KR20190134039A - Voice input processing method and apparatus thereof - Google Patents

Abstract

The present invention relates to a method and an apparatus for processing a voice input. More specifically, the method and apparatus convert a sentence or a phrase inputted by a voice from a user into a text, search for keytalk in the converted text, and provide information related to the field of the searched keytalk. The apparatus comprises: a display unit; an input unit; and a control unit.

Description

Voice input processing method and apparatus {VOICE INPUT PROCESSING METHOD AND APPARATUS THEREOF}

The present invention relates to a method and apparatus for processing a voice input, and more particularly, converts a sentence or phrase input by voice from a user into text, searches for a key talk in the converted text, and retrieves information related to the field of the searched key talk. The present invention relates to a method and an apparatus.

According to a conventional search method, a user may search for and find a desired web document by inputting a search keyword in a search window. For example, a user may enter the movie title "Interstellar" into the search box to retrieve information about the movie "Interstellar". However, if the user does not remember the title of the movie he is looking for, it is required to provide information in a different manner. For example, a user may attempt to search by entering an actor who appeared in a movie he is looking for, a director, a producer, etc. of the movie as a keyword. Movie information sites or movie review sites are often provided with the cast information along with the movie information, so if not fortunate, the user may be able to find a desired movie by keyword, such as actor, director, producer.

However, if the information is provided based on an atypical language, for example, an emotional language, rather than the standardized information, the conventional search method cannot be used. For example, a traditional search engine's response to a search query such as "fun movie" or "movie when watching sad" will search for documents that someone has written containing the keywords "fun movie" or "movie when watching sad." It will only be. However, atypical language requires a different approach from the standardized information such as actors, (formal) movie genres, and release years. Even if someone didn't include the keyword "funny movies" or "movies when sad," there may be movies that many people may find "fun" or "good when sad." Furthermore, other areas besides film may require different approaches to requests for information provision using unstructured languages.

For example, if a user enters the search term "ambient and quiet," the user may want a restaurant that is "atmosphere and quiet," or may want a "ambient and quiet" accommodation. The information needs to be organized and provided for each field.

In addition, once a field is determined and the user wants to search for the new field in the field or combine it with an existing search term, the user receives a new term from the user and searches for the field or combines with the existing search term. It is necessary to perform a search on the relevant field and provide the result to the user.

The present invention is to provide a method and apparatus for converting a sentence or phrase input by voice from a user into text, searching for keytalk in the converted text, and providing information related to the field of the searched keytalk.

An apparatus for processing voice input according to an embodiment of the present invention includes a display unit for displaying an integrated input button for receiving a voice input; An input unit configured to receive the voice input from a user when the integrated input button is selected; And a controller for converting the input voice into text, extracting a key talk and a field related word from the text, and requesting an information providing device for information related to the key talk, wherein the information related to the key talk is displayed for each field. do.

The key torque is set for each field.

The information related to the key talk is displayed together with the overview button for the field and information corresponding to the field.

If the field is selected or the view all button is selected, all the information of the field and the rescan button and the additional input search button are additionally displayed.

The input unit receives a voice input from the user when the re-search button or the additional input search button is selected.

When the rescan button is selected, the controller converts the input voice into text and extracts key talk from the text.

The controller requests the information providing apparatus for information related to the key talk in the field.

When the additional input search button is selected, the controller converts the input voice into text and extracts additional key talk from the text.

The controller requests the information providing apparatus for information related to the key talk and the additional key talk in the field.

A voice input processing method according to an embodiment of the present invention includes: displaying an integrated input button for receiving a voice input; Receiving the voice input from a user when the integrated input button is selected; Converting the input voice into text, extracting a keytalk and a field related word from the text, and requesting an information providing device for information related to the keytalk, wherein the information related to the keytalk is displayed for each field. do.

The key torque is set for each field.

The information related to the key talk is displayed together with the overview button for the field and information corresponding to the field.

If the field is selected or the view all button is selected, all the information of the field and the rescan button and the additional input search button are additionally displayed.

When the re-search button or the additional input search button is selected, a voice input is received from the user.

When the rescan button is selected, the input voice is converted into text and key talk is extracted from the text.

Requests information related to the key talk from the field to the information providing device.

When the additional input search button is selected, the input voice is converted into text and additional key talk is extracted from the text.

In the field, the information providing apparatus requests information related to the key talk and the additional key talk.

The voice input processing method and apparatus according to an embodiment of the present invention may extract a key talk from a voice received from a user, perform a search using the extracted key talk, and provide the result by field.

The voice input processing method and apparatus according to an embodiment of the present invention may be performed by selecting a field, receiving a new voice from the user, extracting a key talk from the received voice, and then using the key torque extracted from the field selected by the user. You can perform a new search and provide the results.

In the voice input processing method and apparatus according to an embodiment of the present invention, a user selects a field, receives a new voice from the user, extracts a key talk from the received voice, and then extracts the existing keytalk from the field selected by the user. Keytalk can be used to perform new searches and provide results.

1 is a network diagram of an information providing system using an attribute language according to an embodiment of the present specification.
2 is a block diagram of a terminal 200 according to an embodiment of the present disclosure.
3 is a block diagram of an information providing apparatus 300 according to an exemplary embodiment.
4 is a flowchart illustrating an information providing process through an information providing interface according to an exemplary embodiment of the present specification.
5 is a detailed flowchart of step 910 according to an embodiment of the present disclosure.
6 is a detailed flowchart of step 510 according to an embodiment of the present disclosure.
7 is a detailed flowchart of step 530 according to an embodiment of the present disclosure.
8 is a flowchart illustrating an information providing process according to another exemplary embodiment of the present specification.
9 is a flowchart illustrating an information providing process according to an exemplary embodiment of the present specification.
10 is an illustration of a stored object-keyword association diagram according to one embodiment of the present disclosure.
11 is an illustration of a basic reserved word-keyword association according to one embodiment of the present specification.
12 is a detailed flowchart of step 940 according to one embodiment of the present specification.
13 is a flowchart illustrating an information providing process according to another exemplary embodiment of the present specification.
14 is a flowchart illustrating an information providing process according to another embodiment of the present specification.
15 is a detailed flowchart of step 1340 according to the first embodiment of the present specification.
16 is a detailed flowchart of step 1340 according to another embodiment of the present disclosure.
17 is a detailed flowchart of step 1340 according to another embodiment of the present disclosure.
18 is a detailed flowchart of step 1320 according to a modification of the present specification.
19 is an example of an interface 1900 generated according to the interface information provided in step 1840.
20 is a diagram illustrating a term hierarchy according to an embodiment of the present specification.
21 is a flowchart illustrating a virtual input interface program installation process according to an embodiment of the present invention.
22 is a diagram illustrating an interface page of a package that performs a voice input processing method according to an embodiment of the present invention.
FIG. 23 is a diagram illustrating an interface page displaying a result of inputting a voice by selecting the integrated voice input button of FIG. 22.
FIG. 24 is a diagram illustrating an interface page displayed when the entire display button is selected in the result display unit of FIG. 23.
25 is a flowchart illustrating a voice input processing method according to an embodiment of the present invention.

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

In describing the embodiments, descriptions of technical contents that are well known in the art to which the present disclosure belongs and are not directly related to the present specification will be omitted. This is to more clearly communicate without obscure the subject matter of the present specification by omitting unnecessary description.

For the same reason, in the accompanying drawings, some components are exaggerated, omitted or schematically illustrated. In addition, the size of each component does not fully reflect the actual size. The same or corresponding components in each drawing are given the same reference numerals.

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

According to some embodiments of the present disclosure, the reserved word may refer to a string that may be expressed or defined through relevance information with other keywords, or that may be expressed and defined in its characteristics, usage, and properties. However, it is also possible to apply a reserved word defined / expressed according to another method, for example, by an administrator's manual input. The detailed configuration of the association information will be described later. The reserved word may also be referred to as keytalk ^TM in the sense of being a key of talk.

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

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

The terminal 200 may be implemented as, for example, an electronic device or a similar device that performs a smartphone, a PDA, a tablet PC, a notebook computer, a laptop computer, a personal computer, or other communication, receives a user input, and outputs a screen. Can be.

The information providing apparatus 300 may be implemented as, for example, a workstation, a server, a general purpose computer, an electronic device capable of performing other communication, or a similar device.

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, Long Term Evolution (LTE), LTE-Advanced (LTE-A), WI-FI, Local Area Network (LAN), Wide Area Network (WAN), Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Wireless Broadband (WiBro), Global System for Mobile Communications (GSM) and the like may be implemented using at least some of the past, present and future communication schemes. Hereinafter, for convenience, the terminal 200 and the information providing apparatus 300 will be described as if they directly communicate without referring to the communication network 150.

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

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

Referring to FIG. 2, the terminal 200 according to an exemplary embodiment of the present specification 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 a user's input operation 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, and other input processing devices that may be available in the present, the past, or the future. The input unit 210 may receive, for example, a user's information input request input and transmit the received information to the controller 250.

The display unit 220 outputs a screen under the control of the controller 250. The display unit 220 may be implemented as, for example, LCDs (liquid crystal displays), LEDs (light emitting diodes), OLEDs (organic light emitting diodes), projectors, and other display devices that may be available now, in the past, or in the future. The display unit 220 may display, for example, an interface page for providing information or an information providing result page. According to an exemplary embodiment, the display unit 220 may be used in place of the display unit 220 using a different method of transmitting information to other users such as voice output or vibration instead of the screen output.

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

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

The controller 250 controls the overall operation of the terminal 200 and each component. In particular, the controller 250 transmits an information providing request and other data to the information providing apparatus 300 according to the information input from the input unit 210 as described below, and according to the page information received from the information providing apparatus 300. The result page and / or the interface page are displayed on the display unit 220.

Operations performed by the controller 250 may be distributed by various computing devices that are physically separated. Some of the operations performed by the controller 250 may be performed by the first server, and other operations may be performed by the second server. In this case, the controller 250 may be implemented as a total of physically separated computing devices.

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

When the control unit 250 or the storage unit 240 is implemented as a total of several physically separated devices, communication between the various devices may be required. For simplicity, the description will be given on the assumption that the storage unit 240 or the controller 250 is implemented as a single object.

When the terminal 200 transmits / receives data, the communication unit 230 may transmit or receive data according to the control of the control unit 250 according to a viewpoint, or the control unit 250 controls the communication unit 230 to transmit / receive data. You can also express it.

A detailed operation of each component of the terminal 200 will be described later with reference to FIGS. 4 to 8.

3 is a block diagram of an information providing apparatus 300 according to an exemplary embodiment.

Referring to FIG. 3, the information providing apparatus 300 according to an embodiment of the present specification 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 device. The communicator 310 transmits the data received from the terminal 200 to the controller 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 or other circumstances of the communication network 150.

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

The controller 320 controls the overall operation of the information providing apparatus 300 and each component. In particular, when receiving the interface page request, the information providing result page request, and other data through the communication unit 310, the controller 320 loads necessary data from the storage unit 330 and generates page information as described below. The page information is transmitted to the terminal 200 through the communication unit 310.

When the information providing device 300 transmits / receives data, the communication unit 310 may express that the data is transmitted / received according to the control of the control unit 320 according to a viewpoint, or the control unit 320 controls the communication unit 310 to transmit the data. It can also be expressed as sending and receiving.

The storage unit 330 may search for a reserved word from an input word input by a user in the terminal 200 to program code for visually distinguishing the reserved word, and / or an installation package (installation program code) for installing the program. Save it. Hereinafter, a program code for searching for a reserved word from an input word input by a user in the terminal 200 and visually distinguishing the reserved word, and / or an installation package (installation program code) for installing the program is called a virtual input interface program. .

A developer or administrator who develops the information providing apparatus 300 may store a virtual input interface program in the storage 330. When the information providing device 300 receives the request message for requesting the virtual input interface program through the communication unit 310, the controller 320 transmits the virtual input interface program to the terminal 200 through the communication unit 310.

A detailed operation of each component of the information providing apparatus 300 will be described later with reference to FIGS. 4 to 8.

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

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

In operation 410, the controller 320 of the information providing apparatus 300 generates interface page information. The interface page information is information necessary for generating an interface page. The interface page is a page for inducing a user's input, receiving the user's input, and delivering 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 may have the form information of the interface page in advance, and only the content corresponding to the content may be transmitted from the information providing apparatus 300 to the terminal 200. In the following description, it is assumed that interface page information or other page information is transmitted in an HTML document format for convenience. However, the scope of the present specification is not limited thereto.

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

In step 430, the controller 250 of the terminal 200 configures the interface page using the interface page information. For example, the controller 250 may configure an interface page in the form of a web page by interpreting an HTML document by driving a web browser. A separate application may be used instead of a web browser.

In operation 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 that allows the user 400 to request information, input and / or select a keyword for providing the information, and configure settings for providing other information.

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

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

In operation 470, the controller 320 of the information providing apparatus 300 generates result page information using a user input (eg, a keyword and / or other information provision setting). A preparation process for generating result page information and a process for generating result page information will be described later with reference to FIGS. 5 to 8. The result page information may be configured, for example, in HTML document format and / or image format.

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

In operation 490, the control unit 250 of the terminal 200 configures a result page using the result page information received by the communication unit 230. For example, the controller 250 may configure the result page by interpreting the result page information in the HTML format.

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

In the embodiment of FIG. 4, it is assumed that a page in a visual form is provided to the user, but it is also possible to provide an interface or result information through voice. In this case, the voice output unit may be used instead of the display unit 220. Instead of a visual / acoustic approach, other currently available or future ways of interface may be used in the relationship with the user 400. In this case, the information providing apparatus 300 may provide the terminal 200 with information converted in another manner instead of page information so as to conform to the interface method.

In an embodiment after FIG. 5, the user 400 wants to be provided with information about an object of a particular interest field of interest. However, the scope of the present specification is not limited thereto.

The field of interest can be, for example, the type of object. For example, if the area of interest is 'celebrity', the objects corresponding to the area of interest may include 'simin', 'yujaeseok', 'taylor swift' and the like. For example, if your area of interest is "movies," your objects of interest may include "Dunkerk," "Spider-Man: Homecoming," and "Superbad 3." For example, if the field of interest is a 'broadcast program', objects corresponding to this field of interest may include 'Infinite Challenge', 'American Idol', and 'Game of Thrones'.

In the following embodiments, a document is collected to evaluate the relationship (association / weighting, etc.) between keywords. The collected documents may all be valued as having an equal value, and newer documents may be rated as having higher value. That is, the correlation between the age of the evaluation point reference document and the keywords appearing in the document may have a negative correlation.

In the process subsequent to FIG. 5 to be described later, values may be evaluated differently according to the freshness of the document. For example, the case where two keywords appear together in a document one day after the evaluation point may be evaluated as 10 times higher as compared with the case where two keywords appear together in the document 10 days after the evaluation point. The age of the document may be evaluated in units of seconds / minutes / hours or in units of days / months / years, for example. The control unit 320 extracts the partial association degree reflecting the document age by dividing the partial association value by the age of the document based on the corresponding document evaluated before the age of the document is reflected, and accumulating the partial association degree between the two keywords. The association of can be extracted.

When a document is generated to check the age of the document, for example, the publication time and / or metadata included in the document may be determined, or periodically crawled to discover a new document not found in a previous crawl. In this case, new documents may be added at the time of the new crawl.

9 is a flowchart illustrating an information providing process according to an exemplary embodiment of the present specification.

In operation 910, the controller 320 stores the object-keyword association degree corresponding to the object item and the representative attribute keyword pair in the storage unit 330.

10 is an illustration of a stored object-keyword association diagram according to one embodiment of the present disclosure.

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

For example, the object-keyword association degree 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, according to some of the embodiments of FIGS. 5 to 8, or a similar process, or a corresponding process. According to another embodiment of the present disclosure, the process of step 910 may be performed by input of an administrator or by receiving an object-keyword association degree determined by an external system through a network or a storage medium.

5 is a detailed flowchart of step 910 according to an embodiment of the present disclosure.

Referring to FIG. 5, in step 510, the controller 320 extracts the representative attribute keyword candidate set from the first set documents. The controller 320 may collect, for example, keywords that frequently appear in documents corresponding to the field of interest in the first set of documents as representative attribute keyword candidate sets.

6 is a detailed flowchart of step 510 according to an embodiment of the present disclosure.

The control unit 320 selects keywords appearing in a document, such as an object keyword representing an object item belonging to a specific field, and keywords appearing in a document, such as a field keyword indicating a specific field, of the first attribute keyword candidate set and the second attribute keyword candidate. Can be set as a set.

For example, if the field of interest is a celebrity, the field keyword may include 'celebrity', 'celebrity', 'movie actor', 'star', 'celeb', 'celeb' and the like. The field keyword may be set by the administrator or recommended and set by the controller 320. The controller 320 may recommend / set a similar keyword, which is obtained by obtaining some field keywords and analyzed similarity with the field keywords more than a preset value, as an additional field keyword.

If the field of interest is a celebrity for providing the information providing service, the object keyword may be an individual person belonging to the field of interest. For example, "Yoo Jae-seok", "Taylor Swift", "Stephan Curry", etc. may be an object keyword corresponding to a celebrity of interest.

To describe the relationship between the field keyword and the object keyword, for example, the field keyword may correspond to an attribute or type of the object keyword. If the field keyword represents a set, the object keyword can be an element representing the set.

The object keyword may be set by an administrator or selected in a manner similar to the field keyword. According to another embodiment, the controller 320 may analyze the context of the collected documents and select keywords determined as elements belonging to the set represented by the field keyword as object keywords.

The popular object keyword and the less popular object keyword may be classified according to the search / collection amount of the corresponding object keyword. The controller 320 may search / collect documents containing each object keyword, set an object keyword having a collection amount more than a specific threshold value as a popular object keyword, and set the remaining object keywords as non-popular object keywords.

Popular field keywords and non-popular field keywords may be classified according to search / collection amounts of corresponding field keywords. The controller 320 may search / collect documents containing respective field keywords, set a field keyword having a collection amount equal to or more than a specific threshold, as a popular field keyword, and set the remaining field keywords as non-popular field keywords. However, a threshold for distinguishing a popular object keyword and a non-popular object keyword and a threshold for distinguishing a popular field keyword from an unpopular field keyword may be different values. For convenience, the popular object keyword and the popular field keyword are collectively referred to as popular field object keywords for convenience. Also, for convenience, both non-popular object keywords and non-popular field keywords are referred to as non-popular field object keywords.

In a modified embodiment, only the popular field keyword or the popular object keyword may be used instead of the popular field object keyword. In a modified embodiment, only the non-popular field keyword or the non-popular object keyword may be used instead of the non-popular field object keyword.

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

The controller 320 may search / collect documents including the popular field object keyword and set keywords included in the collected document as the first attribute keyword candidate set. According to another embodiment, the controller 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 controller 320 may exclude a meaningless keyword, for example, a survey / article, that is set in advance 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 controller 320 searches / collects a document including a popular field object keyword, and searches keywords collected within a predetermined distance from the popular field object keyword or a sentence including the keyword in the collected document. It can also be included in the first attribute keyword candidate set. According to another exemplary embodiment, the controller 320 may search / collect a document including a popular field object keyword, analyze a context, and select keywords used for the purpose of modifying / explaining a popular field object keyword as the first attribute keyword candidate. It can also be included in a set.

The distance between keywords or the distance between a keyword and a sentence may be, for example, the number of words placed between two keywords or keywords and sentences, the number of words placed between two keywords or keywords and sentences, the distance between two keywords or keywords and sentences, It may be determined based on any one or more of the number of words located, two keywords, or the number of characters between the keyword and the sentence.

The controller 320 may first perform morphological analysis for keyword analysis.

In operation 620, the controller 320 sets keywords appearing together in a document such as an unpopular field object keyword as a second attribute keyword candidate set.

The controller 320 may search / collect a document including a non-popular field object keyword and set keywords included in the collected document as a second attribute keyword candidate set. According to another embodiment, the controller 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 controller 320 may exclude a meaningless keyword, for example, a survey / article, that is set in advance from the second attribute keyword candidate set. According to another embodiment, the controller 320 may include, in the second attribute keyword candidate set, a keyword registered in a preset dictionary among keywords included in the collected document.

According to another embodiment, the controller 320 searches / collects a document including a non-popular field object keyword, and searches keywords collected within a predetermined distance from the non-popular field object keyword or a sentence including the keyword in the collected document. It may also be included in the second attribute keyword candidate set. According to another exemplary embodiment, the controller 320 may search for / collect a document including a non-popular field object keyword, analyze the context, and select keywords used to modify / explain the non-popular field object keyword as the 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 may be, for example, the number of words placed between two keywords or keywords and sentences, the number of words placed between two keywords or keywords and sentences, the distance between two keywords or keywords and sentences, It may be determined based on any one or more of the number of words located, two keywords, or the number of characters between the keyword and the sentence.

The controller 320 may first perform morphological analysis for keyword analysis.

In operation 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, keywords used as expressions for modifying popular field object keywords and used as expressions for modifying non-popular field object keywords may be collected in the representative attribute keyword candidate set.

According to another embodiment, in operation 510, the controller 320 may include keywords appearing together with the object keyword and / or the field keyword in the representative attribute keyword candidate set regardless of popularity / unpopularity.

5, in operation 520, the controller 320 extracts two or more sub keywords associated with each representative attribute keyword included in the representative attribute keyword candidate set from the second set of 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 the same document set. For example, the first aggregated document may be a collection including all collectable documents, and the second aggregated document may be a collection including only documents in which a specific area of interest for which an information providing service is to be used as a key keyword. . The controller 320 may analyze the collectable documents and analyze whether each document is a document used as a main keyword for a specific field of interest to provide an information service based on frequently appearing keywords. According to another embodiment, both the first and second aggregated documents may be a collective including all of the related documents that can be collected. According to another embodiment, the first aggregated document may be a collection including all collectable related documents, and the second aggregated document may be a collection including only documents related to a specific area of interest for which an information providing service is to be provided. According to another exemplary embodiment, the second aggregated document may be a collection including all collectable related documents, and the first aggregated document may be a collection including only documents related to a specific area of interest for which an information providing service is to be provided.

For step 520, the control unit 320 may include, for example, documents including a field keyword indicating the field of interest itself so as to generate a set including only documents related to a specific field of interest for which the information providing service is to be provided; Alternatively, documents including object keywords belonging to the corresponding interest may be collected, and documents having a specific gravity or more of the field keywords / object keywords among them may be extracted to generate a set including only documents related to a specific interest. The weight of the field keyword / object keyword may be determined based on the frequency of occurrence of the field keyword / object keyword, the appearance position, and the context. For example, a document in which a field keyword / object keyword frequently appears, a field keyword / object keyword is used as a title of the document, or displayed in large letters or a typeface for emphasis may be classified as a document related to a particular interest.

In operation 520, the controller 320 may analyze, for example, at least a portion of the second set of documents to extract a predetermined number of sub keywords most relevant to each representative attribute keyword, and to extract two or more subordinate items associated with each representative attribute keyword. You can extract keywords.

For example, the controller 320 may determine the degree of association between the representative attribute keyword and the lower keyword in consideration of the frequency of the lower keyword appearing in the same or similar context as the representative attribute keyword. For example, words appearing around keyword A in a particular sentence may appear around words associated with keyword A in other documents.

"I went on a trip with a big heart, but I suffered because the weather was so hot in July."

"I went on a trip with a big heart, but the weather was too humid for July."

Looking at the above two sentences, the word "hot" is replaced with "wet" in the same context. The controller 320 may infer that "hot" and "wet" are words associated with each other.

"I went on a trip with a big heart, but I suffered because the weather was so hot in July."

"I went on vacation after eating a big heart, but I suffered because the weather was so hot in July."

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

"I went on a trip with a big heart, but I suffered because the weather was so hot in July."

"I went on a trip with a big heart, but I suffered because the weather was so hot in August."

Similarly, the control unit 320 may infer that the words "July" and "August" are related through the above two sentences.

The controller 320 indicates that words "hot" and "wet" are related words, "July" and "August" are related words, and "travel" and "vacation" are words related to each other through previously collected documents. You can save it.

"I went on vacation with a big heart, but the weather was too humid for July."

"I went on a trip with a big heart but it was hard because the weather was so hot in August."

If the two sentences are not in the same context but know that "hot" and "wet" are related words, "July" and "August" are related words, and "travel" and "vacation" are words related to each other, the control unit 320 ) Can be learned from the above sentence that "toil" and "tough" are also related words.

Keyword pairs with high frequency appearing in the same / similar context may be determined to be highly related to each other. In addition, it may be determined that the higher the similarity of the context in which the two keywords appear, the higher the correlation between the two keywords. The controller 320 proceeds with learning using the collected documents to set the degree of association between keywords, and sets the degree of association between keywords appearing in the sentence using the context between the set keywords and the context. The correlation between the two can increase the accuracy of the judgment.

Similar learning methods are known as the Neural Net Language Model (NNLM), Recurrent Neural Net Language Model (RNNNLM), word2vec, skipgram, and Continuous Bag-of-Words (CBOW). In particular, when using word2vec, word2vec learns from documents and maps each keyword to a vector, and the similarity between two keywords can be determined by calculating the cosine similarity of the two vectors.

In this manner or in a similar manner, the controller 320 may analyze at least a portion of the second set of documents to extract a predetermined number of lower keywords most relevant to each representative attribute keyword.

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

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

In operation 710, the controller 320 may analyze at least some of the second aggregated documents to extract the degree of correlation between the sub-keywords. For example, assume that 50 collected sub keywords B1 ₁ to B1 ₅₀ are collected as associated sub keywords for the representative attribute keyword A1. In this case, the controller 320 may extract the correlation between the two sub keywords using the frequency of the two sub keywords appearing together in the same document for the 50 sub 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 embodiment B1 ₁ and B1 ₂ is The distance between the addition to the frequency appearing in the same document that affects the degree interrelated, B1 ₁ and B1 if the _two appeared in the same document both keywords B1 ₁ and B1 distance between _two (or a statement that both keywords appear ), The higher the correlation can be recognized. In a similar manner, the correlation between the sub keywords can be extracted. The distance between keywords or the distance between a keyword and a sentence may be, for example, the number of sentences placed between two keywords or keywords and sentences, the number of words placed between two keywords or keywords and sentences, and the distance between two keywords or keywords and sentences. It may be determined based on any one or more of the number of words located, two keywords, or the number of words located between the keyword and the sentence.

In operation 720, the controller 320 may extract an association weight between each representative attribute keyword and the lower keyword based on the degree of correlation between the lower keywords. For example, the control unit 320 may include a correlation between the specific sub-keyword in the sub-keyword set and other sub-keywords in the sub-keyword set with respect to a sub-keyword set corresponding to each representative attribute keyword, and the specific sub-keyword and the An association weight between the specific sub-keyword and the representative attribute keyword may be set such that the association weight between the representative attribute keywords has a positive correlation with each other.

For example, as the degree of correlation between the sub keywords B1 ₁ of the representative attribute keyword A1 and the other sub keywords B1 ₂ to B1 ₅₀ of A1 is higher, an association weight between A1 and B1 ₁ may be set higher. For example, keywords of the other sub-B1 ₁ 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. The geometric mean / harmonic mean may be used instead of the simple arithmetic mean. A cutting mean that is averaged may be used except for the highest two (example) and the lowest two (example) of the correlation between the sub keywords B1 ₁ and the other sub keywords B1 ₂ through B1 ₅₀ . have. Instead of the arithmetic mean of the correlation, a median may be used.

According to some embodiments, B1 to A1_OneUsed to calculate the associative weight of "B1"_OneAnd B1₂end How often they appear in the same article "simply_OneAnd B1₂end Does not depend on the number of documents that appear together (or appear in the same sentence, or appear in close proximity), but B1_OneAnd B1₂end B1 counts the number of documents that appear together (or appear in the same sentence, or appear in close proximity)._OneThe number of documents in which it appears and / or B1₂It can be obtained by dividing by the number of documents appearing. In a similar way, "B1_OneAnd B1₂end Frequency in the same document "_OneAnd B1₂end B1 with a positive correlation with the number of documents that appear together (or in the same sentence, or appear in close proximity)_OneThe number of documents in which it appears and / or B1₂It can be set to have a negative correlation with the number of documents appearing. 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 operation 540, the controller 320 may extract a sub association between an object item and a sub keyword from the first set of documents.

Among the first set of documents, a document such as an object keyword representing an object item (for example, "Taylor Swift"), or sub keywords frequently appearing in the same sentence or a close sentence may be determined to be associated with the object item. The controller 320 may collect documents in which an object keyword of the corresponding object item appears, and extract a lower degree of association between the lower keyword and the object keyword according to the frequency of appearing together in the documents. In particular, when the sub-keyword appears in the same sentence as the object keyword, the controller 320 may set a higher degree of association between the sub-keyword and the object item than when the sub-keyword appears in the other sentence.

The control unit 320 may set that the closer the sentence in which the lower keyword appears is to the sentence in which the object keyword appears, the higher the degree of association between the lower keyword and the object item of the corresponding object keyword is. For example, the proximity of two sentences may include, for example, the number of sentences between two sentences, the number of words between two sentences, the number of words between two sentences, and the number of characters between two sentences. It may be determined based on one or more.

The control unit 320 may set the higher degree of association between the lower keyword and the object item of the corresponding object keyword as the position where the lower keyword appears is closer to the position where the object keyword appears. The proximity between the subkeyword and the object keyword is, for example, the number of sentences located between the subkeyword and the object keyword, the number of words located between the subkeyword and the object keyword, and the word between the subkeyword and the object keyword. It may be determined based on any one or more of the number, the number of characters located between the lower keyword and the object keyword.

In operation 550, the controller 320 may extract an object-keyword relation between the object item and each representative attribute keyword using the sub-association of operation 540 and the association weight of operation 530.

For example, the object-keyword association between object entry C and representative attribute keyword A1 uses the subassociation between C and A1 subkeywords (eg B1 ₁ through B1 ₅₀ ) and the association weight of each subkeyword. Can be extracted. For example, the object-keyword association between the object item C and the representative attribute keyword A1 may be set higher as the lower degree of association between the object item C and B1 ₁ to B1 ₅₀ is higher.

In addition, if a subkey with a higher association weight has a higher subassociation with the object item C in relation to A1, an object C and a representative attribute keyword A1 than a case where the subkey with a low association weight have a high subassociation is higher. The object-keyword association between can be set higher. For example, in the case of Table ₁ , the lower association of the side with higher association weight (B1 ₁ ) is higher than that of Table 2, so the case of Table 1 is the object-object between the object C and the representative attribute keyword A1 than the case of Table 2. Keyword relevance may be set high.

Association weight with A1 Subassociation with C B1 ₁ 0.5 0.5 B1 ₂ 0.2 0.2

Association weight with A1 Subassociation with C B1 ₁ 0.2 0.5 B1 ₂ 0.5 0.2

According to an embodiment of the present disclosure, an object-keyword correlation between the object C and the representative attribute keyword A1 may be obtained as a sum (or using the sum) of a product obtained by multiplying an association weight corresponding to each sub keyword and a sub association degree. In Table 1, 0.5 × 0.5 + 0.2 × 0.2 = 0.29, and in Table 2, 0.2 × 0.5 + 0.5 × 0.2 = 0.20, so that Table 1 is more between object C and representative attribute keyword A1 than in Table 2. The object-keyword association of may be set high. The above-described object-keyword association calculation method is merely exemplary, provided that the sub-association with C in step 540 and the association weight with A1 in step 530 are positively correlated with the object-keyword association between C and A1. Other methods may be used.

Thereafter, when the communication unit 310 receives the information providing request associated with the specific representative attribute keyword, the controller 320 may provide the result item through the communication unit 310 based on the object-keyword association degree extracted in step 550. For example, when receiving a request for providing an information including one representative attribute keyword, the controller 320 may provide information about an object item in the order of highest object-keyword association in relation to the representative attribute keyword. have.

In another embodiment, when receiving an information providing request including two or more representative attribute keywords and corresponding weights, the controller 320 associates object-keywords with the representative attribute keywords included in the information providing request for each object item. The sum (or average) of the values multiplied by the weight (or similarly weighted) may provide information about the object item in the highest order.

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

The embodiment of FIG. 8 further includes two steps 523 and 526 between steps 520 and 530 in addition to the same processes as the embodiment of FIG. 5. Here, only the steps 523 and 526 will be described and the description thereof will be avoided.

In operation 523, the controller 320 determines whether each of the lower keywords extracted in operation 520 corresponds to an emotional language (emotional language). To this end, the storage unit 330 or an external server may hold an emotional dictionary. An emotional word dictionary is a tool for determining whether a word (keyword) is an emotional word, and for example, may hold a list of emotional words. Keywords included in the list of emotional words may be called an emotional word, and keywords that are not included in the emotional word list may not be determined. However, these judgments are based on a dictionary meaning and may not reflect the use of the word of the public that changes with the times. Accordingly, the controller 320 determines whether to use the representative attribute keyword based on whether the representative keywords of the sub-keywords associated with the representative attribute keyword are determined, without determining whether the representative attribute keyword itself is an emotional word.

In another exemplary embodiment, the controller 320 may add another word learned to be highly related (more than a preset value) to the emotional word dictionary registered as the emotional word in the emotional word dictionary.

In operation 526, the controller 320 may leave only the predetermined number of representative attribute keywords in the representative attribute keyword candidate set and remove the remainder in the order of the high emotional language ratio (or number) of the associated lower keywords. This process prevents keywords that are far from the emotional language to be treated like the emotional language.

9, in operation 920, the controller 320 stores the basic reserved word-keyword association degree corresponding to the reserved word and the representative attribute keyword pair in the storage unit 330.

The reserved word may include expressions that may be expressed as weights of the representative attribute keywords. For example, "bored" can also be a reserved word, and "pretty" can be a reserved word.

Representative keyword with high basic reserved word-keyword correlation for reserved word "bored" may include representative attribute keywords that can solve boring situations such as "fun", "interesting", "for time out", and the like. .

Regarding the Reserved Word "Pretty" A representative attribute keyword with a high basic reserved word-keyword association is a representative attribute that describes, for example, "pretty" or similar to "pretty" such as "beautiful", "cute", "eye-catching". May include keywords.

For example, the process of step 920 may be performed by input of an administrator or by receiving a basic reserved word-keyword association degree 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, and news. In addition, the process of step 920 may include a process of reflecting user feedback as described below.

The process of step 920 may be performed in a manner described below with reference to any one of FIGS. 15 to 17.

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

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

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

In operation 930, the communicator 310 receives and acquires a reserved reserved word from the terminal 200, and transmits the received reserved word to the controller 320.

The reserved reserved 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 the converted voice signal to the search apparatus 300. The controller 320 of the search apparatus 300 may analyze the voice signal, convert the voice signal into text, and match the converted text with the reserved word. The controller 320 may also analyze the speech signal and analyze the speech intonation, height, speed, respiratory state, and the like, and use the context information.

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

In operation 940, the controller 320 obtains the reserved word-object association corresponding to the received reserved word and each object item pair by using the object-keyword association and the basic reserved word-keyword association.

12 is a detailed flowchart of step 940 according to one embodiment of the present specification.

Referring to FIG. 12, in operation 1210, the controller 320 obtains a coordinated object-keyword association corresponding to each object item and representative attribute keyword pair for a reserved reserved word.

According to an embodiment of the present disclosure, the controller 320 may correspond to the received reserved word and the representative attribute keyword pair in an object-keyword association degree corresponding to the object item and the representative attribute keyword pair for each object item and the representative attribute keyword pair. By applying a basic reserved word-keyword association, the adjusted object-keyword association corresponding to each object item and representative attribute keyword pair for the received reserved word may be obtained.

In particular, the control unit 320 is a basic reserved word-keyword corresponding to the received reserved word and the representative attribute keyword pair in the object-keyword association degree corresponding to the object item and the representative attribute keyword pair for each object item and the representative attribute keyword pair. The adjusted object-keyword association degree corresponding to each object item and representative attribute keyword pair for the reserved reserved word may be obtained using the value multiplied by the degree of association.

In addition, the control unit 320 has a positive correlation with the object-keyword association corresponding to the object item and the representative attribute keyword pair for each object item and the representative attribute keyword pair, and is applied to the received reserved word and the representative attribute keyword pair. Coordinated object-keyword associations may also be set for the received reserved words, corresponding to each object item and representative attribute keyword pair, to have a positive correlation with the corresponding basic reserved word-keyword association.

In this specification, the values representing the object-keyword association, the basic reserved word-keyword association, the adjusted object-keyword association, the basic reserved word-subkeyword association, and other associations are all assumed to be closer to each other. . In other embodiments, a small correlation value may have a close correlation, while another association value may have a higher correlation value, so that a positive correlation-negative correlation may be appropriate. It may be used as an alternative.

For example, the controller 320 may acquire the object item i ₄ and the representative attribute keyword k ₃ pairs to obtain the adjusted object-keyword association corresponding to the object item i ₄ and the representative attribute keyword k ₃ pair when the reception reserved word is C ₂ . by applying the FIG v _3,4 keyword associated adjustment object-default reserved word corresponding to the received reserved Relevancy C ₂ and k ₃ represent attribute keyword pairs to _4,3 w - equivalent to the object that is able to obtain the keyword associated .

In particular, the method of applying association may be a method of multiplying object-association with basic reserved word-keyword association. For example, the controller 320 may acquire the object item i ₄ and the representative attribute keyword k ₃ pairs to obtain the adjusted object-keyword association corresponding to the object item i ₄ and the representative attribute keyword k ₃ pair when the reception reserved word is C ₂ . corresponding to the object - more even base the reserved word corresponding to the received reserved words C ₂ and k ₃ represent attribute keyword pairs to w _4,3-keyword Relevancy v the value (w _4,3 Хv _{3, 2),} multiplied by the _3,2 Can be used to obtain a coordination object-keyword association. In another embodiment, the control unit 320 may apply a function f (w _4,3 ) to which the adjustment object-keyword correlation has a positive correlation with w _4,3 and v _3,2 instead of multiplying. , v _{3, 2} ) may be used to obtain a coordination object-keyword association. In addition, (w _4,3 Хv _3,2 ) can be used as a coordinating object-keyword association and (w _4,3 Хv _{3, 2} ) is applied as a coordinating object-keyword association by applying corrections based on other factors. Both methods can be used.

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 operation 1220, the controller 320 may obtain a reserved word-object association degree by using the accumulated value of the adjusted object-keyword association degree for a specific object item. For example, the controller 320 may generate the reserved word-object association degree corresponding to the received reserved word and the specific object item pair so as to have a positive correlation with a cumulative value of the adjusted object-keyword association degree for a specific object item. Can be set. The reserved word-object associations corresponding to the pair of object items i ₄ and received reserved words C ₂ , for example

It can be obtained using. f (w _{4, j} , v _{j, 2} ) is a coordinated object-keyword association corresponding to the object item i ₄ and the reception reserved word C ₂ , 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 associations that correspond to the object item i ₄ and incoming reserved word C ₂ pairs are

This can be In another example, the reserved word-object associations that correspond to object item i ₄ and incoming reserved word C ₂ pairs are

It can be a value that is corrected by other factors for.

9, in operation 950, the controller 320 may provide an object item according to the reserved word-object association degree corresponding to the received reserved word. For example, if the reserved word-object association corresponding to the received reserved word C ₂ is as shown in Table 3 below, the controller 320 may provide the object items in the order of Table 4 below.

Object item Reserved word-object association with incoming reserved words i ₁ 0.23 i ₂ 0.33 i ₃ 0.99 i ₄ 0.84

order Object item Reserved word-object association with incoming reserved words One i ₃ 0.99 2 i ₄ 0.84 3 i ₂ 0.33 4 i ₁ 0.23

That is, the controller 320 may provide the object items in the order of the high relevance of the reserved word corresponding to the received reserved word. The terminal 200 provided with the object item may 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 a lower order as necessary. The terminal 200 may provide the user with information about the object item i ₃ by using a speaker instead of the display unit 220.

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

13 to 17 may be performed by partially using the processes of FIGS. 5 to 12 or by partially changing the processes of FIGS. 5 to 12. 13 to 17, the descriptions of FIGS. 5 to 12 may be used when necessary.

Referring to FIG. 13, in operation 1310, the controller 320 extracts a representative attribute keyword candidate set from first collection documents. The controller 320 may collect, for example, keywords that frequently appear in documents corresponding to the field of interest in the first set of documents as representative attribute keyword candidate sets. The process of step 1310 may proceed, for example, similarly or similarly to the process of step 510 of FIG. 5. The process of step 1310 may be performed through the same or similar process as that of FIG. 6. The description of the process of FIG. 6 is not repeated again.

In step 1320, the controller 320 sets a reserved word set. For example, the administrator may manually input a set of reserved words. According to a modification, the controller 320 may provide an interface for setting a word / phrase suitable for a reserved word as a reserved word candidate and setting one or more of the reserved word candidates as a reserved word.

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

In operation 1810, the controller 320 acquires the number of appearances in a document set of one language unit or two or more consecutive language units. 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, any one of word, word, morpheme, syllable and letter. The unit divided by various criteria for distinguishing other sentences may be a language unit in this embodiment.

The controller 320 may divide the documents included in each document of the document set by word units before storing in the form of an array or a list before step 1810. According to an exemplary embodiment, the controller 320 deletes words that are meaningless in each word, for example, some words in Korean, descriptive adjectives such as 'yi', 'low', or other words that are not necessary for analysis or in words / arrays or lists. Can be removed. In addition, according to an embodiment, when a word consists of one word, the controller 320 may convert the word into a basic form (or a preset form).

According to a modified example, the control unit 320 may divide the documents included in each document of the document set in word units and store them in an array or a list before step 1810. According to an embodiment, the controller 320 may convert each word into a basic form (or a preset format). According to an exemplary embodiment, the controller 320 may remove meaningless words, for example, some words in Korean, words which are not necessary for an indication adjective such as 'yi', 'low', or other analysis from the array / list.

Modifications are also possible in which the control unit 320 divides a document into morpheme units, syllable units, or character units.

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

A single language unit can be a reserved word. According to a variant embodiment, two or more consecutive language units may also be reserved words. For example, "neat" (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 less likely to appear than a single language unit, in order to ensure that two or more consecutive language units are selected as reserved words, weights or additional scores for two or more consecutive language units when selecting reserved words Can be given. According to a modification, when selecting a reserved word, a reference value for selecting the reserved word may be more generously for two or more consecutive language units. For example, if a single language unit appears at least a times before it can be recommended as a reserved word candidate, two consecutive language units can be set to be recommended as reserved word candidates even though only b times appear, 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 appear less than b. In the following description, two or more consecutive language units are referred to as continuous language units.

In addition, if consecutive language units are recommended / selected as reserved words, consecutive language units included in the reserved words (candidates) or those reserved words (candidates) and shorter than the reserved words (candidates) should not be recommended as reserved words or recommended as reserved words. Penalties may be deducted when the score is calculated. This is to prevent similar reserved words from being selected or recommended. In the following description, the description of the continuous language unit is omitted to simplify the description, but the description of the single language unit may be equally or similarly applied to the continuous language unit.

In operation 1810, the number of appearance of the language unit may be, for example, the number of documents in which the language unit appears. Even if the language unit appears several times in a document, only one occurrence is allowed. According to another embodiment, the case in which the language unit appears several times in a document may be recognized as the number of appearances, and the number of appearances may be the number of appearances of the language unit.

According to another embodiment, when a language unit appears repeatedly two or more times in a document, the second and subsequent appearances may be recognized as lower appearances than the first appearance. Also, the more repeated the language unit appears in a document, the later appearance can be recognized as a lower score. As the number of appearances increases, the score increases, but the slope gradually decreases. For example, a 1 / r square of the number of appearances (r is a real number greater than 1) may be used as the appearance score of the language unit in the document. For example, (log number of appearances) +1 (where 0 is the number of appearances is 0) may be used. In addition, the appearance score of the language unit in a document may be limited not to exceed a preset upper limit. The cumulative score of the language units for all documents may be the appearance score according to the number of appearances of the language unit. This appearance score may be utilized in step 1830.

In the following description, for the sake of convenience, the number of appearances of a language unit is assumed to be the number of documents in which the language unit appears.

In operation 1820, the controller 320 obtains a number of times that an emotional word is located within a predetermined distance from a language unit. The distance between a language unit and an emotional word is, for example, the number of words located between the language unit and the emotional word, the number of words located between the language unit and the emotional word, and the number of characters located between the language unit and the emotional word. It may be determined based on any one or more.

Also, if the language unit and the emotional word belong to different sentences, the controller 320 may determine that the emotional word is not located within a predetermined distance from the language unit regardless of the number of words / phrases / characters located between the language unit and the emotional word. It may be. According to another variation, when the language unit and the emotional word belong to different sentences, the controller 320 may calculate the distance by adding a predetermined number to the distance calculated by excluding the judgment on the sentence. This is because it is preferable to evaluate the distance more than the number of words / phrases / letters located between the emotional words because the probability that the language unit and the emotional word belong to different sentences is relatively high.

Whether or not a specific word (word phrase) is an emotional word can be checked by checking a dictionary of registered emotional words.

The number of times an emotional 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 emotional word are located together within a certain distance. Even if the language unit and the appraiser appear together within a certain distance in a document several times, the number of appearances is recognized only once. According to another embodiment of the present disclosure, when multiple language units are located within a certain distance of a corresponding language unit of a document, all of the various language units may be recognized as the number of times the emotional words are located within a certain distance from the language unit. Hereinafter, a language unit in which an emotional word is located within a predetermined distance is referred to as an emotional word location language unit.

According to another embodiment, when an emotional language location language unit appears repeatedly two or more times in a document, the second and subsequent appearances may be regarded as lower appearances than the first appearance. In addition, the more repeated the appearance of the corresponding language of language position in the document, the later appearance can be recognized as a lower score. As the number of appearances increases, the score increases, but the slope gradually decreases. For example, a 1 / r square of the number of appearances (r is a real number greater than 1) may be used as the appearance score of the language unit of the location of the corresponding word in the document. For example, (log number of appearances) +1 (where 0 is the number of appearances is 0) may be used. In addition, the appearance score of the language unit of the appraiser position in a document may be limited not to exceed a preset upper limit. A cumulative value of appearance scores of the language unit for all documents may be an appearance score according to the number of appearances of the corresponding language unit. This appearance score can then be utilized in step 1830.

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

In the following description, for the sake of convenience, the number of appearances of an emotional language position language unit is assumed to be the number of documents in which an emotional word appears within a predetermined distance from the language unit.

In operation 1830, the controller 320 sets a reserved word candidate in consideration of the number of appearances of the corresponding language unit and the number of times the emotional word is located within a predetermined distance from the language unit.

For example, the controller 320 may multiply the number of appearances of the language unit and the number of times the emotional words are located within a predetermined distance (or according to a calculation method having a positive correlation with two other variables) to convert the emotional word scores. The predetermined number may be set as the reserved word candidates in order of increasing score. Alternatively, a language unit corresponding to a predetermined score or more may be set as a reserved word candidate.

Language unit Appearance Appraiser Position Count score First language unit 3003 1122 3369366 Second language unit 2001 1820 3641820 Third language unit 3121 1300 4057300 4th language unit 200 110 22000

In the example of Table 5, candidates for reserved words may be selected in the order of third language unit-> second language unit-> first language unit-> fourth language unit. If the controller 320 recommends two reserved word candidates, the third language unit and the second language unit may be recommended. If the control unit 320 recommends a language unit having a score of 3 million points or more as a reserved word candidate, the third language unit, the second language unit, and the first language unit are sequentially recommended as reserved word candidates. According to another embodiment of the present disclosure, the controller 320 extracts the first number of language units in order of the number of appearance of the language units, and then based on the number of times that the emotional words appear within a certain distance of the extracted language units (or the score of the emotional words). A predetermined number of reserved word candidates can be extracted in one order. In the example of Table 5, if three language units are extracted in the order of appearance, the first to third language units may be extracted. Based on the number of appearances of the emotional words, the reserved word candidates may be recommended in the order of the second language unit, the third language unit, and the first language unit.

According to another embodiment, the controller 320 extracts the first number of language units in order of the number of appearance of the language units, and then based on the number of times that the emotional words appear within a certain distance of the extracted language units (or the score of the emotional words). The reserved word candidates of the second number (where the second number is less than the first number) may be extracted in the order. In the example of Table 5, if three language units are extracted in the order of appearance, the first to third language units may be extracted. If two language units are extracted based on the number of occurrences of an emotional word, reserved word candidates may be recommended in the order of the second language unit and the third language unit.

The control unit 320 is similar to or slightly different from the above-described method, in which a score for selecting a reserved word candidate is positively correlated with the number of occurrences of the language unit in step 1810, and an emotional word is located within a preset distance in step 1820. A score for selecting a reserved word candidate may be set to have a positive correlation with the number of occurrences of a language unit, and the reserved word candidate may be recommended using this score.

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

In operation 1840, the information providing apparatus 300 provides the terminal 200 with interface information for generating a reserved word selection interface including reserved word candidate information. The interface information can be a document in html format, for example. 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 the dynamic information in a page form previously stored in the terminal 200. To provide users with a page with an interface.

The controller 320 may generate page information for generating a page including a reserved word selection interface including reserved word candidate information, and the communication unit 310 may provide the page information to the terminal 200. The terminal 200 may render and display a page including the corresponding interface to the user. According to a modified embodiment, instead of an interface included in the visual page, an interface by sound or an interface by a technology known or known in the future may be provided. In the following description, it is assumed that an interface included in a visual page is provided for convenience.

19 is an example of an interface 1900 generated according to the interface information provided in step 1840.

Referring to FIG. 19, the interface 1900 includes a table including a checkbox column 1910, a reserved word candidate column 1920, and a detail view column 1930. In addition, the interface 1900 may include a reserved word add button 1940, a candidate delete button 1950, and a put-in box 1960. The user selects the desired reserved word candidate (s) in the checkbox column 1910 and selects one of the add reserved word button 1940, the delete candidate button 1950, and the put in the archive button 1960 to process the reserved word candidate. Can be. When any one button is selected, the terminal 200 may transmit input information obtained by converting a user's input to the information providing apparatus 300.

The information providing apparatus 300 may process the reserved word candidate according to the input information received from the terminal 200. For example, when a user selects a check box 1910 of some reserved word candidates (hereinafter, 'selected candidates') and selects an add reserved word button 1940, the controller of the information providing apparatus 300 that has received input information related thereto ( 320 may add the selected candidates to the reserved word set and delete the selected candidates from the reserved word candidate set. The controller 320 controls not to recommend the language unit (s) included in the reserved word set and the language unit (s) substantially the same as the language unit (s) included in the reserved word set as a reserved word candidate.

According to another example, when a user selects a check box 1910 of some reserved word candidates and selects a reserved word candidate deletion button 1950, the controller 320 of the information providing apparatus 300 that receives the input information related thereto is excluded from the reserved words. Selected candidates may be added to the set and selected candidates may be deleted from the reserved word candidate set. The controller 320 controls not to recommend language unit (s) included in the reserved word exclusion set and language unit (s) substantially the same as the language unit (s) included in the reserved word exclusion set in the future as a reserved word candidate recommendation. do.

According to another example, when a user selects a check box 1910 of a candidate candidate for a reserved word and selects a put button 1960, the controller 320 of the information providing apparatus 300 that receives input information related to the reserved word is a reserved word. Selected candidates may be added to the candidate storage set and selected candidates may be deleted from the reserved word candidate set. The controller 320 may not recommend language unit (s) included in the reserved word candidate storage set and language unit (s) substantially the same as the language unit (s) included in the reserved word candidate storage set when the reserved word candidate is recommended in the future. To prevent it.

Instead of the buttons 1940, 1950, 1960, another interface similar to the button or another interface that can take the role of the button may be utilized.

In addition, the controller 320 may provide an interface for deleting some of the reserved words from the set of reserved words. The controller 320 may provide an interface for removing a language unit from the reserved word exclusion set such that some language units of the reserved word exclusion set are not excluded from the recommendation. The controller 320 may provide a 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 some language units of the reserved word candidate storage set to the reserved word. In addition, the user may include some language units of the reserved word candidate set in the reserved word exclusion set and delete the reserved word candidate set through the list interface. In this case, the corresponding language unit is no longer provided through the list interface of the reserved word candidate storage set and is not recommended as the reserved word candidate through the interface 1900 of FIG. 19. In addition, the user may simply delete some language units of the reserved word candidate storing set from the reserved word candidate storing set through the list interface. In this case, the corresponding language unit is no longer provided through the list interface of the reserved word candidate storage set, but may be recommended as the reserved word candidate through the interface 1900 of FIG. 19.

In addition, the interface 1900 may include a previous page button 1970 and / or a next page button 1980 for moving pages in case the reserved word candidate is not displayed on one page. The previous page button 1970 and / or next page button 1980 may optionally be provided according to the number of actual candidates and the current page position. In addition, an interface that is expanded through scrolling may be provided instead of the previous page button 1970 and / or the next page button 1980. In some interfaces, only tables containing items 1910, 1920, and 1930 may be scrolled and buttons 1940, 1950, 1960, 1970, and 1980 may be excluded from scrolling.

The user may select the detailed view 1930 to inquire in detail about the background or related information for which the reserved word candidate is recommended. When the detailed view 1930 is selected, the interface provided by the controller 320 includes information about the reserved word candidate and the interface for adding the reserved word candidate to the reserved word, adding to the reserved word candidate storage set, or adding the reserved word exclusion set. can do.

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

Returning to FIG. 18, in operation 1850, the controller 320 may add the selected reserved word candidate to the set of reserved words according to an input for selecting the reserved word.

Returning to FIG. 13, in operation 1330, the controller 320 stores an object-keyword association corresponding to the object item and the representative attribute keyword pair.

10 is an illustration of a stored object-keyword association diagram according to one embodiment of the present disclosure.

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

For example, the object-keyword association degree 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 of the present disclosure, the process of step 1330 may be performed by input of an administrator or by receiving an object-keyword association degree 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 descriptions thereof 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. 13. Therefore, even when the embodiments of FIGS. 5 to 8 are used in performing the process of step 1330, the process of step 510 is not performed again and the result of step 1310 may be recycled.

In operation 1340, the control unit 320 uses the associated weight corresponding to the representative attribute keyword and the lower keyword pair, and the basic reserved word corresponding to the reserved word and the representative attribute keyword pair using the default reserved word-sub keyword association degree corresponding to the reserved word and the lower keyword pair. Store the keyword association in the storage unit 330. For example, the process of step 1340 may be performed according to an input of an administrator or according to one or more embodiments of FIGS. 15 to 17.

Before the process of step 1340, or during the process of step 1340, the sub-keyword is determined, the association weight corresponding to the representative attribute keyword and the sub-keyword pair is determined, and the default reserved word-sub-keyword association degree corresponding to the reserved word and the sub-keyword pair. Should be determined.

The lower 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 subkeyword of step 520 may be used in step 1340. If the lower keyword is not determined in operation 1330, the lower keyword may be determined through the same or similar process as that of the operation 520 and the previous process of FIG. 5.

The association weight 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 weight of step 530 may be used in step 1340. If the association weight is not determined in operation 1330, the association weight may be determined through a process that is the same as or similar to that of operation 530 and the process of FIG. 5.

The basic reserved word-sub keyword association may be calculated, for example, by considering the frequency of the reserved word and the lower keyword appearing in the same or similar context.

Hereinafter, an example of obtaining the basic reserved word-keyword association degree v _3,2 between the reserved word C ₂ and the representative attribute keyword k ₃ will be described in the description of FIGS. 15 to 17. For example, it is assumed that sub keywords of the representative attribute keyword k ₃ are B3 ₁ to B3 ₅₀ . For a reserved word, a representative attribute keyword, and a basic reserved word-keyword association, refer to the example described above with reference to FIG. 11. A basic reserved word-sub keyword association corresponding to the reserved word C _j and the lower keyword Bg _h pairs is expressed as x _{j, h} . An association weight corresponding to the lower keyword Bg _h and the representative attribute keyword k _g pair is expressed as y _{g, h} . The adjusted reserved word-subkey word association degree corresponding to the reserved word C _j , the representative attribute keyword k _g, and the lower keyword Bg _h combination is expressed as x _{j, g, h} .

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

Referring to FIG. 15, in step 1510, the controller 320 obtains an adjusted reserved word-sub keyword association degree by applying an association weight to a basic reserved word-sub keyword association degree.

In operation 1510, the control unit 320 applies an association weight corresponding to the sub-keyword and the representative attribute keyword pair to a basic reserved word-sub-keyword association degree corresponding to the reserved word and the sub-keyword pair for each reserved word and the sub-keyword pair. The adjusted reserved word-sub keyword association degree corresponding to each reserved word and sub keyword pair for the representative attribute keyword may be obtained.

For example, when the reserved word is C ₂ , the controller 320 may include C ₂ − − corresponding to the lower keyword B _{3 4} and the representative attribute keyword k ₃ pair. 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 The adjusted reserved word- _subkeyword association degree x _2,3,4 may be obtained by applying the corresponding association weight y _3,4 .

In particular, the method of applying the association weight may be a method of multiplying the basic reserved word-sub keyword association degree x _2,4 by the association weight y _3,4 corresponding to the lower keyword B3 ₄ and the representative attribute keyword k ₃ pairs. For example, when the reserved word is C ₂ , the controller 320 may include C ₂ − − corresponding to the lower keyword B _{3 4} and the representative attribute keyword k ₃ pair. B3 ₄ -adjusted Reserved word-subkey word association x _2,3,4 in order to obtain reserved keyword C ₂ and subkeyword B3 ₄ base reserved word-subkey word association x _2,4 corresponding to _subkeyword B3 ₄ and representative The adjusted reserved word-subkeyword association degree x _2,3,4 may be obtained using (x _2,4 Хy _{3, 4} ), which is a value multiplied by the association weight y _3,4 corresponding to the attribute keyword k ₃ pairs. In another embodiment, the control unit 320 may apply another operation / use scheme such that the adjusted reserved word-sub keyword association degree x _2,3,4 has a positive correlation with x _2,4 and y _3,4 instead of multiplication. The adjusted reserved word- _subkeyword association degree x _2,3,4 may be obtained using the function f (x _{2, 4} , y _{3 , 4} ). We can also use (x _2,4 Хy _{3, 4} ) as an adjusted reserved word-subkey association x _2,3,4 and add a correction based on other factors to (x _2,4 Хy _{3, 4} ) _-Any sub-keyword association x _2,3,4 can be used.

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 operation 1520, the controller 320 may set a basic reserved word-keyword association degree between the reserved word C ₂ and the representative attribute keyword k ₃ by using the cumulative value of the adjusted reserved word-sub keyword association degree x _{2, 3, f} . That is, the default reserved word-keyword association between the reserved word C ₂ and the representative attribute keyword k ₃ is

Can be 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 ,} to obtain a ₂ x _{f, 3, f} by accumulating the reserved words C ₂ and the representative keyword attribute default reserved word between k ₃ - can be obtained also associated keywords. According to another embodiment, the default reserved word-keyword association between the reserved word C ₂ and the representative attribute keyword k ₃ is

This can be a value added by correction by other factors. According to another embodiment, the default reserved word-keyword association between the reserved word C ₂ and the representative attribute keyword k ₃ is

Can be a positively correlated value. Here, 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 may be different from 50 in the formula.

16 is a detailed flowchart of step 1340 according to another embodiment of the present disclosure. Processes of steps 1510 and 1520 of FIG. 16 are the same as those described above with reference to FIG. 15 and thus will not be repeated.

Referring to FIG. 16, in operation 1530, the controller 320 may determine a representative keyword whose basic reserved word-keyword association degree of the reserved word and the corresponding representative keyword pair among the representative keywords corresponding to a specific reserved word is equal to or less than a reference basic reserved word-keyword association degree. You can delete the default reserved word-keyword associations. Representative keywords corresponding to a specific reserved word indicate keywords in which a basic reserved word-keyword association degree is set in relation to the specific reserved word. The reference default reserved word-keyword association may be preset. According to another embodiment, the reference basic reserved word-keyword association is set by using an average value of the basic reserved word-keyword association corresponding to the specific reserved word or when the basic reserved word-keyword association corresponding to the specific reserved word is arranged in size order. The default reserved word-keyword association degree of the rank may be set. The reference basic reserved word-keyword association may be a specific value having a positive correlation with the basic reserved word-keyword associations corresponding to other specific reserved words. The reference basic reserved word-keyword association degree may be set differently according to the reserved word or may be set identically for all reserved words. If the default reserved word-keyword association is deleted, it is set as no association between the reserved word and the representative keyword. The control unit may set the association degree to 0, and the base reserved word-keyword association degree of the reserved word and the corresponding representative keyword pair in the list indicating the basic reserved word-keyword association degree (which may be replaced by an array or other data structure). The default reserved word-keyword association may be deleted by deleting the related information. A method of adding information indicating that the other basic reserved word-keyword association has been deleted (or the association has been deleted) may be used.

In step 1530, the default reserved word-keyword association with a relatively fine association is deleted (i.e., set as no association) to perform an overly complex operation or to make the association practically meaningless to the user / administrator. It can be suppressed.

17 is a detailed flowchart of step 1340 according to another embodiment of the present disclosure.

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

In operation 1540, the controller 320 may normalize the basic reserved word-keyword association that remains without being deleted. For example, in order to include the average value of the default reserved word-keyword association associated with a specific representative attribute keyword in a specific range, the default reserved word-keyword association associated with a specific representative attribute keyword is multiplied by a predetermined coefficient or raised. Can be reduced. In another example, the base reserved word-keyword association associated with a specific representative keyword is stored by multiplying by a predetermined coefficient so that the sum of the basic reserved word-keyword associations stored and linked to the specific representative attribute keyword is included in a specific range. Can be reduced or reduced. That is, in order to prevent a case in which a specific representative attribute keyword has a high default reserved word-keyword association, and no reserved word is selected, only that specific representative attribute keyword is recommended / used or a specific representative attribute keyword is rarely used. Appropriate adjustments can be made.

According to another embodiment, in step 1540, instead of multiplying the basic reserved word-keyword association by a predetermined coefficient, the controller 320 may add a predetermined coefficient or combine operations such as logarithmic and squared to perform normalization. According to another embodiment, the normalization may be performed by partially reducing only basic reserved word-keyword associations above a certain reference value or partially increasing only basic reserved word-keyword associations below a certain reference value.

According to another embodiment, in operation 1540, the controller 320 may perform normalization to include an average value (or sum) of basic reserved word-keyword associations stored in connection with a specific reserved word in a specific range.

A modified example in which step 1530 is omitted in the process of FIG. 17 and a basic reserved word-keyword association degree normalization process is performed is also possible.

In operation 1350, the communication unit 310 receives and acquires a reserved reserved word from the terminal 200, and transmits the reserved reserved word to the control unit 320.

The reserved reserved 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 the converted voice signal to the search apparatus 300. The controller 320 of the search apparatus 300 may analyze the voice signal, convert the voice signal into text, and match the converted text with the reserved word. The controller 320 may also analyze the speech signal and analyze the speech intonation, height, speed, respiratory state, and the like, and use the context information.

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

In operation 1360, the controller 320 obtains a reserved word-object association corresponding to the received reserved word and each object item pair by using an object-keyword association and a basic reserved word-keyword association. The process of step 1360 may be performed according to the method of step 940 of FIG. 9 or the method of FIG. 12. The description of the same content is omitted.

In operation 1370, the controller 320 may provide an object item according to a reserved word-object association degree corresponding to the received reserved word. The process of step 1370 may be performed in the same manner as the process of step 950. The same description is omitted.

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

Since steps 1310, 1320, 1330, 1340, 1350, 1360, and 1370 of FIG. 14 are the same as steps 1310, 1320, 1330, 1340, 1350, 1360, and 1370 of FIG. 13, the same description is not repeated.

Step 1333 added 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 concurrently / in parallel with step 1330 and the process of step 1333 may be performed during step 1330.

In operation 1333, the controller 320 stores an association weight between the representative attribute keyword and the lower keyword. If the process of setting the association weight is not performed in step 1330, the association weight between the representative attribute keyword and the lower keyword may be set through a process similar to the process of step 530 of FIG. 5. According to another embodiment, the controller 320 may store the association weight between the representative attribute keyword and the lower keyword in a manner of calling the association weight set in operation 1330.

In operation 1337, the controller 320 obtains a basic reserved word-subkeyword association degree between the lower keyword and the reserved word.

For example, the controller 320 may determine an association degree between the reserved word and the lower keyword in consideration of the frequency of the lower keyword appearing in the same or similar context as the reserved word. For example, words appearing around keyword A in a particular sentence may appear around words associated with keyword A in other documents.

"I went on a trip with a big heart, but I suffered because the weather was so hot in July."

"I went on a trip with a big heart, but the weather was too humid for July."

Looking at the above two sentences, the word "hot" is replaced with "wet" in the same context. The controller 320 may infer that "hot" and "wet" are words associated with each other.

"I went on a trip with a big heart, but I suffered because the weather was so hot in July."

"I went on vacation after eating a big heart, but I suffered because the weather was so hot in July."

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

"I went on a trip with a big heart, but I suffered because the weather was so hot in July."

"I went on a trip with a big heart, but I suffered because the weather was so hot in August."

Similarly, the control unit 320 may infer that the words "July" and "August" are related through the above two sentences.

The controller 320 indicates that words "hot" and "wet" are related words, "July" and "August" are related words, and "travel" and "vacation" are words related to each other through previously collected documents. You can save it.

"I went on vacation with a big heart, but the weather was too humid for July."

"I went on a trip with a big heart but it was hard because the weather was so hot in August."

If the two sentences are not in the same context but know that "hot" and "wet" are related words, "July" and "August" are related words, and "travel" and "vacation" are words related to each other, the control unit 320 ) Can be learned from the above sentence that "toil" and "tough" are also related words.

Keyword pairs with high frequency appearing in the same / similar context may be determined to be highly related to each other. In addition, it may be determined that the higher the similarity of the context in which the two keywords appear, the higher the correlation between the two keywords. The controller 320 proceeds with learning using the collected documents to set the degree of association between keywords, and sets the degree of association between keywords appearing in the sentence using the context between the set keywords and the context. The correlation between the two can increase the accuracy of the judgment.

Similar learning methods are known as the Neural Net Language Model (NNLM), Recurrent Neural Net Language Model (RNNNLM), word2vec, skipgram, and Continuous Bag-of-Words (CBOW). In particular, when using word2vec, word2vec learns from documents and maps each keyword to a vector, and the similarity between two keywords can be determined by calculating the cosine similarity of the two vectors.

20 is a diagram illustrating a term hierarchy according to an embodiment of the present specification.

When the process of FIG. 13 or 14 is completed, a hierarchical relationship between reserved words C ₁ to C _q , representative attribute keywords k ₁ to k _n , and sub keywords BX ₁ to BX ₅₀ is set. .

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 sub-keyword. Using such a hierarchical relationship, the controller 320 may recommend an appropriate object or select a new reserved word candidate according to the reserved word. In addition, the hierarchical relationship of FIG. 20 may reflect new materials, and thus, the hierarchical relationship may be modified or supplemented through learning while repeating the processes of FIGS. 13 and 14.

21 is a flowchart illustrating a virtual input interface program installation process according to an embodiment of the present invention.

In step 2110, the terminal 200 transmits a package request message to the information providing apparatus 300. The package request message is a message for requesting a virtual input interface program. In operation 2120, the information providing apparatus 300 delivers a virtual input interface program (package) to the terminal 200.

In step 2130, the terminal 200 installs the received virtual input interface program (package). If installation is unnecessary and the received virtual input interface program can be executed as it is, step 2130 may be omitted.

22 is a diagram illustrating an interface page of a package that performs a voice input processing method according to an embodiment of the present invention.

4, 21 and 22, the interface page for performing the voice input processing method according to an embodiment of the present invention is the interface page of the package installed in the terminal in accordance with the package installation 2130 in the terminal of FIG. Can be.

Referring to FIG. 22, an interface page 2200 for performing a voice input processing method according to an embodiment of the present invention includes a title display unit 2201, a popular key talk display unit 2202, a field display unit 2203, and a text input unit ( 2204 and integrated voice input button 2205.

The display unit 220 of the terminal 200 displays the interface page 2200.

The title display portion 2201 displays the title of the package. In addition, the title display unit 2201 displays a notice or a hint about key talk. The hint for the key talk may be a key value for the property and a setting value for the property set for each field.

For example, if the field is a star, the set values for the attributes are appearance, personality, mood, intelligence, singularity, coinage, etc., and the keystrokes for appearance are "handsome", "cute", "gorgeous", "dazzling" "," Noticeable "," pretty "," elegant "," sensuous "," sensual "," shiny "," chubby "," clean "," fresh "," beautiful "and the like.

 The popular key talk display unit 2202 displays a key key and a field to which the key talk is currently applied. For example, the popular key talk display unit 2202 may display the key talk "for couples" and the field "movie" to which the key talk is applied. Popularity may be determined using, for example, information on the frequency of selection of the service users or popular search words of other services.

The field display unit 2203 displays a field. The field may be an item from which information is extracted based on key talk. For example, star, film, broadcasting, webtoon, restaurant, lodging, cooking, wine, art, beer, travel (domestic), activities (domestic), clothes, cosmetics, hair, gifts, music, books, jewelry, Items such as watches, fashion accessories, eyewear, shoes, convenience foods, convenience facilities, franchise restaurants, coffee shops, bags, perfumes, and the like. The field display unit 2203 may display an image of the field together with the field, and displays a field list including a small image or text that is not shown in the drawing but may be selected by the user.

The text input unit 2204 receives the text for the key talk from the user instead of the voice or displays the key talk set by the user as a favorite.

   When the integrated voice input button 2205 is selected, the terminal 200 in which the package for executing the voice input processing method according to an embodiment of the present invention is installed performs an operation for receiving a voice input from the user. That is, the controller 250 of the terminal 200 operates the input unit 210 and indicates that the voice input is in operation through the display unit 220. For example, the terminal 200 may indicate that the voice input is operating to the user by listening to, for example. The input unit 210 may include a voice input device such as a microphone. The input unit 210 of the terminal 200 receives an input word by voice from a user, and the input word may be a word, a phrase, or a sentence.

The terminal 200 installed with the package recognizes the input voice and converts it into text. Recognizing the input voice and converting the input voice into text is performed by the controller 250 of the terminal 200 in which the package is installed or the information providing apparatus 300 receives data about the voice input to the terminal 200 and receives the voice. After recognizing and converting the text, the converted text may be transmitted to the terminal 200.

The control unit 250 of the terminal 200 in which the package is installed extracts a field related word or a key talk from the converted text. For example, if the user inputs a voice saying, "Is there a delicious and clean restaurant", the control unit 250 of the terminal 200 in which the package is installed has a key talk "clean" and a key talk "delicious" in the user input by voice. Extract "delicious" similar to and extract the "delicious" corresponding to the field.

The terminal 200 installed with the package extracts information corresponding to each field based on the extracted key talk. The extracted information may be object information corresponding to each field.

The terminal 200 may display object information corresponding to each field for each field. For example, if the field is "delicious", restaurant information in the "delicious" field provided by the information providing device 300 is displayed, if the field is "celebrity" image and name of the actor, talent, singer, etc. Can be displayed.

FIG. 23 is a diagram illustrating an interface page displaying a result of inputting a voice by selecting the integrated voice input button of FIG. 22.

The interface page 2300 includes a result goal display unit 2301, a recommendation key talk display unit 2302, a result display unit 2303, a text input unit 2204, and an integrated voice input button 2205.

The result target time section 2301 displays a title indicating that the search result.

Since the text input unit 2204 and the integrated voice input button 2205 have been described with reference to FIG. 22, repeated descriptions thereof will be omitted.

The recommended key torque display portion 2302 displays the recommended key torque.

For example, if a user inputs a voice input from a user saying "Is there a delicious and clean restaurant" and outputs a result, a key talk such as "a lot of menus" or "a lot to see nearby" may be recommended. In addition, the recommended key talk and the recommended field may be displayed together.

The result display unit 2303 includes a field display unit 2306 and a full view button 2037.

The field display unit 2306 displays a field arranged by priority and the number of objects searched in relation to the field. The highest priority sector may be placed at the top or leftmost. The user may scroll upwards or scroll leftwards while touching each result display unit 2303 to search for results for other fields.

If there is a term directly pointing to a field in the voice input input by the user, the field has the highest priority, and the result display unit of the field is disposed at the top or left side.

If there is no field searched after the field input by the user or a field input by the user, the searched field is arranged according to the relevance to the key talk input by the user.

Keytalk is set separately for each field. For example, in the field of "Gourmet", key talks such as "Refreshing, delicious, simple, fun, exotic," and "Star", in the field, Keytalk can be set.

Document pools of basic documents used to set keytalk for each field are managed separately for each field, and keytalk, representative attribute keywords, and subkeywords are extracted and set from document pools of the corresponding base documents for each field. Alternatively, the administrator can set keytalk for each field.

Since the method of extracting and setting the key talk, the representative attribute keyword, and the sub-keyword, respectively, from the document pool of the basic documents is as described above, a description thereof will be omitted.

Document pools of basic documents in each field may be identical to each other or may be different for each field.

As more keytalks are input by the user in the field, the field is placed on the upper or left side.

The control unit 320 of the information providing apparatus 300 calculates the association degree reserved word-keyword association degree and the association degree object-keyword association degree between the basic key talk and the representative attribute keyword described above for each field, Using this, keyword and reserved word-object association corresponding to each object item pair are obtained.

The controller 320 of the information providing apparatus 300 may provide the object items to the terminal 200 in order of the high degree of the reserved word-object association according to the reserved word-object association degree obtained for each field.

When the user selects the all view button 2307, all the results for the corresponding field are output.

FIG. 24 is a diagram illustrating an interface page displayed when the entire display button is selected in the result display unit of FIG. 23.

The interface page 2400 includes a field target time unit 2401, a recommendation key torque display unit 2302, a result display unit 2403, a text input unit 2204 and an integrated voice input button 2205, a rescan button 2404, and additional input search. A button 2405 and a choice display portion 2406 are included.

The field target time section 2401 displays a field in which a result is found.

Since the recommended key torque display unit 2302 has been described with reference to FIG. 23, repeated description thereof will be omitted.

The option display unit 2406 displays options for outputting a search result for a field as a user's selection. For example, the option display unit 2406 may display options such as "inside", "menu", "taste", and the like, when "inside" is selected, outputs restaurants around me. In addition, when "menu" is selected as an option, a result of additionally reflecting items such as "light", "ulk", "sharp", "savory", and "spicy" may be output.

Options may vary by sector. For example, options such as “menu” and “taste” may be displayed in the restaurant area, but options such as “stay type” and “subsidiary facilities” may be displayed in the accommodation field.

These options may be set by the administrator or by the controller 320 of the information providing apparatus 300.

The user selects the rescan button 2404 to input a new key talk corresponding to the current field by voice. In this case, the control unit 250 of the terminal 200 in which the package is installed converts the voice input by the user into text as described above, extracts the key torque, and the key input by the user only in the current field without separate judgment on the field. Use torque to request and print information. For example, if the current field is the "Gourmet" field outputted by the user entering "Is there a delicious and neat restaurant", the user selects the rescan button 2404 and enters "big and tasty stew" and the package is installed. The terminal 200 may request the information providing apparatus 300 for information related to the key talks “ulk” and “tasteful” in the current “delicious” field, and receive and output the result.

In addition, the terminal 200 does not convert the voice input by the user into text and extracts the key talk, but converts the voice input by the user into text and transmits the converted text to the information providing apparatus 300 or input by the user. Transmitting one voice data to the information providing apparatus 300, the information providing apparatus 300 converts the received voice data into text, and then the information providing apparatus 300 is the text or information providing apparatus converted by the terminal 200 Keytalk may be extracted from the text converted by 300, and information related to the keytalk may be retrieved.

The user may select an additional input search button 2405 to input additional key talk corresponding to the current field by voice. In this case, the controller 250 of the terminal 200 in which the package is installed converts the voice input by the user into text as described above, extracts the key talk, and the user inputs the existing field only in the current field without any judgment on the field. By adding a key torque extracted from a voice additionally input by a user to a key talk, the information providing apparatus 300 may request information related to the key talk, and receive and output the result.

For example, if the current field is the "Gourmet" field outputted by the user inputting "Is there a delicious and clean restaurant", the user selects the additional input search button 2405 and additionally enters "Plain Noodle" and the package is installed. The terminal may extract and output information related to keytalks "delicious", "clean", "light" and "noodle" in the current "delicious" field.

In addition, the terminal 200 does not convert the voice input by the user into text and extract the key talk, but instead converts the voice input by the user into text and transmits the converted text to the information providing apparatus 300 or input by the user. The text or information providing apparatus converted by the terminal 200 after the voice data is transmitted to the information providing apparatus 300 and the information providing apparatus 300 converts the received voice data into text. The keytor may be extracted from the text converted by 300, and the newly extracted keytalk and information related to the existing keytalk may be searched.

25 is a flowchart illustrating a voice input processing method according to an embodiment of the present invention.

Referring to FIG. 25, in step S2000, the controller 250 of the terminal 200 determines whether the integrated input button is selected.

When the integrated input button is selected, the input unit 210 of the terminal 200 receives a voice input from the user in step S2501.

In operation S2502, the controller 250 of the terminal 200 converts the received voice input into text. Alternatively, the controller 250 of the terminal 200 may transmit the received voice input to the information providing apparatus 300 and receive a result of the information providing apparatus 300 converted into text to extract key talk.

Keytalk and field related words are extracted from the converted text in step S2503.

In operation S2504, the control unit 250 of the terminal 200 requests the information providing apparatus 300 for key talk related information, receives the result, and displays the result on the terminal 200. Alternatively, the information providing apparatus 300 converts a voice input received from the terminal 200 into text, extracts a keytalk and a field related word from the converted text, searches for information related to the extracted keytalk and the field, and then the result. May be transmitted to the terminal 200.

Received results are displayed by sector.

In operation S2505, it is determined whether the information all view button belonging to the specific field is selected. When the overview button is selected, the control unit 300 of the terminal 200 displays all information corresponding to the field in step S2506 and displays a re-search button, an integrated input button, and an additional input search button.

In step S2507, it is determined whether the rescan button is selected.

If the re-search button is selected, the terminal 200 receives a voice input from the user in step S2508.

In step S2509, the control unit 250 of the terminal 200 converts the input voice input into text and extracts the key talk. Alternatively, the controller 250 of the terminal 200 may transmit the received voice input to the information providing apparatus 300 and receive a result of the information providing apparatus 300 converted into text to extract key talk.

In operation S2510, the terminal 200 requests the information providing apparatus for the key talk related information extracted for the corresponding field, and displays the result on the terminal 200. Alternatively, the information providing apparatus 300 converts a voice input received from the terminal 200 into text, extracts a keytalk from the converted text, searches for keytalk related information extracted only from a corresponding field, and then outputs the result to the terminal 200. Can also be sent.

If the rescan button is not selected in step S2507, it is determined whether the additional input search button is selected in step S2511.

If the additional input search button is selected, the terminal 200 receives a voice input from the user in step S2512.

In step S2513, the control unit 250 of the terminal 200 converts the input voice into text and extracts the key talk. Alternatively, the control unit 250 of the terminal 200 may transmit the received voice input to the information providing apparatus 300 and receive a result of the information providing apparatus 300 converting the text to extract the key talk.

In step S2514, the terminal 200 requests the information providing apparatus for the information related to the extracted key talk and the existing key talk, and displays the result on the terminal 200. Alternatively, the information providing apparatus 300 converts the voice input received from the terminal 200 into text, extracts the key talk from the converted text, searches for the key torque related information extracted only from the corresponding field, and then outputs the result to the terminal 200. Can also be sent.

At this point, it will be understood that each block of the flowchart illustrations and combinations of flowchart illustrations may be performed by computer program instructions. These computer program instructions may be mounted on a processor of a general purpose computer, special purpose computer, or other programmable data processing equipment, such that the instructions being performed through the processor of the computer or other programmable data processing equipment are described in the flowchart block (s). It creates a means to perform the functions. These computer program instructions may be stored in a computer usable or computer readable memory that can be directed to a computer or other programmable data processing equipment to implement functionality in a particular manner, and thus the computer usable or computer readable memory. It is also possible for the instructions stored in to produce an article of manufacture containing instruction means for performing the functions described in the flowchart block (s). Computer program instructions may also be mounted on a computer or other programmable data processing equipment, such that a series of operating steps may be performed on the computer or other programmable data processing equipment to create a computer-implemented process to create a computer or other programmable data. Instructions for performing the processing equipment may also provide steps for performing the functions described in the flowchart block (s).

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

In this case, the term '~ part' used in the present embodiment refers to software or a hardware component such as an FPGA or an ASIC, and '~ part' performs certain roles. However, '~' is not meant to be limited to software or hardware. '~ Portion' may be configured to be in an addressable storage medium or may be configured to play one or more processors. Thus, as an example, '~' means components such as software components, object-oriented software components, class components, and task components, and processes, functions, properties, procedures, and the like. Subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables. The functionality provided within the components and the 'parts' may be combined into a smaller number of components and the 'parts' or further separated into additional components and the 'parts'. In addition, the components and '~' may be implemented to play one or more CPUs in the device or secure multimedia card.

Those skilled in the art will appreciate that the present disclosure may be embodied in other specific forms without changing the technical spirit or essential features thereof. Therefore, it should be understood that the embodiments described above are exemplary in all respects and not restrictive. The scope of the present specification is indicated by the following claims rather than the detailed description, and all changes or modifications derived from the meaning and scope of the claims and the equivalent concept are included in the scope of the present specification. Should be interpreted.

On the other hand, the present specification and drawings have been described with respect to the preferred embodiments of the present specification, although specific terms are used, it is only used in a general sense to easily describe the technical content of the present specification and to help the understanding of the invention, 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 spirit of the present disclosure can be implemented in addition to the embodiments disclosed herein.

Claims (18)

A display unit displaying an integrated input button for receiving a voice input;
An input unit configured to receive the voice input from a user when the integrated input button is selected;
And a controller for converting the input voice into text, extracting key talk and field-related words from the text, and requesting an information providing device for information related to the key talk.
And the information related to the key talk is displayed for each field. The method of claim 1,
And the key talk is set for each field. The method of claim 1,
And information related to the key talk is displayed together with a view button for the field and information corresponding to the field. The method of claim 3, wherein
When the field is selected or the all view button is selected, all information, a re-search button, and an additional input search button of the field are additionally displayed. The method of claim 4, wherein
And the input unit receives a voice input from the user when the re-search button or the additional input search button is selected. The method of claim 5,
And when the rescan button is selected, the controller converts the input voice into text and extracts key talk from the text. The method of claim 6,
And the control unit requests the information providing device for information related to the key talk in the field. The method of claim 5,
And when the additional input search button is selected, the controller converts the input voice into text and extracts additional key talk from the text. The method of claim 8,
And the controller requests the information providing device for information related to the key talk and the additional key talk in the field. Displaying an integrated input button for receiving a voice input;
Receiving the voice input from a user when the integrated input button is selected;
Converting the input voice into text, extracting a keytalk and a field related word from the text, and requesting an information providing device for information related to the keytalk;
And the information related to the key talk is displayed for each field. The method of claim 9,
And the key talk is set for each field. The method of claim 9,
And the information related to the key talk is displayed together with a view button for the field and information corresponding to the field. The method of claim 12,
The voice input processing method of selecting the field or selecting the all view button further displays all information of the field, a research button, and an additional input search button. The method of claim 13,
And receiving the voice input from the user when the re-search button or the additional input search button is selected. The method of claim 14,
And when the rescan button is selected, converts the input voice into text and extracts key talk from the text. The method of claim 15,
The voice input processing method of requesting the information providing device for information related to the key talk in the field. The method of claim 14,
And when the additional input search button is selected, converts the input voice into text and extracts additional key talk from the text. The method of claim 17,
And requesting information related to the key talk and the additional key talk from the field.

Images