. [ For example, the control unit 250 can interpret the result page information in the HTML format to form a result page.

In step 495, the display unit 220 of the terminal 200 provides the result page to the user 400.

In the embodiment of FIG. 4, it is assumed that a page of a visual form is provided to a user, but it is also possible to provide an interface or result information through voice. In this case, the audio output unit may be used instead of the display unit 220. [ Other methods of interfacing that may be presently or in the future instead of a visual / audible approach may be used in connection with the user 400. In this case, the information providing apparatus 300 may provide the terminal 200 with information converted in a different manner instead of the page information in accordance with the interface method.

5, the user 400 desires to receive information on an object of a specific area of interest to which the user 400 is interested. However, the scope of the present invention is not limited thereto.

The area of interest can be, for example, the type of the object. For example, if the area of interest is a 'celebrity,' the objects of interest may include 'Yushimin', 'Yoo Jae-seok', 'Taylor Swift', and so on. For example, if your interests are 'movies,' then those objects that interest you can include 'Dunkirk,' 'Spiderman: Homecoming,' and 'Super Bad 3.' For example, if the area of interest is a "program," the objects of interest may include "Infinite Challenge," "American Idol," "Game of the Throne," and so on.

In the following embodiments, documents are collected to evaluate the relationships (associativity / weight, etc.) between the keywords. Collected documents may all be evaluated as having equivalent value, and the latest documents may be evaluated as having higher value. That is, the age of the reference document at the evaluation time point and the relevance between the keywords appearing in the document may have a negative correlation.

The value can be evaluated differently according to the latestness of the document even in the process of FIG. 5 and later to be described later. For example, two keywords appearing in a document that is one day old at the evaluation time point may be evaluated as 10 times higher than when two keywords appear together in a document that is 10 days old at the evaluation time point. The age of the document can be evaluated, for example, in seconds / minutes / hours, or in days / months / years. The control unit 320 extracts a partial relevance degree reflecting the document age by dividing one partial relevance value of the document evaluated before the age of the document is reflected by the age of the document, Can be extracted.

The document creation time for identifying the age of the document can be determined, for example, by using the publication time and / or metadata included in the document, or by periodically crawling, a document that is not found in the previous crawl is newly found You may also find that a new document has been added at the time of the new crawl.

9 is a flowchart of an information providing process according to an embodiment of the present invention.

In step 910, the control unit 320 stores the object-keyword association diagram corresponding to the object attribute and the representative attribute keyword pair in the storage unit 330.

10 is an illustration of a stored object-keyword association diagram in accordance with one embodiment of the present disclosure.

In the embodiment of FIG. 10, there are m (i ₁ to i _m ) object items and n (k ₁ to k _n ) are all representative attribute keywords.

For example, the object-keyword association of the object item i ₅ and the representative attribute keyword k ₃ is w _5,3 .

The process of step 910 may be performed, for example, in accordance with some of the embodiments of FIGS. 5-8, or a similar process, or a corresponding process. According to another embodiment, the process of step 910 may be performed by inputting by an administrator or by receiving an object-keyword association determined by an external system through a network or a storage medium.

5 is a detailed flowchart of step 910 in accordance with one embodiment of the present disclosure.

Referring to FIG. 5, in step 510, the control unit 320 extracts a representative attribute keyword candidate set from the first set of documents. For example, the control unit 320 may collect keywords frequently appearing in a document corresponding to a field of interest among the first set of documents as a representative attribute keyword candidate set.

6 is a detailed flowchart of step 510 in accordance with one embodiment of the present disclosure.

The control unit 320 assigns keywords appearing in a document such as keywords appearing in a document such as an object keyword indicating an object item belonging to a specific field and a field keyword indicating a specific field to a first attribute keyword candidate set and a second attribute keyword candidate It can be set as a set.

For example, if you are interested in providing informational services, the domain keywords may include "celebrities," "entertainers," "movie stars," "stars," "selves," "celebs, The field keyword can be set by the administrator or recommended and set by the control unit 320. The control unit 320 may acquire some field keywords and recommend / set a similar keyword in which the degree of association with the keyword is greater than or equal to a predetermined value as an additional field keyword.

If the interest to provide the information service is a celebrity, the object keyword can be an individual person belonging to the interest field. For example, "Yoo Jae-suk", "Taylor Swift", "Stephanie Currie" and so on can be object keywords corresponding to interest celebrities.

To describe the relationship between the field keyword and the object keyword, for example, the field keyword may correspond to the attribute or type of the object keyword. If the field keyword represents a set, then the object keyword can represent the elements belonging to that set.

Object keywords can be set by an administrator or in a similar way to field keywords. According to another embodiment of the present invention, the controller 320 may analyze the context of the collected documents and may select, as object keywords, keywords judged as elements belonging to the set represented by the field keyword.

The popularity object keyword and the non-human object keyword can be classified according to the search / collection amount of the object keyword. The control unit 320 may search for / collect a document including each object keyword, set an object keyword whose collection amount is not less than a threshold value as a popular object keyword, and set the remaining object keywords as a non-target object keyword.

Keywords for popular keywords and non-keyword keywords can be classified according to the search / collection of keywords in the relevant field. The control unit 320 may search for / collect documents containing the keyword of each field, set a keyword of a category having a collection amount of a certain threshold value or more as a popular keyword, and set the keyword of the other keyword as a non- keyword. However, the threshold value for distinguishing the popular object keyword from the non-human object keyword may be different from the threshold value for distinguishing the popular keyword keyword and the non-keyword keyword. For convenience, the popular object keyword and the popular keyword are collectively referred to as a popular field object keyword. For convenience, the non-object keyword and the non-object keyword are collectively referred to as the non-object object keyword.

In an alternative embodiment, only the popular field keywords or popular object keywords may be used instead of the popular field object keywords. In an alternative embodiment, only the non-human field keyword or the non-human object keyword may be used instead of the non-human field object keyword.

In step 610, the control unit 320 sets keywords appearing together in a document such as a popular field object keyword as a first attribute keyword candidate set.

The control unit 320 can search / collect documents including the popular field object keyword and set the keywords included in the collected document as the first attribute keyword candidate set. According to another embodiment, the control unit 320 may exclude the field keyword and the object keyword among the keywords included in the collected document from the first attribute keyword candidate set. In addition, the control unit 320 may exclude keywords having no meaning set in advance, for example, survey / article, from the first attribute keyword candidate set. According to another embodiment, the controller 320 may include a keyword registered in a preset dictionary among the keywords included in the collected document in the first attribute keyword candidate set.

According to another embodiment, the control unit 320 searches / collects a document including a popular field object keyword, and searches for keywords in a collected document, which are located within a predetermined distance from a sentence containing the popular field object keyword or the keyword May be included in the first attribute keyword candidate set. According to another embodiment, the control unit 320 searches / collects documents containing popular field object keywords, analyzes the contexts, and assigns keywords used for expressing / describing popular field object keywords to first attribute keyword candidates It can also be included in a set.

The distance between keywords, or the distance between a keyword and a sentence, can be, for example, the number of sentences between two keywords or between a keyword and a sentence, the number of words located between two keywords or between a keyword and a sentence, The number of words to be located, the number of words between two keywords, or the number of characters between a keyword and a sentence.

The control unit 320 may first perform morpheme analysis for keyword analysis.

In step 620, the control unit 320 sets the keywords appearing together in the same document as the non-human field keyword as the second attribute keyword candidate set.

The control unit 320 may search / collect documents including non-human field keywords, and may set keywords included in the collected documents as second attribute keyword candidates. According to another embodiment, the control unit 320 may exclude the field keyword and the object keyword among the keywords included in the collected document from the second attribute keyword candidate set. In addition, the control unit 320 may exclude keywords having no meaning set in advance, for example, survey / article, from the second attribute keyword candidate set. According to another embodiment, the controller 320 may include a keyword registered in a preset dictionary among the keywords included in the collected document in the second attribute keyword candidate set.

According to another embodiment, the control unit 320 searches / collects a document including an unacceptable field object keyword, and extracts keywords that are located within a preset distance from a non-target field object keyword or a sentence including the keyword in the collected document And may be included in the second attribute keyword candidate set. According to another embodiment, the control unit 320 searches / collects a document including an object keyword in the non-object domain, analyzes the context, and assigns keywords used for expressing / describing the object keyword in the non-object domain to a second attribute keyword candidate It can also be included in a set.

The distance between keywords, or the distance between a keyword and a sentence, can be, for example, the number of sentences between two keywords or between a keyword and a sentence, the number of words located between two keywords or between a keyword and a sentence, The number of words to be located, the number of words between two keywords, or the number of characters between a keyword and a sentence.

The control unit 320 may first perform morpheme analysis for keyword analysis.

In step 630, the controller 320 may set keywords belonging to both the first attribute keyword candidate set and the second attribute keyword candidate set as the representative attribute keyword candidate set. That is, the keywords used for expressing the keyword of the popular field object and used for expressing the keyword of the non-human field object keyword can be collected in the representative attribute keyword candidate set.

According to another embodiment, in step 510, the control unit 320 may include the keywords appearing together with the object keyword and / or the field keyword in the representative attribute keyword candidate set, regardless of whether the popularity / non-popularity is determined.

Returning to FIG. 5, in step 520, the control unit 320 extracts two or more lower keywords associated with each representative attribute keyword included in the representative attribute keyword candidate set from the second set documents.

The second set document used for extracting the lower keyword in step 520 and the first set document used for extracting the representative attribute keyword candidate set in step 510 may be different document sets or may be the same document set. For example, the first set of documents may be a set that includes all of the collectable documents, and the second set of documents may be a set that includes only documents that are used as the main keywords of a particular area of interest to provide the information providing service . The control unit 320 analyzes the collectible documents and analyzes whether the specific interest field in which each document provides the information providing service is the main keyword based on frequently appearing keywords. According to another embodiment, both the first aggregation document and the second aggregation document may be a collection including all of the related documents that can be collected. According to another embodiment, the first aggregation document may be a set that includes all of the related documents that can be collected, and the second aggregation document may be a set that includes only documents related to a specific area of interest in which the information providing service is to be provided. According to another embodiment, the second aggregation document may be a set including all the related related documents that can be collected, and the first aggregation document may be a set that includes only documents related to a specific interest area in which the information providing service is to be provided.

For step 520, the control unit 320 determines whether the document containing the field keyword representing the interest field itself and / or the document containing the field of interest representing the interest field itself, for example, to generate a set containing only documents related to the particular field of interest, Or a document including an object keyword belonging to the interest field, and extracts documents having a proportion of a field keyword / object keyword of a predetermined value or more among them to generate a set including only documents related to a specific field of interest. The weight of the field keyword / object keyword can be judged through the appearance frequency, appearance position, and context of the field keyword / object keyword. For example, a document in which a field keyword / object keyword appears frequently, a field keyword / object keyword is used as a title of the document, or a letter in a form for large letters or emphasis is classified as a document related to a specific field of interest.

In step 520, the control unit 320 analyzes at least a part of the second set of documents, extracts a predetermined number of sub-keywords having the highest relevance with each representative attribute keyword, and extracts two or more sub- Keywords can be extracted.

The controller 320 may determine the degree of association between the representative attribute keyword and the lower keyword in consideration of, for example, the frequency with which the lower keyword appears in the same or similar context as the representative attribute keyword. For example, words appearing in the vicinity of keyword A in a certain sentence may appear in the vicinity of the word associated with keyword A in another document.

"I went on a big trip, but it was July, and the weather was so hot that I suffered."

"I went on a big trip, but it was July, so the weather was too humid."

Looking at the two sentences above, the word "hot" was replaced by the word "wet" in the same context. The control unit 320 can infer that "hot" and "wet" are related words.

"I went on a big trip, but it was July, and the weather was so hot that I suffered."

"I had a great vacation and went on vacation, but it was July, so I had a hard time because the weather was so hot."

Similarly, the control unit 320 can infer that the words " travel " and " vacation " are related words through the above two sentences.

"I went on a big trip, but it was July, and the weather was so hot that I suffered."

"I went on a big trip, but it was August and the weather was so hot that I suffered."

Similarly, the control unit 320 can infer that the words " July " and " August " are associated with each other through the above two sentences.

The control unit 320 is a word associated with "hot" and "wet" through previously collected documents, and a word associated with "July" and "August", and a word associated with "travel" and "vacation" It is assumed that the following sentence is collected.

"I had a great vacation and went on a vacation, but it was July, and the weather was too humid."

"I went on a big trip, but it was hard because it was too hot in August."

If the two sentences are not the same context but know that "hot" and "humid" are related words, "July" and "August" are related words, and "travel" ) Will be able to learn that "suffering" and "hardship" are also related words through the above sentence.

It is possible to determine that a keyword pair having a high frequency of appearing in the same / similar context has a high degree of correlation with each other. In addition, the higher the degree of similarity of the context in which the two keywords appear, the higher the degree of association between the two keywords can be judged. The control unit 320 performs learning using the collected documents to set the degree of association between the keywords and sets the degree of association of the keywords appearing in the corresponding sentence by using the relation between the set keywords and the context of the sentence. It is possible to improve the accuracy of the determination of the inter-linkage.

Similar learning methods are known, such as Neural Net Language Model (NNLM), Recurrent Neural Net Language Model (RNNLM), word2vec, skipgram, and Continuous Bag-of-Word (CBOW) schemes. In particular, when word2vec is used, word2vec learns by using documents and maps each keyword to a vector, and the similarity between two keywords can be grasped through calculation of the cosine similarity of two vectors.

In this manner or the like, the control unit 320 may analyze at least a part of the second set of documents and extract a predetermined number of the sub-keywords having the highest relevance with each representative attribute keyword.

In step 530, the controller 320 may extract association weight corresponding to each representative attribute keyword and a lower keyword pair in the representative attribute keyword candidate set from the second set documents.

7 is a detailed flowchart of step 530 according to one embodiment of the present disclosure.

In step 710, the control unit 320 may analyze at least a part of the second set of documents to extract the degree of correlation between the sub-keywords. For example, it is assumed that the keyword of the 50 sub-B1 ₁ to B1 ₅₀ collected as the associated sub-keyword for representing attribute keywords A1. In this case, the controller 320 can extract the degree of correlation between the two lower keywords using the frequencies of the two lower keywords appearing in the same document for the 50 lower keywords. B1 is also interconnected between the _first B1 and B1 ₂ and B1 is determined in accordance with the _first frequency to the _second appearing in the same document. According to another aspect of the B1 and B1 _{1 2} In addition to the frequency of appearance in the same document affecting the correlation, if B1 ₁ and B1 ₂ appear in the same document, the distance between the two keywords B1 ₁ and B1 ₂ (or the distance between the sentences where the two keywords appear) ), The higher the degree of correlation can be recognized. In a similar manner, the correlation between sub-keywords can be extracted. The distance between keywords, or the distance between a keyword and a sentence, can be, for example, the number of sentences between two keywords or between a keyword and a sentence, the number of words located between two keywords or between a keyword and a sentence, The number of words to be located, the number of words to be located, and the number of characters located between two keywords or between a keyword and a sentence.

In step 720, the control unit 320 may extract association weights between the representative attribute keywords and the sub-keywords based on the degree of correlation between the lower keywords. For example, the control unit 320 determines, for a set of lower keywords corresponding to each representative attribute keyword, a correlation between a specific lower keyword in the lower keyword set and another lower keyword in the lower keyword set, The association weights between the specific sub keyword and the representative attribute keyword can be set so that the association weights between the representative attribute keywords have a positive correlation with each other.

For example, the higher the degree of correlation between the lower keyword B1 ₁ of the representative attribute keyword A1 and the other lower keywords B1 ₂ to B1 ₅₀ of A1, the higher the association weight between A1 and B1 ₁ can be set. For example, the other sub-keyword of ₁ B1 and A1 (B1 B1 ₂ to ₅₀₎ is also the arithmetic mean (or total) correlation between this association may be a weight between ₁ and B1 A1. Instead of simple arithmetic mean, geometric mean / harmonic mean may be used. It is also possible to use a cutting average that averages except for the two highest (example) and lowest (example) interrelation between the sub keyword B1 ₁ and the other sub keywords (B1 ₂ to B ₁₅₀ ) of A1 have. The median may be used instead of the arithmetic mean of the correlation.

According to some embodiments, the "B1 B1 ₁ and ₂ are used to calculate the associated weight for the ₁ B1 for A1 Frequently appearing in the same document "is simply B1 ₁ and B1 ₂ It does not depend on the number of documents that appear (or appear in the same sentence or appear in the same sentence), but B1 ₁ and B1 ₂ The number of documents that appear together (or appear in the same sentence or appear in close proximity) can be found by dividing by the number of documents in which B1 ₁ appears and / or the number of documents in which B1 ₂ appears. In a similar manner "B1 ₁ and B1 ₂ Frequently appearing in the same document "is B1 ₁ and B1 ₂ It appeared to as (to appear or are in the same sentence, or close to the emergence of) correlation of the number and the notes of the document to the number and the positive correlation between the B1 ₁ appeared number of documents and / or to have B1 ₂ appeared in the document Can be set to have a relationship. It is a kind of normalization to prevent a simple commonly used word from having a high associative weight on the representative attribute keyword A1.

Returning to FIG. 5, in step 540, the control unit 320 may extract a lower degree of association between the object item and the lower keyword from the first set of documents.

A document such as an object keyword (e.g., " Taylor Swift ") representing an object item among the first set documents, or a lower keyword frequently appearing in the same sentence or a nearby sentence, can be determined to be associated with the object item. The control unit 320 may collect the document in which the object keyword of the corresponding object item is present and extract the lower degree of association between the lower keyword and the object keyword according to the frequency of coexistence in the documents. In particular, when the lower keyword appears in the same sentence as the object keyword, the control unit 320 can set the higher degree of association between the lower keyword and the object item than when the lower keyword appears in a sentence different from the object keyword.

The control unit 320 can set the higher the degree of association between the lower keyword and the object item of the object keyword as the sentence in which the lower keyword appears comes closer to the sentence in which the object keyword appears. The degree of proximity of two sentences can be, for example, the number of sentences located between two sentences, the number of words located between two sentences, the number of words located between two sentences, and the number of characters located between two sentences Can be judged on the basis of one or more.

The control unit 320 can set the higher the degree of association between the lower keyword and the object item of the object keyword as the position where the lower keyword appears is closer to the position where the object keyword appears. The degree of proximity between the lower keyword and the object keyword is determined by, for example, the number of sentences located between the lower keyword and the object keyword, the number of words located between the lower keyword and the object keyword, Number of characters, number of characters located between the lower keyword and the object keyword, and the like.

In step 550, the controller 320 may extract an object-keyword association between the object item and each representative attribute keyword using the lower association diagram of step 540 and the association weights of step 530.

For example, the object-keyword association between the object item C and the representative attribute keyword A1 may be determined by using the sub-associations between C and A1 lower keywords (for example, B1 ₁ to B1 ₅₀ ) and the association weights of the respective lower keywords . For example, the object-keyword association between the object item C and the representative attribute keyword A1 may be set to be higher as the lower degree of association between the object item C and B1 ₁ to B1 ₅₀ is higher.

In addition, when the subordinate association degree with respect to the object item C is high for the subordinate keyword having a higher association weight in relation to A1, the relation between the object C and the representative attribute keyword A1 May be set to a higher value. For example, in Table 1, since the lower associativity of the higher correlation weight (B1 ₁ ) is higher than that of Table 2, the case of Table 1 is larger than that of the object C between the object C and the representative attribute keyword A1, The keyword association can be set high.

Association weights with A1 Sub-associativity with C B1 ₁ 0.5 0.5 B1 ₂ 0.2 0.2

Association weights with A1 Sub-associativity with C B1 ₁ 0.2 0.5 B1 ₂ 0.5 0.2

According to an exemplary embodiment, the object-keyword association between the object C and the representative attribute keyword A1 may be obtained by summing (or using the sum) the values obtained by multiplying the association weights corresponding to the respective lower keywords and the lower association degrees. 0.5 × 0.5 + 0.2 × 0.2 = 0.29 in Table 1, and 0.2 × 0.5 + 0.5 × 0.2 = 0.20 in Table 2, so that the case of Table 1 is between the object C and the representative attribute keyword A1 The object-keyword association degree of the object-keyword can be set high.

The above-described object-keyword association calculation method is merely an example. If the association weights between A1 and A5 of step 540 and the object-keyword association between A1 and A1 are positive Other methods may be used.

When the communication unit 310 receives the information providing request associated with the specific representative attribute keyword, the control unit 320 can provide the result item through the communication unit 310 based on the object-keyword association extracted in operation 550. For example, when an information provision request including any one representative attribute keyword is received, the control unit 320 can provide information on the object item in the order of highest object-keyword association in relation to the representative attribute keyword have.

In another embodiment, upon receiving an information provision request including two or more representative attribute keywords and corresponding weight values, the control unit 320 determines an object-keyword association with representative attribute keywords included in the information provision request for each object item Information about an object item can be provided in the order in which the sum (or average) of values obtained by multiplying a figure by a weight (or similarly weighted addition) is highest.

8 is a flowchart of an information providing process according to another embodiment of the present invention.

The embodiment of FIG. 8 further includes two steps 523 and 526 between steps 520 and 530, in addition to the same procedures as the embodiment of FIG. Here, duplicate explanation is avoided and only steps 523 and 526 are described.

In step 523, the control unit 320 determines whether each of the lower keywords extracted in step 520 corresponds to the emotion language (emotion word). For this, the storage unit 330 or an external server may have a dictionary of emotions. The speech dictionary is a tool for determining whether a word (keyword) is an emotional word, for example, a word list. The keyword included in the evaluation word list can be said to be an emotional word, and the other keywords can be judged not to be an emotional word. However, these judgments are based on lexical meaning and may not reflect the popular word usage that varies with the times. Accordingly, the control unit 320 determines whether to use the representative attribute keyword based on whether or not the emotion word of the lower keyword associated with the representative attribute keyword is used, without determining whether the representative attribute keyword itself is the emotion word.

In another embodiment, the control unit 320 may add another word learned to be associated with a predetermined number of words or more (higher than a predetermined value), which is registered as an emotion word in the emotion word dictionary, to the emotion word dictionary.

In step 526, the control unit 320 may leave only a predetermined number of representative attribute keywords in the representative attribute keyword candidate set in descending order of the emotional language ratio (or number) of the associated sub keywords, and remove the rest. Through this process, it is possible to prevent the keyword, which is distant from the emotional language, from being treated as an emotional language.

Returning to FIG. 9, in step 920, the control unit 320 stores the basic word-keyword associations corresponding to the reserved words and representative keyword pairs in the storage unit 330. FIG.

The reserved word may include expressions that can be expressed by weights of representative attribute keywords. For example, "boredom" can be a reserved word, and "pretty" can be a reserved word.

The representative attribute keyword having a basic reserved word-keyword association with respect to the reserved word " boredom " may include representative attribute keywords capable of solving a bored situation such as "interesting", "exciting", " .

Basic reserved word for the reserved word "pretty" - The representative attribute keyword having a high degree of keyword association is a representative attribute describing similar or "pretty" to "pretty" such as "beautiful", "cute", "eye- Keywords.

The process of step 920 may be performed, for example, by an input of an administrator or by receiving a basic keyword-keyword association determined by an external system through a network or a storage medium. According to another embodiment, the process of step 920 may be performed in a manner similar to that of FIGS. 5 to 8 by analyzing collectable documents such as the Internet, SNS, news, and the like. In addition, the process of step 920 may include a process of reflecting user's feedback as described later.

The process of step 920 may be performed in a manner to be described later with reference to any one of Figs.

11 is an illustration of a basic reserved word-keyword association according to one embodiment of the present disclosure.

In the embodiment of FIG. 11, there are q words (C ₁ to C _q ) of all reserved words and n (k ₁ to k _n ) of representative keyword keywords.

For example, the basic word-keyword association of the reserved word C ₅ and the representative attribute keyword k ₃ is v _3,5 .

In step 930, the communication unit 310 receives and acquires a receiving reservation word from the terminal 200, and transmits the receiving reservation word to the control unit 320. [

The reception reservation word is a reserved word received by the terminal 200 from the search user. The terminal 200 may convert the voice input into an electrical signal (voice signal) and transmit it to the search device 300. The control unit 320 of the search apparatus 300 may analyze the voice signal, convert the voice signal into text, and match the converted text with a reserved word. The control unit 320 may also analyze the voice signal and analyze the voice intonation, the height, the speed, the breathing state, and the like to use the information as context information.

According to another embodiment, the terminal 200 may convert the voice input into text and transmit the converted text to the search device 300. The terminal 200 can analyze the received voice input, the height, the speed, the breathing state, and the like and transmit the analysis information to the search device 300. The search device 300 may utilize the analysis information as a kind of context information.

In step 940, the control unit 320 obtains a reserved word-object association map corresponding to the received reservation word and each object item pair using the object-keyword association map and the basic reserved word-keyword association map.

12 is a detailed flowchart of step 940 in accordance with one embodiment of the present disclosure.

Referring to FIG. 12, in step 1210, the control unit 320 obtains an adjustment object-keyword association corresponding to each object item and a representative attribute keyword pair with respect to a received reservation word.

According to one embodiment, the control unit 320 assigns object-keyword associativity corresponding to the object item and the representative attribute keyword pair to each of the object item and the representative attribute keyword pair, Keyword associations corresponding to the respective object items and the representative attribute keyword pairs for the reception reservation word can be obtained by applying the basic reserved word-keyword association.

In particular, the control unit 320 assigns, to each object item and representative attribute keyword pair, an object-keyword association corresponding to the object item and the representative attribute keyword pair based on the reception reservation word and a basic reserved word keyword corresponding to the representative attribute keyword pair Keyword association degree corresponding to each object item and the representative attribute keyword pair with respect to the reception reservation word can be obtained by using a value obtained by multiplying the correlation value by the correlation value.

Also, the controller 320 has a positive correlation with the object-keyword association corresponding to the object item and the representative attribute keyword pair with respect to each object item and the representative attribute keyword pair, Keyword associations corresponding to the respective object items and representative attribute keyword pairs for the reception reservation word so as to have a positive correlation with the corresponding basic reserved word-keyword association.

In this specification, it is assumed that the values of object-keyword association, basic reserved word-keyword association, coordination object-keyword association, basic reserved word-lower keyword association and other association relations are more closely related to each other . In other embodiments, some association values are more closely related to each other as the association values become smaller and the association values become larger as the association values become larger. May be used instead.

For example, in order to obtain an adjustment object-keyword association corresponding to the object item i ₄ and the representative attribute keyword k ₃ pair when the reception reservation word is C ₂ , the control unit 320 sets the pair of the object item i ₄ and the representative attribute keyword k ₃ Keyword association can be obtained by applying the basic reserved word-keyword association v ₃ , ₄ corresponding to the pair of the reception reservation word C ₂ and the representative attribute keyword k ₃ to the object-association map w ₄ , .

In particular, the method of applying the association may be a method of multiplying the association between the object-association and the basic reserved word-keyword. For example, in order to obtain an adjustment object-keyword association corresponding to the object item i ₄ and the representative attribute keyword k ₃ pair when the reception reservation word is C ₂ , the control unit 320 sets the pair of the object item i ₄ and the representative attribute keyword k ₃ (W _{4 , 3} Хv _{3, 2} ) which is a value obtained by multiplying the object-association degree w _4,3 corresponding to the received reserved word C ₂ and the representative attribute keyword k ₃ corresponding to the basic reserved word-keyword association v _3,2 To obtain the adjusted object-keyword association. In another embodiment, the control unit 320 may use the function f (w _4,3 (x, y)), which applies another operation / utilization scheme such that the coordination object-keyword association has a positive correlation with w _4,3 and v _3,2 instead of the multiplication , v _{3, 2} ) may be used to obtain the adjusted object-keyword association. In addition (w _4,3 Хv _3,2) to adjust the object - how to utilize as the keyword association, and (w _4,3 Хv _{3, 2)} applying a correction based on other factors in the adjustment object - to utilize keywords as Relevancy All of the methods can be utilized.

를 이용하여 획득될 수 있다. f(w_4,j,v_j,2)는 객체 항목 i₄ 및 수신 예약어 C₂, 키워드 k_j에 대응하는 조정 객체-키워드 연관도이다.In step 1220, the control unit 320 may obtain the reserved word-object association using the accumulated value of the adjustment object-keyword association for the specific object item. For example, the control unit 320 may associate the reserved word-object association diagram corresponding to the reception reservation word and the specific object item pair so as to have a positive correlation with the accumulated value of the adjustment object- Can be set. The reserved word-object association diagram corresponding to the pair of the object item i ₄ and the reception reservation word C ₂ is, for example,

. ≪ / RTI > f (w _{4, j} , v _{j, 2} ) is an adjustment object-keyword association diagram corresponding to the object item i ₄ , the reception reservation word C ₂ , and the keyword k _j .

이 될 수 있다. 다른 예에서 객체 항목 i₄ 및 수신 예약어 C₂ 쌍에 대응하는 예약어-객체 연관도는

에 대해 기타 요인에 의한 보정을 가한 값이 될 수 있다.For example, the reserved word-object association diagram corresponding to the pair of object item i ₄ and reception reservation word C ₂

. In another example, the reserved word-object association diagram corresponding to the pair of object item i ₄ and reception reservation word C ₂

And the correction by other factors.

Returning to FIG. 9, in step 950, the control unit 320 may provide the object item according to the reserved word-object association map corresponding to the received reservation word. For example, if the reserved word-object association corresponding to the reception reservation word C ₂ is as shown in Table 3 below, the control unit 320 can provide the object item in the order of Table 4. [

Object item Reserved words with recipient reserved words - Object association diagram i ₁ 0.23 i ₂ 0.33 i ₃ 0.99 i ₄ 0.84

order Object item Reserved words with recipient reserved words - Object association diagram One i ₃ 0.99 2 i ₄ 0.84 3 i ₂ 0.33 4 i ₁ 0.23

That is, the control unit 320 can provide the object items in the descending order of the reserved word-object associations corresponding to the reserved reserved words. The terminal 200 provided with the object item can provide information about the object item i ₃ to the user through the display unit 220. The terminal 200 may also provide information on other object items in descending order as needed. The terminal 200 can provide the user with information about the object item i ₃ by voice through the speaker instead of the display unit 220.

13 is a flowchart of an information providing process according to another embodiment of the present invention.

13 to 17 may be performed by using some of the processes of FIGS. 5 to 12 or by partially modifying the processes of FIGS. 5 to 12. FIG. The description of FIGS. 5 to 12 can be used when necessary while explaining the processes of FIGS. 13 to 17. FIG.

Referring to FIG. 13, in step 1310, the control unit 320 extracts a representative attribute keyword candidate set from the first set documents. For example, the control unit 320 may collect keywords frequently appearing in a document corresponding to a field of interest among the first set of documents as a representative attribute keyword candidate set. The process of step 1310 may proceed in the same or similar process as, for example, step 510 of FIG. The process of step 1310 may be performed through the same or similar process as the process of FIG. The description of the process of FIG. 6 is not repeated.

In step 1320, the control unit 320 sets a reserved word set. For example, a set of reserved words may be set by the administrator manually. According to the modified example, the control unit 320 may set an appropriate word / phrase as a reserved word as a reserved word candidate and provide an interface by which one or more of the reserved word candidates can be set as a reserved word.

18 is a detailed flowchart of step 1320 according to a modification of the present disclosure.

In step 1810, the control unit 320 obtains the number of occurrences in the document set of one language unit or two or more consecutive language units. Here, the document set may be the same document set as the document set used in step 510 of FIG. 5, or may be another document set.

The language unit may be, for example, a word / word / morpheme / syllable / letter. A unit divided into various criteria for distinguishing other sentences can be the language unit in this embodiment.

The control unit 320 may divide the documents included in the respective documents of the document set before the step 1810 into units of words and store them in the form of an array or a list. The control unit 320 deletes words that are not meaningful in each word, for example, some of the surveys in the Korean language, the instruction adjectives such as 'I' and 'Low', and words not necessary for other analysis, Can be removed. In addition, the control unit 320 may convert the word into a basic type (or a preset type) when the word is composed of one word according to the embodiment.

According to the modified example, the control unit 320 may divide the documents included in each document of the document set before the step 1810 into words, and store them in the form of an array, a list, or the like. The control unit 320 can convert each word into a basic type (or a preset type) according to an embodiment. The control unit 320 may remove words that are not meaningful according to the embodiment, for example, some of the surveys in Korean, directive adjectives such as 'i' and 'low', and words not necessary for analysis.

It is also possible that the control unit 320 divides the document into morpheme units, syllable units, or character units.

In the following embodiments, it is assumed that the control unit 320 divides a document into units of words, for convenience, and each word is a language unit.

A single language unit can be a reserved word. According to an alternative embodiment, two or more consecutive language units may also be reserved words. For example, "clean" (a single language unit) can be a reserved word, and "good mood" (two consecutive language units) can be a reserved word. However, since two or more consecutive language units are frequently unavailable compared to a single language unit, in order to select two or more consecutive language units as reserved words according to the embodiment, a weight or an additional score . According to the modified example, a reference value for selecting a reserved word can be set more generously for two or more consecutive language units when a reserved word is selected. For example, if a single language unit can be recommended as a reserved word candidate by appearing at least a times, two consecutive language units can be set to be recommended as reserved word candidates even if the number of times b, which is much less than a, The language unit may be set to be recommended as a reserved word candidate even if only c times less than b appears. Hereinafter, two or more consecutive language units are referred to as consecutive language units.

Also, when consecutive language units are recommended / selected as reserved words, consecutive language units included in the language unit included in the reserved words (candidates) or reserved words (candidates) and shorter than the reserved words (candidates) are not recommended as reserved words, You can deduct points when you score points. This is to prevent similar reserved words from being selected or recommended. Hereinafter, in order to simplify the description, the description of the continuous language unit is omitted, but the description of the single language unit can be equally or similarly applied to the continuous language unit.

In step 1810, the number of occurrences of the language unit may be, for example, the number of documents in which the corresponding language unit appears. Even if a language unit appears multiple times in a document, the number of occurrences is recognized only once. According to another embodiment, the case where the corresponding language unit appears plural times in one document is regarded as the number of occurrences, and the number of occurrences can be the number of occurrences of the language unit.

According to another embodiment, when a language unit repeatedly appears in a document repeatedly two or more times, the second and subsequent occurrences can be recognized as lower values than the first appearances. Also, if repeatedly appearing in the same language unit in a document, the later appearance can be recognized as a lower score. As the number of appearances increases, the score increases, but the gradient becomes gradual. For example, the number of occurrences of 1 / r squared (r is a real number greater than 1) may be used as an entry point for that language unit in the document. For example, the log value of the number of appearances may be +1 (if the number of occurrences is 0, the appearance score is 0) or the like may be used. Also, the appearance score of each language unit in a document may be limited to not exceed a preset upper limit value. The accumulated value of the appearance score of the corresponding language unit with respect to all the documents may be the appearance score according to the appearance frequency of the corresponding language unit. Such an appearance score can be utilized in step 1830.

For the sake of convenience, the number of occurrences of a language unit is assumed to be the number of documents in which a corresponding language unit appears.

In step 1820, the control unit 320 obtains the number of times the emotion word is located within a predetermined distance from the language unit. The distance between the language unit and the emotion word may be, for example, the number of words located between the language unit and the emotion word, the number of words located between the language unit and the emotion word, the number of characters located between the language unit and the emotion word It can be judged based on any one or more.

Also, when the language unit and the emotion word belong to different sentences, the control unit 320 determines that the emotion word is not located within a predetermined distance from the language unit regardless of the word / word / character positions located between the language unit and the emotion word It is possible. According to another modified example, when the language unit and the emotion word belong to different sentences, the control unit 320 may calculate distances by adding a certain number to the distance calculated by excluding judgment on the sentence. When the language unit and the emotional word belong to different sentences, it is preferable to evaluate the distance more than the number of words / phrases / characters located between the emotional words because the probability that they are not related to each other is relatively high.

Whether or not an emotion word of a specific word (a word phrase) can be confirmed by inquiring in a registered emotion word dictionary.

The number of times the emotion word is located within a certain distance from the language unit may be, for example, the number of documents in which the language unit and the emoticon are located together within a predetermined distance. Even if a language unit and a sentence appear in a document several times within a certain distance, the number of appearances can be recognized only once. According to another embodiment, when there are a plurality of language units in which a language is located within a certain distance of a language unit of a document, all the language units can be recognized as the number of positions where the language is located within a certain distance from the language unit. Hereinafter, a language unit in which an emotion word is located within a certain distance is referred to as an emotion word position language unit.

According to another embodiment, when a language unit of a sentence is repeated twice or more in a document, the appearance after the second can be recognized as the appearance of a lower value than the first appearance. In addition, the repetition of the appearance of the corresponding unit of language in a document can be recognized as a lower score as the repetition is repeated. As the number of appearances increases, the score increases, but the gradient becomes gradual. For example, 1 / r squared (r is a real number greater than 1) of the number of occurrences may be used as the appearance score of the corresponding emotional word position language unit in the document. For example, a log value of the number of appearances + 1 (where the appearance score is 0 if the number of occurrences is 0) may be used. In addition, the appearance score of the emotion word position language unit in one document may be limited not to exceed a preset upper limit value. The accumulated value of the appearance score of the corresponding language position language unit for all the documents will be the appearance score according to the appearance frequency of the corresponding language unit. And such an appearance score may be utilized at step 1830. [

According to another embodiment, the more the number of emotion words located within a certain distance with respect to one language unit, the higher the appearance score may be. In addition, if there is an emotional word within a shorter distance for a single language unit, the higher the score of appearance can be. In addition, an appearance score (the number of appearances) may be recognized only when there are two or more predetermined number of emotion words for one language unit.

Hereinafter, it is assumed that the number of occurrences of the unit of language in the sentence position language is the number of documents in which the sentence has appeared within a predetermined distance from the corresponding language unit.

In step 1830, the control unit 320 sets a reserved word candidate in consideration of the appearance frequency of the language unit and the number of times the emotion word is located within a certain distance from the language unit.

For example, the control unit 320 may multiply the number of occurrences of the language unit and the number of times the emotion word is located within a certain distance (or according to a calculation method having a positive correlation with respect to the other two variables) as the emotion word score. And a predetermined number of preset words can be set as reserved word candidates in descending order of scores. Alternatively, a language unit corresponding to a predetermined score or more may be set as a reserved word candidate.

Language unit Number of appearances Number of positions score First language unit 3003 1122 3369366 Second language unit 2001 1820 3641820 Third language unit 3121 1300 4057300 Fourth language unit 200 110 22000

In the example of Table 5, a reserved word candidate can be a third language unit -> a second language unit -> a first language unit -> a fourth language unit sequence. If the control unit 320 recommends two reserved word candidates, the third language unit and the second language unit will be recommended. The third language unit, the second language unit, and the first language unit will be recommended as reserved word candidates in order if the control unit 320 recommends a language unit of 3 million or more points as a reserved word candidate.

According to another embodiment, the control unit 320 extracts a first number of language units in order of the number of times of language unit appearance, and then, based on the number of times the emotion word appears (or the score of the emotion word) within a certain distance of the extracted language units A predetermined number of reserved word candidates can be extracted in one sequence. In the example of Table 5, if three language units are extracted in order of appearance frequency, the first to third language units can be extracted. Based on the number of appearance of emotion words, a reserved word candidate can be recommended in order of a second language unit, a third language unit, and a first language unit.

According to another embodiment, the control unit 320 extracts a first number of language units in the order of the number of appearance of the language unit, and then sets the number of times (or the number of the words of the sentence) appearing in the predetermined distance of the extracted language units The second number (the second number is less than the first number) can be extracted. In the example of Table 5, if three language units are extracted in order of appearance frequency, the first to third language units can be extracted. If two language units are extracted based on the number of emotion words, a reserved word candidate may be recommended in order of the second language unit and the third language unit.

The control unit 320 determines whether the score for the reserved word candidate selection is positively correlated with the number of appearances of the language unit of the step 1810 and the emotional word is located within a predetermined distance of the step 1820 in a manner similar or somewhat different from the above- A score for selecting a reserved word can be set so as to have a positive correlation with the number of occurrences of the language unit, and the reserved word candidate can be recommended using the score.

In addition, the control unit 320 may process the language unit already included in the reserved word set so as not to be added as a reserved word candidate. In addition, the control unit 320 may process the language unit substantially the same as the reserved words included in the reserved word set so as not to be added as a reserved word candidate.

In step 1840, the information providing apparatus 300 provides the terminal 200 with interface information for generating a reserved word selection interface including the reserved word candidate information. The interface information can be, for example, an HTML document. According to another embodiment, the interface information includes only dynamic information (recommended reserved word candidates, etc.) necessary for generating an interface, and the terminal 200 reflects this dynamic information to the page type previously stored in the terminal 200 To provide the user with a page including the interface.

The control unit 320 may generate page information for generating a page including a reserved word selection interface including the reserved word candidate information and the communication unit 310 may provide page information to the terminal 200. [ The terminal 200 may render a page including the corresponding interface to the user and display the rendered page. According to an alternative embodiment, an interface by sound instead of an interface included in the visual page, or an interface by a technology known to the future or known in the future may be provided. For the sake of simplicity, it is assumed that the interface included in the visual page is provided below.

FIG. 19 is an illustration of interface 1900 generated in accordance with the interface information provided in step 1840.

Referring to FIG. 19, the interface 1900 includes a check box column 1910, a reserved word candidate column 1920, and a detail view column 1930. The interface 1900 may also include a reserved word addition button 1940, a candidate deletion button 1950, and a store button 1960. The user selects a desired reserved word candidate (s) from the check box column 1910 and selects either the reserve word add button 1940, the candidate delete button 1950, or the store button 1960 to process reserved word candidates . When one of the buttons is selected, the terminal 200 can transmit the input information converted by the user to the information providing apparatus 300.

The information providing apparatus 300 can process a reserved word candidate according to input information received from the terminal 200. [ For example, if the user selects a check box 1910 of a reserved word candidate (hereinafter, 'selected candidate') and selects a reserved word adding button 1940, the control unit of the information providing apparatus 300 320 may add selected candidates to the reserved word set and delete selected candidates from the reserved word candidate set. The control unit 320 controls not to recommend the language unit (s) included in the reserved word set and the language unit (s) substantially identical to the language unit (s) included in the reserved word set as reserved word candidates in the future.

According to another example, when the user selects a check box 1910 of a certain reserved word candidate and selects a reserved word candidate deletion button 1950, the control unit 320 of the information providing apparatus 300, which receives the input information related thereto, Add the selected candidates to the set and delete the selected candidates from the reserved word candidate set. The control unit 320 controls the language unit (s) included in the reserved word excluded set and the language unit (s) substantially included in the reserved word excluded set to be not recommended as reserved word candidates do.

According to another example, when the user selects a check box 1910 of a certain reserved word candidate and selects a put in storage box 1960, the control unit 320 of the information providing apparatus 300, which receives the input information related thereto, The selected candidates may be added to the candidate storage set and the candidates selected from the reserved word candidate set may be deleted. The control unit 320 does not recommend the language unit (s) included in the reserved word candidate storage set and the language unit (s) substantially the same as the reserved language candidate (s) included in the reserved word candidate storage set .

Other interfaces may be used instead of the buttons 1940, 1950, and 1960, which may take the place of other interfaces or buttons similar to buttons.

In addition, the control unit 320 may provide an interface for partially deleting a reserved word from the reserved word set. The control unit 320 may provide an interface for removing the language unit from the reserved word exclusion set so that some language units of the reserved word exclusion set are not excluded from the recommendation. The control unit 320 may provide the language unit of the reserved word candidate storage set in the form of a list interface similar to that of FIG. 19, and allow the user to add a partial language unit of the reserved word candidate storage set as a reserved word. In addition, a user may delete some of the language units of the reserved word candidate storage set from the reserved word candidate storage set so as to be included in the reserved word elimination set through the list interface. In this case, the language unit is no longer provided through the list interface of the reserved word candidate archive set, nor is it recommended as a reserved word candidate through the interface 1900 of FIG. Also, the user may delete some language units of the reserved word candidate storage set from the reserved word candidate storage set through the list interface. In this case, the language unit is no longer provided through the list interface of the reserved word candidate archive set, but it can be recommended as a reserved word candidate through the interface 1900 of FIG.

In addition, the interface 1900 may include a previous page button 1970 and / or a next page button 1980 for page movement in case the reserved word candidate can not be displayed on one page. The previous page button 1970 and / or the next page button 1980 may be optionally provided according to the number of actual candidates and the current page position. An interface may also be provided that expands through scrolling instead of the previous page button 1970 and / or next page button 1980. [ In some interfaces, only the table containing entries 1910, 1920 and 1930 may be scrolled and the buttons 1940, 1950, 1960, 1970, 1980 may be excluded from scrolling.

The user may select the detailed view 1930 to inquire in detail the background or related information for which the reserved word candidate is recommended. The interface provided by the control unit 320 when the detailed view 1930 is selected includes an interface for adding information about the reserved word candidate and the reserved word candidate to a reserved word or adding it to a reserved word candidate storage set or adding it to a reserved word excluding set can do.

An interface for managing other reserved word candidates may be provided to the user through the terminal 200 by the control unit 320. [

Returning to FIG. 18, in step 1850, the control unit 320 may add the selected reserved word candidate to the reserved word set according to an input for selecting a reserved word.

Returning to FIG. 13, in step 1330, the control unit 320 stores the object-keyword association diagram corresponding to the object item and the representative attribute keyword pair.

10 is an illustration of a stored object-keyword association diagram in accordance with one embodiment of the present disclosure.

In the embodiment of FIG. 10, there are m (i ₁ to i _m ) object items and n (k ₁ to k _n ) are all representative attribute keywords.

For example, the object-keyword association of the object item i ₅ and the representative attribute keyword k ₃ is w _5,3 .

The process of step 1330 may be performed, for example, according to some of the embodiments of FIGS. 5 to 8, or a similar process, or a corresponding process. According to another embodiment, the process of step 1330 may be performed by inputting by an administrator or by receiving an object-keyword association determined by an external system through a network or a storage medium.

Since the embodiments of Figs. 5 to 8 have been described above, repeated description will be omitted. However, the process performed in step 510 of FIGS. 5 and 8 is substantially the same as the process of step 1310 of FIG. Therefore, even if the embodiments of FIGS. 5 to 8 are utilized in performing the process of step 1330, the process of step 510 is not performed again and the result of step 1310 can be recycled.

In step 1340, the control unit 320 generates a basic reserved word corresponding to the reserved word and the representative attribute keyword pair using the associative weight corresponding to the representative attribute keyword and the lower keyword pair, The keyword association is stored in the storage unit 330. The process of step 1340 may be performed, for example, according to an input of an administrator, or may be performed according to any one or more of the embodiments of Figs.

Prior to the process of step 1340 or during the process of step 1340, the lower keywords are determined, the association weights corresponding to the representative attribute keywords and the lower keyword pairs are determined, and the basic reserved word-lower keyword associations Should be determined.

The sub keyword used in the process of step 1340 may be determined while step 520 is performed during the process of step 1330. [ In this case, the lower keyword of step 520 may be used in step 1340. If the lower keyword is not determined in step 1330, the lower keyword may be determined through a process similar to or similar to the process of step 520 of FIG.

The association weights used in the process of step 1340 may be determined while step 530 is performed during the process of step 1330. [ In this case, the association weights of step 530 may be used in step 1340. If the association weights are not determined in step 1330, the association weights can be determined through a process that is the same as or similar to the process of step 530 and preceding steps of FIG.

The basic reserved word-lower keyword association can be calculated, for example, in consideration of the frequency in which a reserved word and a lower keyword appear in the same or similar context.

Hereinafter, an example of acquiring the basic reserved word-keyword association v _3,2 between the reserved word C ₂ and the representative attribute keyword k ₃ will be described with reference to FIGS. 15 to 17. For example, it is assumed that the lower keywords of the representative attribute keyword k ₃ are B3 ₁ to B3 ₅₀ . For the reserved word, the representative attribute keyword, and the basic reserved word-keyword association, reference is made to the example described above with reference to FIG. The sub-keyword association degree is expressed as x _{j, h} - reserved word reserved word base corresponding to C _j and sub-keyword pair Bg _h. The association weights corresponding to the pair of the sub keyword Bg _h and the representative attribute keyword k _g are expressed as y _{g, h} . The sub-keyword association degree is expressed as x _{j, g, h} - reserved word C _j, k _g represent attribute keywords and keyword sub Bg _h combination adjustment reserved word corresponding to.

15 is a detailed flowchart of step 1340 according to the first embodiment of the present invention.

Referring to FIG. 15, in step 1510, the control unit 320 obtains the adjusted reserved word-lower keyword association by applying association weights to the basic reserved word-lower keyword association.

In step 1510, the control unit 320 applies associative weights corresponding to the lower keyword and the representative attribute keyword pair to the basic reserved word-lower keyword association corresponding to the reserved word and the lower keyword pair for each reserved word and lower keyword pair It is possible to obtain the adjustment reservation word-lower keyword association diagram corresponding to each of the reserved words and the lower keyword pairs for the representative attribute keyword.

For example, controller 320 is C ₂ to reserved words that correspond to the sub-keyword B3 ₄ and shorthand keyword k ₃ pairs when C ₂ days The sub-keyword Relevancy sub keyword B3 ₄ and representing the property keywords x _2,4 k ₃ ssang-B3 ₄ between adjustment reserved - to obtain a lower degree keywords associated reserved words C ₂ and sub-keyword B3 ₄ ssang corresponding default reserved for the We can obtain the adjusted word-lower keyword association x _2,3,4 by applying the corresponding association weight y _3,4 .

In particular, the method of applying the association weight can be a method of multiplying the basic reserved word-lower keyword association x _2,4 by the association weight y _3,4 corresponding to the lower keyword B3 ₄ and the representative attribute keyword k ₃ pair. For example, controller 320 is C ₂ to reserved words that correspond to the sub-keyword B3 ₄ and shorthand keyword k ₃ pairs when C ₂ days B3 _{4 Intermediate} reserved words - Basic reserved words corresponding to ₄ pairs of reserved words C ₂ and B ₃ to obtain subordinate keyword associations x ₂ , ₃ and ₄ - Subordinate keyword associations x _{2 4} and subordinate keywords B _{3 4} and representative value is multiplied by the associated weight _3,4 y corresponding to the attribute keywords ssang k ₃ (x _2,4 Хy _{3, 4)} is adjusted by using the reserved - more sub keywords, it is possible to obtain the x _2,3,4. In another embodiment, the control unit 320 applies a different calculation / utilization scheme such that the adjusted reserved word-lower keyword relevance x ₂ , ₃ , ₄ is x _{2, 4,} and y ₃ , ₄ , function f adjusted by using the _{(x 2, 4, y 3} , 4) reserved words - can be obtained also x _2,3,4 associated sub-keyword. In addition, (x _2,4 Хy _{3, 4)} for adjusting the reserved word-bottom keywords associated with a method of utilizing as _2,3,4 x, and (x _2,4 Хy _{3, 4)} applying a correction based on other factors to adjust reserved - All of the methods used as sub-keyword associations x _{2, 3, 4} can be utilized.

가 될 수 있다. 즉, 하위 키워드 B3₁ 내지 B3₅₀에 대해 예약어 C₂와의 기본 예약어-하위키워드 연관도 x_2,f를 구한 뒤 각각에 대해 해당 하위 키워드에 대한 연관 가중치 y_3,f를 반영하여 x_{2, 3,f}를 구하고 x_{2, 3,f}를 누적하여 예약어 C₂ 및 대표 속성 키워드 k₃ 간의 기본 예약어-키워드 연관도를 구할 수 있다. 다른 실시 예에 따르면, 예약어 C₂ 및 대표 속성 키워드 k₃ 간의 기본 예약어-키워드 연관도는

에 기타 요인에 의한 보정을 가한 값이 될 수 있다. 다른 실시 예에 따르면, 예약어 C₂ 및 대표 속성 키워드 k₃ 간의 기본 예약어-키워드 연관도는

과 양의 상관 관계를 가지는 값이 될 수 있다. 여기서는 하나의 대표속성 키워드에 연결돼 있는 하위 키워드가 50개인 것으로 가정하였으나, 대표속성 키워드에 연결돼 있는 하위 키워드의 개수가 달라지면 수식에서 f의 누적 범위가 50이 아닌 다른 값이 될 수 있을 것이다.In step 1520, the control unit 320 may set a basic word-keyword association between the reserved word C ₂ and the representative attribute keyword k ₃ using the cumulative value of the adjusted reserved word-lower keyword association degree x _{2, 3, and f} . That is, the basic reserved word-keyword association between the reserved word C ₂ and the representative attribute keyword k ₃ is

. That is, the sub-keyword B3 ₁ to base the reserved word with the reserved word C ₂ for B3 ₅₀ - lower keyword Relevancy x _2, after obtaining a _f for each reflecting the associated weight y _{3, f} for its sub-keyword x _{2, 3 , f} , and accumulating x _{2, 3, and f} to obtain a basic word-keyword association between the reserved word C ₂ and the representative keyword k ₃ . According to another embodiment, the basic reserved word-keyword association between the reserved word C ₂ and the representative attribute keyword k ₃ is

And a correction based on other factors. According to another embodiment, the basic reserved word-keyword association between the reserved word C ₂ and the representative attribute keyword k ₃ is

And < / RTI > In this case, it is assumed that there are 50 sub-keywords connected to one representative attribute keyword. However, if the number of sub-keywords connected to the representative attribute keyword is different, the cumulative range of f in the formula may be different from 50.

16 is a detailed flowchart of step 1340 according to another embodiment of the present disclosure. The process of steps 1510 and 1520 of FIG. 16 is the same as the process described above with reference to FIG.

Referring to FIG. 16, in step 1530, the controller 320 determines whether or not a representative keyword having a basic reserved word-keyword association degree of the reserved word and the corresponding representative keyword pair of the representative keywords corresponding to a specific reserved word You can delete the default reserved words - keyword associations for. The representative keywords corresponding to a specific reserved word indicate keywords for which a basic reserved word-keyword association is set in relation to the specific reserved word. The reference basic reserved word-keyword association may be set in advance. According to another embodiment, the reference basic word-keyword association is set by using an average value of the basic word-keyword association corresponding to a specific reserved word or when a basic word-keyword association corresponding to a specific reserved word is arranged in the order of magnitude Can be set using the basic reserved word-keyword association of the ranking. A specific value having a positive correlation with the basic reserved word-keyword associations corresponding to other specific reserved words may be a reference basic reserved word-keyword association. The reference basic reserved word-keyword association may be set differently according to a reserved word or may be set to the same for all reserved words. The deletion of the basic reserved word-keyword association is set to have no association between the reserved word and the representative keyword. The control unit may set the degree of association to 0, or may associate the basic word-keyword association of the reserved word with the corresponding representative keyword pair in a list (which may be replaced with an array or other data structure) Keyword association may be deleted in such a manner as to delete information about the keyword. Other basic reserved words - a method of adding information indicating that a keyword association has been deleted (or an association has been deleted) may be used.

After the process of step 1530, the basic reserved word-keyword association having a relatively fine association is deleted (i.e., the association is set to have no association), or an overly complicated operation is performed or an association with no substantial meaning to the user / It can be prevented from being displayed.

17 is a detailed flowchart of step 1340 in accordance with another embodiment of the present disclosure.

The process of steps 1510 and 1520 of FIG. 17 is the same as the process described above with reference to FIG. 15, and thus description thereof will not be repeated. The process of step 1530 of FIG. 17 is the same as the process described above with reference to FIG. 16, and thus description thereof will not be repeated.

In step 1540, the control unit 320 may normalize the remaining basic reserved word-keyword association without being deleted. For example, in order to make the average value of the basic reserved word-keyword association stored in association with a specific representative attribute keyword included in a specific range, the basic reserved word-keyword associations connected to and stored in the specific representative attribute keyword are multiplied by a certain coefficient, . For example, in order to ensure that the sum of the basic reserved word-keyword associations stored in association with the specific representative attribute keyword is included in the specific range, the basic reserved word-keyword associations connected to the specific representative attribute keywords are stored, . In other words, in order to prevent a case in which only a specific representative attribute keyword is recommended / used or a specific representative attribute keyword is rarely used even when a certain reserved word is selected because a high-value basic reserved word- Appropriate adjustments can be made.

According to another embodiment, in step 1540, the control unit 320 may perform normalization by adding a predetermined coefficient instead of multiplying the basic word-keyword association by a certain coefficient, or by combining operations such as logarithmic and square. According to another embodiment, the normalization may be performed by reducing only a part of the basic reserved word-keyword associations beyond a certain reference value or by partially raising only the basic reserved word-keyword associations below a certain reference value.

According to another embodiment, in step 1540, the control unit 320 may normalize the average value (or the sum) of the stored basic word-keyword associations stored in a specific reserved word to be included in a specific range.

It is also possible that the step 1530 is omitted in the process of FIG. 17 and the basic reserved word-keyword association normalization process is performed.

In step 1350, the communication unit 310 receives the reception reservation word from the terminal 200, acquires it, and transmits the reception reservation word to the control unit 320.

The reception reservation word is a reserved word received by the terminal 200 from the search user. The terminal 200 may convert the voice input into an electrical signal (voice signal) and transmit it to the search device 300. The control unit 320 of the search apparatus 300 may analyze the voice signal, convert the voice signal into text, and match the converted text with a reserved word. The control unit 320 may also analyze the voice signal and analyze the voice intonation, the height, the speed, the breathing state, and the like to use the information as context information.

According to another embodiment, the terminal 200 may convert the voice input into text and transmit the converted text to the search device 300. The terminal 200 can analyze the received voice input, the height, the speed, the breathing state, and the like and transmit the analysis information to the search device 300. The search device 300 may utilize the analysis information as a kind of context information.

In step 1360, the control unit 320 obtains a reserved word-object association map corresponding to the reception reservation word and each object item pair using the object-keyword association map and the basic reserved word-keyword association map. The process of step 1360 may be performed according to the method of step 940 of FIG. 9 or the method of FIG. The description of the same contents is omitted.

In step 1370, the control unit 320 may provide the object item according to the reserved word-object association map corresponding to the received reservation word. The process of step 1370 may be performed in the same manner as the process of step 950. The same explanation will be omitted.

14 is a flowchart of an information providing process according to another embodiment of the present invention.

Steps 1310, 1320, 1330, 1340, 1350, 1360, and 1370 in FIG. 14 are the same as steps 1310, 1320, 1330, 1340, 1350, 1360, and 1370 in FIG. 13,

The added step 1333 in FIG. 14 may be performed at any time after the representative attribute keyword and the lower keyword are determined. For example, step 1333 may be performed in parallel / concurrent with step 1330, and the process of step 1333 may be performed during step 1330.

In step 1333, the control unit 320 stores association weights between the representative attribute keyword and the lower keyword. If the process of setting the association weights is not performed in step 1330, an association weight between the representative attribute keyword and the lower keyword may be set through a process similar to that of step 530 of FIG. According to another embodiment, the controller 320 may store association weights between the representative attribute keyword and the lower keyword in a manner that the association weights set in step 1330 are retrieved.

In step 1337, the control unit 320 obtains a basic keyword-lower keyword association between the lower keyword and the reserved word.

The control unit 320 can determine the degree of association between the reserved word and the lower keyword in consideration of, for example, the frequency with which the lower keyword appears in the same or similar context as the reserved word. For example, words appearing in the vicinity of keyword A in a certain sentence may appear in the vicinity of the word associated with keyword A in another document.

"I went on a big trip, but it was July, and the weather was so hot that I suffered."

"I went on a big trip, but it was July, so the weather was too humid."

Looking at the two sentences above, the word "hot" was replaced by the word "wet" in the same context. The control unit 320 can infer that "hot" and "wet" are related words.

"I went on a big trip, but it was July, and the weather was so hot that I suffered."

"I had a great vacation and went on vacation, but it was July, so I had a hard time because the weather was so hot."

Similarly, the control unit 320 can infer that the words " travel " and " vacation " are related words through the above two sentences.

"I went on a big trip, but it was July, and the weather was so hot that I suffered."

"I went on a big trip, but it was August and the weather was so hot that I suffered."

Similarly, the control unit 320 can infer that the words " July " and " August " are associated with each other through the above two sentences.

The control unit 320 is a word associated with "hot" and "wet" through previously collected documents, and a word associated with "July" and "August", and a word associated with "travel" and "vacation" It is assumed that the following sentence is collected.

"I had a great vacation and went on a vacation, but it was July, and the weather was too humid."

"I went on a big trip, but it was hard because it was too hot in August."

If the two sentences are not the same context but know that "hot" and "humid" are related words, "July" and "August" are related words, and "travel" ) Will be able to learn that "suffering" and "hardship" are also related words through the above sentence.

It is possible to determine that a keyword pair having a high frequency of appearing in the same / similar context has a high degree of correlation with each other. In addition, the higher the degree of similarity of the context in which the two keywords appear, the higher the degree of association between the two keywords can be judged. The control unit 320 performs learning using the collected documents to set the degree of association between the keywords and sets the degree of association of the keywords appearing in the corresponding sentence by using the relation between the set keywords and the context of the sentence. It is possible to improve the accuracy of the determination of the inter-linkage.

Similar learning methods are known, such as Neural Net Language Model (NNLM), Recurrent Neural Net Language Model (RNNLM), word2vec, skipgram, and Continuous Bag-of-Word (CBOW) schemes. In particular, when word2vec is used, word2vec learns by using documents and maps each keyword to a vector, and the similarity between two keywords can be grasped through calculation of the cosine similarity of two vectors.

Figure 20 is a diagram illustrating a hierarchy of terms in accordance with one embodiment of the present disclosure.

13 or 14, a hierarchical relationship between the reserved words C ₁ to C _q , the representative attribute keywords k ₁ to k _n , and the lower keywords BX ₁ to BX ₅₀ is established .

A basic reserved word-keyword association is set between the reserved word and the representative attribute keyword, and an association weight is set between the representative attribute keyword and the lower keyword. Using this hierarchical relationship, the control unit 320 can perform an operation such as recommending an appropriate object according to a reserved word or selecting a new reserved word candidate. In addition, the hierarchical relationship of FIG. 20 can be corrected or supplemented through learning while repeating the process of FIG. 13 and FIG. 14 by reflecting new data.

At this point, it will be appreciated that the combinations of blocks and flowchart illustrations in the process flow diagrams may be performed by computer program instructions. These computer program instructions may be loaded into a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, so that the instructions, which may be executed by a processor of a computer or other programmable data processing apparatus, Thereby creating means for performing functions. These computer program instructions may also be stored in a computer usable or computer readable memory capable of directing a computer or other programmable data processing apparatus to implement the functionality in a particular manner so that the computer usable or computer readable memory The instructions stored in the block diagram (s) are also capable of producing manufacturing items containing instruction means for performing the functions described in the flowchart block (s). Computer program instructions may also be stored on a computer or other programmable data processing equipment so that a series of operating steps may be performed on a computer or other programmable data processing equipment to create a computer- It is also possible for the instructions to perform the processing equipment to provide steps for executing the functions described in the flowchart block (s).

In addition, each block may represent a module, segment, or portion of code that includes one or more executable instructions for executing the specified logical function (s). It should also be noted that in some alternative implementations, the functions mentioned in the blocks may occur out of order. For example, two blocks shown in succession may actually be executed substantially concurrently, or the blocks may sometimes be performed in reverse order according to the corresponding function.

Herein, the term " part " used in the present embodiment means a hardware component such as software or an FPGA or an ASIC, and 'part' performs certain roles. However, 'part' is not meant to be limited to software or hardware. &Quot; to " may be configured to reside on an addressable storage medium and may be configured to play one or more processors. Thus, by way of example, 'parts' may refer to components such as software components, object-oriented software components, class components and task components, and processes, functions, , Subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays, and variables. The functions provided in the components and components may be further combined with a smaller number of components and components or further components and components. In addition, the components and components may be implemented to play back one or more CPUs in a device or a secure multimedia card.

It will be understood by those skilled in the art that the present specification may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive. The scope of the present specification is defined by the appended claims rather than the foregoing detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalents are included in the scope of the present specification Should be interpreted.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, It is not intended to limit the scope of the specification. It will be apparent to those skilled in the art that other modifications based on the technical idea of the present invention are possible in addition to the embodiments disclosed herein.

Claims (18)

Extracting a representative attribute keyword candidate set including a representative attribute keyword;
Setting a reserved word set including a reserved word;
Storing an object-keyword association corresponding to an object item and a representative attribute keyword pair;
Storing associative weight corresponding to the representative attribute keyword and the lower keyword pair and a basic reserved word-keyword association corresponding to the reserved word and representative attribute keyword pair using the basic reserved word-lower keyword association corresponding to the reserved words and the lower keyword pair ;
Obtaining a receiving reservation word;
Acquiring a reserved word-object association map corresponding to the reception reservation word and each object item pair using the object-keyword association diagram and the basic reserved word-keyword association diagram; And
Providing an object item according to the reserved word and the reserved word-object association diagram corresponding to each object item pair. The method according to claim 1,
Storing associative weight corresponding to the representative attribute keyword and the lower keyword pair and a basic reserved word-keyword association corresponding to the reserved word and representative attribute keyword pair using the basic reserved word-lower keyword association corresponding to the reserved words and the lower keyword pair Before,
Storing association weight corresponding to the representative attribute keyword and the lower keyword pair; And
Further comprising the step of acquiring the basic reserved word-lower keyword associations between each of the lower keywords corresponding to the representative attribute keywords and the reserved words. 3. The method of claim 2,
The step of acquiring the basic reserved word-lower keyword association between each of the lower keywords corresponding to the representative attribute keywords and the reserved words,
Determining a basic word-to-sub keyword association between the reserved word and each of the sub keywords in consideration of the frequency with which each sub keyword appears in the same or similar context as the reserved word. The method according to claim 1,
Storing associative weight corresponding to the representative attribute keyword and the lower keyword pair and a basic reserved word-keyword association corresponding to the reserved word and representative attribute keyword pair using the basic reserved word-lower keyword association corresponding to the reserved words and the lower keyword pair Quot;
Applying associative weights corresponding to the lower keyword and the representative attribute keyword pair to the basic reserved word and lower keyword association diagram corresponding to the reserved word and the lower keyword pair for each of the reserved word and the lower keyword pair, And obtaining an adjustment reservation word-lower keyword association corresponding to the lower keyword pair; And
Keyword association degree corresponding to the reserved word and the specific representative attribute keyword pair so as to have a positive correlation with the cumulative value of the adjustment reserved word-lower keyword association with respect to the specific representative attribute keyword Information delivery method. 5. The method of claim 4,
Applying associative weights corresponding to the lower keyword and the representative attribute keyword pair to the basic reserved word and lower keyword association diagram corresponding to the reserved word and the lower keyword pair for each of the reserved word and the lower keyword pair, And acquiring an adjustment reserved word-lower keyword association corresponding to the lower keyword pair,
For each of the reserved word and the lower keyword pair, a value obtained by multiplying the basic word-lower keyword association corresponding to the reserved word and the lower keyword pair by the association weight corresponding to the lower keyword and the representative attribute keyword pair, And obtaining an adjustment reservation word-lower keyword association corresponding to each of the reserved words and the lower keyword pair for the corresponding keyword. 5. The method of claim 4,
Storing associative weight corresponding to the representative attribute keyword and the lower keyword pair and a basic reserved word-keyword association corresponding to the reserved word and representative attribute keyword pair using the basic reserved word-lower keyword association corresponding to the reserved words and the lower keyword pair Quot;
A basic reserved word corresponding to a specific reserved word and a representative keyword having a keyword association among the representative keywords having a keyword association, a basic reserved word for a representative keyword pair corresponding to the specific reserved word, Further comprising the step of deleting the figure. 5. The method of claim 4,
Storing associative weight corresponding to the representative attribute keyword and the lower keyword pair and a basic reserved word-keyword association corresponding to the reserved word and representative attribute keyword pair using the basic reserved word-lower keyword association corresponding to the reserved words and the lower keyword pair Quot;
Deletes the basic reserved word-keyword associations for the representative keywords excluding the predetermined number of representative keywords selected from the representative keywords corresponding to the specific reserved words in the descending order of the basic reserved word-keyword associations of the specific reserved words and the corresponding representative keyword pairs Further comprising the steps of: 8. The method according to claim 6 or 7,
Storing associative weight corresponding to the representative attribute keyword and the lower keyword pair and a basic reserved word-keyword association corresponding to the reserved word and representative attribute keyword pair using the basic reserved word-lower keyword association corresponding to the reserved words and the lower keyword pair Quot;
And normalizing the basic reserved word-keyword association map of the basic reserved word-keyword association map corresponding to the specific reserved word so as to complement the basic reserved word-keyword association map. The method according to claim 1,
The step of setting a reserved word set including a reserved word includes:
The number of times that one language unit or two or more consecutive language units appear in a document included in a document set, and an emotion word included in the evaluation dictionary within a preset distance from the one language unit or two or more consecutive language units Setting a reserved word candidate in consideration of the number of times a word is reserved;
Providing interface information for generating a reserved word selection interface including reserved word candidate information; And
Receiving the reserved word selection input via the reserved word selection interface, and adding the selected reserved word candidate to the reserved word set according to the selected input. Extracting a representative attribute keyword candidate set including a representative attribute keyword, setting a reserved word set including a reserved word, storing an object-keyword association corresponding to the object item and the representative attribute keyword pair in the storage unit, Keyword associations corresponding to the reserved words and the representative attribute keyword pairs are stored in the storage unit by using the association weights corresponding to the pairs of the lower-level keywords and the basic reserved word-lower-keyword associations corresponding to the reserved words and the lower- A control unit;
A storage unit for storing the object-keyword associativity and the basic reserved word-keyword associativity; And
And a communication unit for obtaining a reception reservation word,
Wherein the control unit obtains a reserved word-object association map corresponding to the reception reservation word and each object item pair using the object-keyword association map and the basic reserved word-keyword association map,
Wherein the control unit provides an object item according to the reserved word and the reserved word-object association map corresponding to each object item pair. 11. The method of claim 10,
The controller associates the basic keyword-keyword associations corresponding to the reserved words and the representative keyword pairs using the association weights corresponding to the representative attribute keywords and the lower keyword pairs and the basic word-lower keyword associations corresponding to the reserved words and the lower keyword pairs Before storing in the storage,
The control unit stores the associative weight corresponding to the representative attribute keyword and the lower keyword pair in the storage unit and obtains the information of acquiring the basic reserved word-lower keyword association between each lower keyword corresponding to the representative attribute keyword and the reserved word Device. 12. The method of claim 11,
Wherein,
And determines the basic reserved word-lower keyword association between the reserved word and each of the lower keywords in consideration of the frequency with which each lower keyword appears in the same or similar context as the reserved word. 11. The method of claim 10,
Wherein,
Applying associative weights corresponding to the lower keyword and the representative attribute keyword pair to the basic reserved word and lower keyword association diagram corresponding to the reserved word and the lower keyword pair for each of the reserved word and the lower keyword pair, And an adjustment reservation word-lower keyword association corresponding to the lower keyword pair,
Keyword associations corresponding to the reserved word and the specific representative attribute keyword pair so as to have a positive correlation with the accumulated value of the adjusted reserved word-lower keyword association with respect to the specific representative attribute keyword. 14. The method of claim 13,
Wherein,
For each of the reserved word and the lower keyword pair, a value obtained by multiplying the basic word-lower keyword association corresponding to the reserved word and the lower keyword pair by the association weight corresponding to the lower keyword and the representative attribute keyword pair, To obtain an adjusted reservation word-lower keyword association corresponding to each of the reserved words and the lower-level keyword pairs. 14. The method of claim 13,
Wherein,
A basic reserved word corresponding to a specific reserved word and a representative keyword having a keyword association among the representative keywords having a keyword association, a basic reserved word for a representative keyword pair corresponding to the specific reserved word, The information providing apparatus comprising: 14. The method of claim 13,
Wherein,
Deletes the basic reserved word-keyword associations for the representative keywords excluding the predetermined number of representative keywords selected from the representative keywords corresponding to the specific reserved words in the descending order of the basic reserved word-keyword associations of the specific reserved words and the corresponding representative keyword pairs Information providing device. 17. The method according to claim 15 or 16,
Wherein,
And normalizes the basic reserved word-keyword association diagram of the basic reserved word-keyword association diagram corresponding to the specific reserved word so as to complement the basic reserved word-keyword association diagram. 11. The method of claim 10,
Wherein,
The number of times that one language unit or two or more consecutive language units appear in a document included in a document set, and an emotion word included in the evaluation dictionary within a preset distance from the one language unit or two or more consecutive language units And sets a reserved word candidate in consideration of the number of times that the reserved word is selected;
Wherein the communication unit provides interface information for generating a reserved word selection interface including reserved word candidate information,
Wherein the control unit adds the selected reserved word candidate to the reserved word set according to the selection input when the communication unit receives the reserved word selection input through the reserved word selection interface.

Images