KR102192376B1 - Method and system for extracting foreign synonym using transliteration model - Google Patents

Abstract

Disclosed is a method and system for automatically extracting foreign synonyms using a transliteration model. The synonym extraction method includes: extracting a keyword list related to the target keyword for the target keyword; And extracting a synonym corresponding to the original language notation of the target keyword from the keyword list using a transliteration model.

Description

Method and system for automatically extracting foreign synonyms using transliteration model {METHOD AND SYSTEM FOR EXTRACTING FOREIGN SYNONYM USING TRANSLITERATION MODEL}

The following description relates to a technology for extracting synonyms, and in particular, to a technology for extracting the original language notation of a foreign word.

In general, when a user wants to collect information through the Internet, he accesses a server providing information and obtains the information through search.

If you want to purchase a specific product, you can get information about prices or characteristics using a search site, or you can search for a website that sells the product.

In order to search for a product, a product registration process including a process of matching product models and categories is required. For example, Korean Patent Publication No. 10-0705410 (registration date April 03, 2007) discloses a technology for registering product information by automatically matching product models and categories.

Provides a method and system for constructing a multi-graph structure representing the relationship between keywords for keywords extracted from product information, and automatically extracting the original language notation as a synonym for foreign words and adding them to the multi-graph.

Provides a method and system for extracting the original language notation of the corresponding foreign word from words appearing in a product name or product-related query using a transliteration model for foreign words and processing them as synonyms.

A synonym extraction method executed in a computer system, wherein the computer system includes at least one processor configured to execute computer readable instructions contained in a memory, and the synonym extraction method comprises: by the at least one processor, an object Extracting a keyword list related to the target keyword for the keyword; And extracting, by the at least one processor, a synonym corresponding to the original language notation of the target keyword from the keyword list using a transliteration model.

According to an aspect, the extracting of the keyword list includes: extracting at least one of a product name list clicked after entering the target keyword as a search query, another query list related to each click product, and a tag list related to each click product can do.

According to another aspect, in the step of extracting the synonym, the synonym may be extracted using a transliteration model using a minimum unit of pronunciation.

According to another aspect, the extracting the synonym may include separating the target keyword into a minimum unit of pronunciation using a data format of LBG (linked bi-gramed).

According to another aspect, the extracting of the synonym may include extracting, as the synonym, a word matching the target keyword among words extracted from the keyword list through the transliteration model.

According to another aspect, the extracting of the synonym may further include extracting a synonym candidate by substituting the matched word for the target keyword.

According to another aspect, the extracting the synonym may further include extracting a word of a specific language from the keyword list.

According to another aspect, the step of extracting the synonym may further include separating the target keyword into a single word when the target keyword is composed of a compound word.

According to another aspect, in the step of extracting the synonym, at least one of a product name list clicked after entering the target keyword as a search query, another query list related to each click product, and a tag list related to each click product is used. It may further include the step of verifying the synonym.

According to another aspect, the extracting of the synonym includes at least one of a product name list clicked after entering the target keyword as a search query, another query list related to each click product, and a tag list related to each click product. The method may further include determining the synonym based on whether or not a word matching the target keyword appears or the frequency of appearance.

According to another aspect, the method for extracting synonyms includes: generating, by the at least one processor, a multi-graph indicating a relationship between keywords in a product dictionary; And additionally registering the synonym in the multi-graph by the at least one processor.

According to another aspect, the generating may include converting product information matching a corresponding product for each product included in the product dictionary into a keyword; And extracting relationship data between information matching the product from the product dictionary as relationship data between keywords.

According to another aspect, the step of registering additionally includes adding a relationship between the target keyword and the synonym on the multi-graph, or entering the target keyword among keywords on the multi-graph as a search query and clicking the product name A relationship with keywords appearing together with the synonym may be added to at least one of a list, another query list related to each click product, and a tag list related to each click product.

A non-transitory computer-readable recording medium in which a program for executing the synonym extraction method is recorded on a computer is provided.

A computer system comprising: at least one processor configured to execute computer-readable instructions contained in a memory, wherein the at least one processor extracts a keyword list related to the target keyword for the target keyword, and then generates a transliteration model. A computer system including a synonym extractor for extracting synonyms corresponding to the original language representation of the target keyword from the keyword list is provided.

According to embodiments of the present invention, it is possible to construct a multi-graph structure representing the relationship between keywords with respect to keywords extracted from product information, and in the case of foreign words, it is possible to automatically extract the original language notation as a synonym and add it to the multi-graph.

According to embodiments of the present invention, the original language notation of the foreign word may be extracted from words appearing in a product name or product-related query using a transliteration model for the foreign word and processed as synonyms.

1 is a diagram showing an example of a network environment according to an embodiment of the present invention.
2 is a block diagram illustrating an internal configuration of an electronic device and a server according to an embodiment of the present invention.
3 is a block diagram illustrating an example of components that may be included in a processor of a server according to an embodiment of the present invention.
4 is a flowchart showing an example of a method that a server can perform according to an embodiment of the present invention.
5 shows an example of the configuration of a product dictionary for keyword management in an embodiment of the present invention.
6 is an exemplary diagram for explaining an example of a process of converting product information into keywords according to an embodiment of the present invention.
7 is an exemplary diagram for explaining an example of a process of constructing a multi-graph showing a relationship between keywords in an embodiment of the present invention.
8 shows an example expressed in Korean vector numeric characters in an embodiment of the present invention.
9 is an exemplary diagram for explaining a separation method using an LBG data format in an embodiment of the present invention.
10 is an exemplary diagram illustrating a process of extracting a data set in units of MWS according to an embodiment of the present invention.
11 to 12 are exemplary diagrams for explaining a sound range processing process according to an embodiment of the present invention.
13 to 16 are exemplary views for explaining an English synonym extraction process using a transliteration model in an embodiment of the present invention.
17 is an exemplary diagram for explaining a process of adding an original language notation synonym to a multi-graph according to an embodiment of the present invention.

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

Embodiments of the present invention relate to a technique for automatically extracting the original language notation of a foreign word as a synonym.

Embodiments including those specifically disclosed in this specification can automatically extract the original language notation of a foreign language and process it as a synonym, and through this, significant in terms of keyword management automation, work resource reduction, service quality improvement, service expansion, etc. Advantages can be achieved.

1 is a diagram showing an example of a network environment according to an embodiment of the present invention. The network environment of FIG. 1 shows an example including a plurality of electronic devices 110, 120, 130, and 140, a plurality of servers 150 and 160, and a network 170. 1 is an example for explaining the present invention, and the number of electronic devices or servers is not limited as in FIG. 1.

The plurality of electronic devices 110, 120, 130, and 140 may be a fixed terminal implemented as a computer system or a mobile terminal. Examples of the plurality of electronic devices 110, 120, 130, 140 include smart phones, mobile phones, navigation systems, computers, notebook computers, digital broadcasting terminals, personal digital assistants (PDAs), portable multimedia players (PMPs). ), tablet PC, game console, wearable device, internet of things (IoT) device, virtual reality (VR) device, and augmented reality (AR) device. As an example, in FIG. 1, the shape of a smartphone is shown as an example of the electronic device 110, but in the embodiments of the present invention, the electronic device 110 is substantially different through the network 170 It may mean one of various physical computer systems capable of communicating with the electronic devices 120, 130, and 140 and/or the servers 150 and 160.

The communication method is not limited, and not only a communication method using a communication network (for example, a mobile communication network, wired Internet, wireless Internet, broadcasting network, satellite network, etc.) that the network 170 may include, but also short-range wireless communication between devices is included. I can. For example, the network 170 includes a personal area network (PAN), a local area network (LAN), a campus area network (CAN), a metropolitan area network (MAN), a wide area network (WAN), and a broadband network (BBN). , Internet, and the like. In addition, the network 170 may include any one or more of a network topology including a bus network, a star network, a ring network, a mesh network, a star-bus network, a tree or a hierarchical network, etc. Not limited.

Each of the servers 150 and 160 is a computer device or a plurality of computers that communicates with a plurality of electronic devices 110, 120, 130, and 140 through a network 170 to provide commands, codes, files, contents, services, etc. It can be implemented with devices. For example, the server 150 may be a system that provides a first service to a plurality of electronic devices 110, 120, 130, 140 accessed through the network 170, and the server 160 is also a network ( It may be a system that provides a second service to a plurality of electronic devices 110, 120, 130, and 140 accessed through 170). As a more specific example, the server 150 is a service (for example, a keyword management service, etc.) targeted by the application through an application as a computer program installed and driven in a plurality of electronic devices 110, 120, 130, 140 May be provided to a plurality of electronic devices 110, 120, 130, and 140 as a first service. As another example, the server 160 may provide a service for distributing the above-described application installation and running file to the plurality of electronic devices 110, 120, 130, and 140 as the second service.

2 is a block diagram illustrating an internal configuration of an electronic device and a server according to an embodiment of the present invention. In FIG. 2, internal configurations of the electronic device 110 and the server 150 are described as examples of the electronic device. In addition, other electronic devices 120, 130, 140 or server 160 may also have the same or similar internal configuration as the electronic device 110 or server 150 described above.

The electronic device 110 and the server 150 may include memories 211 and 221, processors 212 and 222, communication modules 213 and 223, and input/output interfaces 214 and 224. The memories 211 and 221 are non-transitory computer-readable recording media, such as random access memory (RAM), read only memory (ROM), disk drive, solid state drive (SSD), flash memory, and the like. It may include a permanent mass storage device. Here, a non-destructive mass storage device such as a ROM, SSD, flash memory, or disk drive may be included in the electronic device 110 or the server 150 as a separate permanent storage device that is separate from the memories 211 and 221. In addition, the memories 211 and 221 include an operating system and at least one program code (for example, a browser installed and driven in the electronic device 110 or a code for an application installed in the electronic device 110 to provide a specific service). Can be saved. These software components may be loaded from a computer-readable recording medium separate from the memories 211 and 221. Such a separate computer-readable recording medium may include a computer-readable recording medium such as a floppy drive, disk, tape, DVD/CD-ROM drive, and memory card. In another embodiment, software components may be loaded into the memories 211 and 221 through communication modules 213 and 223 other than a computer-readable recording medium. For example, at least one program is a computer program installed by files provided by a file distribution system (for example, the server 160 described above) that distributes the installation files of developers or applications through the network 170 It may be loaded into the memories 211 and 221 based on (for example, the above-described application).

The processors 212 and 222 may be configured to process instructions of a computer program by performing basic arithmetic, logic, and input/output operations. The instructions may be provided to the processors 212 and 222 by the memories 211 and 221 or the communication modules 213 and 223. For example, the processors 212 and 222 may be configured to execute a command received according to a program code stored in a recording device such as the memories 211 and 221.

The communication modules 213 and 223 may provide a function for the electronic device 110 and the server 150 to communicate with each other through the network 170, and the electronic device 110 and/or the server 150 A function for communicating with an electronic device (for example, the electronic device 120) or another server (for example, the server 160) may be provided. As an example, a request generated by the processor 212 of the electronic device 110 according to a program code stored in a recording device such as the memory 211 is sent to the server 150 through the network 170 under the control of the communication module 213. ) Can be delivered. Conversely, control signals, commands, contents, files, etc. provided under the control of the processor 222 of the server 150 are transmitted via the communication module 223 and the network 170 to the communication module 213 of the electronic device 110. ) Through the electronic device 110. For example, control signals, commands, contents, files, etc. of the server 150 received through the communication module 213 may be transmitted to the processor 212 or the memory 211, and the contents or files may be transmitted to an electronic device ( 110) may be stored in a storage medium (permanent storage device described above) that may further include.

The input/output interface 214 may be a means for an interface with the input/output device 215. For example, the input device may include a device such as a keyboard, a mouse, a microphone, and a camera, and the output device may include a device such as a display, a speaker, and a haptic feedback device. As another example, the input/output interface 214 may be a means for interfacing with a device in which input and output functions are integrated into one, such as a touch screen. The input/output device 215 may be configured with the electronic device 110 and one device. In addition, the input/output interface 224 of the server 150 may be a means for an interface with a device (not shown) for input or output that is connected to the server 150 or that the server 150 may include. As a more specific example, when the processor 212 of the electronic device 110 processes a command of a computer program loaded in the memory 211, a service configured by using data provided by the server 150 or the electronic device 120 A screen or content may be displayed on the display through the input/output interface 214.

In addition, in other embodiments, the electronic device 110 and the server 150 may include more components than those of FIG. 2. However, there is no need to clearly show most of the prior art components. For example, the electronic device 110 may be implemented to include at least some of the input/output devices 215 described above, or other components such as a transceiver, a global positioning system (GPS) module, a camera, various sensors, and a database. May include more. As a more specific example, when the electronic device 110 is a smartphone, an acceleration sensor or a gyro sensor, a camera module, various physical buttons, a button using a touch panel, an input/output port, a vibrator for vibration generally included in the smartphone Various components such as may be implemented to be further included in the electronic device 110.

Hereinafter, a specific embodiment of a synonym extraction method and system using a transliteration model will be described.

3 is a block diagram showing an example of components that may be included in a processor of a server according to an embodiment of the present invention, and FIG. 4 is an example of a method performed by a server according to an embodiment of the present invention. Is a flow chart showing.

The server 150 according to the present embodiment serves as a platform that provides a service for managing keywords used on the Internet. In particular, the server 150 can build a multi-graph structure representing the relationship between related keywords for each product included in the product dictionary, and in the case of foreign words included in the product dictionary, the original language notation is automatically extracted and added as synonyms to the multi-graph. can do.

The processor 222 of the server 150 is a component for performing the synonym extraction method according to FIG. 4, and as shown in FIG. 3, a graph generation unit 310, a synonym extraction unit 320, and a synonym addition unit It may include (330). Depending on the embodiment, components of the processor 222 may be selectively included or excluded from the processor 222. Further, according to an embodiment, components of the processor 222 may be separated or merged to express the function of the processor 222.

Components of the processor 222 and the processor 222 may control the server 150 to perform steps (S410 to S430) included in the synonym extraction method of FIG. 4. For example, the processor 222 and the components of the processor 222 may be implemented to execute an instruction according to the code of the operating system included in the memory 221 and the code of at least one program.

Here, the components of the processor 222 may be expressions of different functions of the processor 222 performed by the processor 222 according to an instruction provided by the program code stored in the server 150. . For example, the graph generation unit 310 may be used as a functional expression of the processor 222 controlling the server 150 according to the above-described command so that the server 150 generates a multi-graph representing the relationship between keywords. .

The processor 222 may read necessary commands from the memory 221 loaded with commands related to control of the server 150. In this case, the read command may include a command for controlling the processor 222 to execute steps S410 to S430 to be described later.

In step S410, the graph generation unit 310 may generate a multi-graph indicating a relationship between keywords extracted from product information for each product included in the product dictionary. The product dictionary is a database system for managing keywords, and may be composed of keywords, dictionary types, and synonym sets. The product dictionary may include product information such as a seller, brand, category, attribute, etc., matched to the corresponding product for each product.

The product dictionary described above is product management (automatic product brand matching, product category automatic matching, product quality check, etc.), shopping, search, keyword auto completion, tag dictionary, advertisement, store, NLU (natural language understanding), AI (artificial intelligence). ) Can be applied to various services.

For example, referring to FIG. 5, assuming that the product dictionary 500 is divided into a sales place 510, a brand 520, and a category 530, the sales place-brand, brand- It is possible to extract the relationship data of category, category-seller.

As shown in FIG. 6, the graph generating unit 310 relates to a product A having a seller ID of S1, a brand ID of B1, and a category ID of C1, the seller name'ABC' and the brand name matched to the corresponding product. 'DEF' and category name'GHI' can be converted into keywords, respectively. The graph generation unit 310 may convert product information matching product A with respect to product A into keywords “ABC”, “DEF”, and “GHI”, and then create a multi-graph indicating a relationship between the converted keywords. The graph generator 310 may generate a multi-graph by extracting relationship data between information matching the product from the product dictionary as relationship data between keywords converted from the corresponding information.

Referring to FIG. 7, the graph generating unit 310 includes a connection relationship of a vendor-brand between the keywords'ABC' and'DEF', a connection relationship of a brand-category between the keywords'DEF' and'GHI', and the keyword'ABC'. It is possible to build a multi-graph 700 between'and'GHI' through a connection relationship called a vendor-category.

The multi-graph 700 may be constructed as a tree structure having a single depth as well as a tree structure having a plurality of depths in relation between keywords. For example, when there are multiple series within the same brand, that is, sub-brands, a brand-series connection relationship may be added.

As described above, the graph generation unit 310 may generate a database by accumulating relationship data between keywords, and this may be defined as a structure of the multi-graph 700 representing the relationship between keywords.

In FIG. 4 again, in step S420, the synonym extracting unit 320 may extract the original word notation of the corresponding foreign word from the product name or product-related query as a synonym using the transliteration model for the foreign word included in the product dictionary. .

Some words used in shopping, etc. contain foreign words assimilated from various foreign languages such as English, French, Italian, and German.

In the case of these foreign words, there is a problem that it is difficult to specify the source (ie, country) of the word. For example, in the case of “Chanel”, when pronounced in English, it may be transliterated into “channel” or “channel”, and when pronounced in French, it may be transliterated into “Chanel”, “Chanel”, and “Chanel”.

In order to use the keyword "Chanel", transliteration coverage that can accommodate all pronunciations of various countries is required.

Although there exist learning models that support transliteration through deep learning, the existing learning models are heavy and cannot support pronunciations of various countries at once. English language learning data is required, French language learning data is required, and there is a limit to passing transliteration processing of each language.

In contrast, in the present embodiment, synonyms for foreign words may be extracted using a transliteration model in the unit of MWS (minimal word by sound). Unlike existing learning models, MWS can minimize training data and enable single transliteration processing even for multilingual pronunciation.

MWS is a concept different from syllable, and may mean the smallest unit of pronunciation. Table 1 shows the example word'happy' divided into syllables and MWS.

happy Syllable MWS [ha, ppy] [h, a, ppy], [h, ap, py],
[ha, ppy],
[hap, py]

Each country, language, and person can see the same word and express various pronunciations in the smallest unit. Table 2 shows the MWS of the exemplary word'happy' in Korean notation.

happy [ha, ppy]: Happy or happi → MWS: [ha] = [Ha, ha], [ppy] = [P]
[h, ap, py]: H.AP, H.AP → MWS: [h] = [H], [ap] = [app], [py] = [P, pie]
[hap, py]: happy, ?c pie → MWS: [hap] = [hap,?c], [py] = [pi, pie]

Hangul can be expressed as vector numeric characters when viewed as a dimensional division of the initial, neutral, and final. For example, basic numerical information in Korean is the same as the table in Fig. 8. In order to easily separate words into MWS, a data format of LBG (linked bi-gramed) can be used.

happy ha-ap-pp-py happy c18u1_u1o0_o0c17_c17u20_u20o0

When MWS is [at] = [Atlant], [lant] = [Lant] for [at, lant], which is one of the smallest pronunciation units of the English notation'atlant' corresponding to the Korean notation'Atlant', [at ] = In the case of [Attle], there is no "L" or "R", but you can see that it is transliterated as "Attle". This problem can be solved through a separation method using the LBG data format as shown in the table of FIG. 9.

For example, as shown in FIG. 10, after collecting a learning data set 1001 consisting of a pair of English words and Korean pronunciation in a product dictionary, the MWS data set 1002 is extracted through separate learning in MWS units. I can.

Hereinafter, the process of processing the transliteration range will be described. 11 to 12 illustrate a transliteration processing process for the English word'naver'.

The processor 222 may divide the English word'naver' into candidate groups in all cases that can be separated by MWS (S1).

The processor 222 may extract the MWS existing in the MWS data set 1002 previously constructed from the candidate group separated in step S1 (S2).

The processor 222 may map the MWS extracted in step S2 to each candidate group separated in step S1 (S3).

The processor 222 may connect and merge the MWS mapped in step S3 with each other (S4). In this case, if even one of the candidate groups does not exist in the MWS data set 1002, the corresponding candidate group may be excluded.

The processor 222 may transliterate the merged MWS in step S4 (S5). In other words, the processor 222 may obtain a transliteration for an English word by decoding a Korean vector mapped to the merged MWS. In the result of process S5 of FIG. 12, the number in parentheses may mean the number of corresponding Korean words extracted from the MWS learning data set of FIG. 10, and may be used as a basis for extracting various synonyms if necessary.

An example of the transliteration processing result may be as follows, but is not limited thereto.

-Korean synonyms for gajet: Gadget, Gajet, Gajet, Geojet, Gamjet, Gamjet, Gamjet, Gajet, Kazette, Ganjet

-Korean synonyms of maleficent: Marepicent, Maripicent, Marepicent, Marupicent, Maraficent, Marpicent, Marenpicent, Marepicent, Marepisant, Maripisant, Marinepisent, Maru Pi Sant, Mara Pi Sant, Mar Pi Sant, Maroon Pe Sant

As an example of an English synonym, it is also possible to collect Korean words and English synonyms that did not pass the transliteration model, extract the data set in common with the MWS data set extracted from the existing product dictionary, and then merge and apply it with the final MWS data set. have.

The synonym extractor 320 may extract synonyms of the original language notation for foreign words included in the product dictionary through the transliteration processing described above.

Hereinafter, a specific embodiment will be described using English notation of foreign words written in Korean, that is, English synonyms.

The synonym extracting unit 320 includes a list of product names clicked by users who input a target keyword as a search query for each target keyword by using a foreign word as a target keyword among keywords included in the product dictionary, a tag list related to each click product, and At least one keyword list can be extracted from another query list related to the click product (for example, a query entered by other users who clicked on the same product, another query entered by the same user before clicking the same product, etc.) have. In this specification, a click may mean all types of selection input by a user. 13 exemplarily shows a list of click product names 1311 related to the target keyword'airism (keyword_seq = 102800453)' and a list of click product names 1321 related to'loop top tent (keyword_seq = 438121)'.

The synonym extracting unit 320 may separate the target keyword into a single word when the target keyword is composed of a compound word. For example, in the case of'rooftop tent (keyword_seq = 438121)', it can be divided into'root top tent','rooftop tent','rooftop', and'tent'.

The synonym extracting unit 320 may extract a word of a specific language, that is, an English word, from at least one keyword list of a click product name list, a click product related tag list, and a query list extracted for the target keyword.

14 is an English word list 1411 extracted from a click product name list 1311 related to the target keyword'airism (keyword_seq = 102800453)' and a click product name list 1321 related to'rooftop tent (keyword_seq = 438121)' An English word list 1421 extracted from) is shown as an example.

The synonym extracting unit 320 may extract an English word matching the target keyword from the previously extracted English word list by executing the transliteration model.

15 is a matching result 1511 between a target keyword'airism (keyword_seq = 102800453)' and an English word list 1411, and a matching between a target keyword'loop top tent (keyword_seq = 438121)' and an English word list 1421 The result 1521 is shown as an example.

The synonym extractor 320 may extract an English synonym candidate by replacing an English word matched with the target keyword.

16 shows an English synonym candidate 1611 for the target keyword'airism (keyword_seq = 102800453)' and an English synonym candidate 1621 for'loop top tent (keyword_seq = 438121)'.

The synonym extractor 320 may determine a final English synonym based on whether or not an English synonym candidate appears in a list of click product names related to a target keyword and a tag/query list related to a click product, or frequency of appearance. When the word of the English synonym candidate matches the pattern appearing in the click product name list related to the target keyword and the tag/query list related to the click product, the synonym extractor 320 determines the final English synonym, or there are several English synonym candidates. In this case, a word having a pattern that most often appears in the list of click product names related to the target keyword and the tag/query list related to the click product may be determined as the final English synonym. In other words, the synonym extracting unit 320 may verify an English synonym that is an original language notation of the target keyword by using the click product name list related to the target keyword and the click product related tag/query list.

In FIG. 4 again, in step S430, the synonym adding unit 330 may additionally register the original language representation of the foreign word as a synonym on the multi-graph, and at this time, set a relationship between the existing keyword and the synonym in the multi-graph.

Referring to FIG. 17, between the keywords'ABC' and'DEF', the linking relationship of vendor-brand, the linking relationship of brand-category between the keywords'DEF' and'GHI', and between the keywords'ABC' and'GHI' In the multi-graph 700 created by the connection relationship of vendor-category, if the seller's name'ABC' corresponds to a foreign word, add the original word notation'XXX' of the foreign word'ABC' on the multi-graph 700 as a synonym (1701) Thus, a synonym relationship can be established between the keywords'ABC' and'XXX'.

In addition, among keywords on the multi-graph 700, a relationship between a click product name list related to the target keyword “ABC” and a keyword appearing together with the synonym 1701 in the click product related tag/query list may be additionally set.

Therefore, in the present embodiment, it is possible to automatically register synonyms of the original language for foreign words registered in the product dictionary.

Through the above-described method, keywords in the original language are extracted as synonyms for keywords registered in the product dictionary and registered on the multi-graph, thereby solving synonym processing between keywords and automatically expanding the multi-graph.

By using the transliteration model, the original language notation of foreign words is automatically processed as synonyms, thereby realizing the automation of keyword management tasks, which can significantly reduce related work resources.

In the above, the synonym processing process using the transliteration model is described as a subordinate process performed after the process of generating the multi-graph, but is not limited thereto, and depending on the embodiment, it is performed as a separate and independent process regardless of the multi-graph. It is also possible.

As described above, according to embodiments of the present invention, a multi-graph structure representing the relationship between keywords with respect to the keyword extracted from product information can be constructed, and in the case of foreign words, the original language notation as synonyms can be automatically extracted and added to the multi-graph. According to embodiments of the present invention, the original language notation of the foreign word may be extracted from words appearing in a product name or product-related query using a transliteration model for the foreign word and processed as synonyms.

The apparatus described above may be implemented as a hardware component, a software component, and/or a combination of a hardware component and a software component. For example, the devices and components described in the embodiments are a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), a programmable gate array (PLU). It may be implemented using one or more general purpose computers or special purpose computers, such as a logic unit), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications executed on the operating system. In addition, the processing device may access, store, manipulate, process, and generate data in response to the execution of software. For the convenience of understanding, although it is sometimes described that one processing device is used, one of ordinary skill in the art, the processing device is a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that it may include. For example, the processing device may include a plurality of processors or one processor and one controller. In addition, other processing configurations are possible, such as a parallel processor.

The software may include a computer program, code, instructions, or a combination of one or more of these, configuring the processing unit to behave as desired or processed independently or collectively. You can command the device. Software and/or data may be embodyed in any type of machine, component, physical device, computer storage medium or device to be interpreted by the processing device or to provide instructions or data to the processing device. have. The software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. In this case, the medium may be one that continuously stores a program executable by a computer, or temporarily stores a program for execution or download. In addition, the medium may be a variety of recording means or storage means in a form in which a single or several pieces of hardware are combined, but is not limited to a medium directly connected to a computer system, but may be distributed on a network. Examples of media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical recording media such as CD-ROMs and DVDs, magnetic-optical media such as floptical disks, and And a ROM, RAM, flash memory, and the like, and may be configured to store program instructions. In addition, examples of other media include an app store that distributes applications, a site that supplies or distributes various software, and a recording medium or a storage medium managed by a server.

As described above, although the embodiments have been described by the limited embodiments and drawings, various modifications and variations are possible from the above description by those of ordinary skill in the art. For example, the described techniques are performed in a different order from the described method, and/or components such as a system, structure, device, circuit, etc. described are combined or combined in a form different from the described method, or other components Alternatively, even if substituted or substituted by an equivalent, an appropriate result can be achieved.

Therefore, other implementations, other embodiments, and claims and equivalents fall within the scope of the claims to be described later.

Claims (20)

In the synonym extraction method executed in a computer system,
The computer system includes at least one processor configured to execute computer readable instructions contained in a memory,
The synonym extraction method,
Generating, by the at least one processor, a multi-graph indicating a relationship between keywords in a product dictionary;
Extracting, by the at least one processor, a keyword list related to the target keyword for a target keyword corresponding to a foreign word among keywords included in the product dictionary;
Extracting, by the at least one processor, an original language notation keyword of the target keyword from the keyword list as a synonym of the target keyword using a transliteration model; And
Adding the synonym to the multi-graph by the at least one processor
Including,
The step of additionally registering,
Adding a relationship between the target keyword and the synonym on the multi-graph
Synonym extraction method, characterized in that. The method of claim 1,
The step of extracting the keyword list,
After entering the target keyword as a search query, extracting at least one of a list of product names clicked, a list of other queries related to each click product, and a list of tags related to each click product
Synonym extraction method, characterized in that. The method of claim 1,
The step of extracting as a synonym of the target keyword,
Extracting the synonym using a transliteration model using the minimum unit of pronunciation
Synonym extraction method, characterized in that. The method of claim 3,
The step of extracting as a synonym of the target keyword,
Separating the target keyword into a minimum unit of pronunciation using a data format of LBG (linked bi-gramed)
Synonym extraction method comprising a. The method of claim 1,
The step of extracting as a synonym of the target keyword,
Extracting, as the synonym, a word matching the target keyword among words extracted from the keyword list through the transliteration model
Synonym extraction method comprising a. The method of claim 5,
The step of extracting as a synonym of the target keyword,
Extracting a synonym candidate by substituting the matched word for the target keyword
Synonym extraction method further comprising a. The method of claim 5,
The step of extracting as a synonym of the target keyword,
Extracting a word of a specific language from the keyword list
Synonym extraction method further comprising a. The method of claim 5,
The step of extracting as a synonym of the target keyword,
Separating the target keyword into a single word when it is composed of a compound word
Synonym extraction method further comprising a. The method of claim 5,
The step of extracting as a synonym of the target keyword,
Entering the target keyword as a search query and verifying the synonym using at least one of a clicked product name list, another query list related to each click product, and a tag list related to each click product
Synonym extraction method further comprising a. The method of claim 5,
The step of extracting as a synonym of the target keyword,
Based on the appearance or frequency of occurrence of a word matching the target keyword in at least one of a list of product names clicked after entering the target keyword as a search query, a list of other queries related to each click product, and a tag list related to each click product Determining the synonym with
Synonym extraction method further comprising a. delete The method of claim 1,
The generating step,
Converting product information matched with a corresponding product for each product included in the product dictionary into a keyword; And
Extracting relationship data between information matched with the product from the product dictionary as relationship data between keywords
Synonym extraction method comprising a. The method of claim 1,
The step of additionally registering,
The target keyword among the keywords on the multi-graph is inputted as a search query, and the keyword appears together with the synonym in at least one of a list of product names clicked, a list of other queries related to each click product, and a tag list related to each click product. Adding relationships
Synonym extraction method, characterized in that. A non-transitory computer-readable recording medium on which a program for executing the synonym extraction method of any one of claims 1 to 10, 12, and 13 is recorded on a computer. In a computer system,
At least one processor configured to execute computer readable instructions contained in memory
Including,
The at least one processor,
A graph generator for generating a multi-graph indicating a relationship between keywords in the product dictionary;
After extracting a keyword list related to the target keyword for a target keyword corresponding to a foreign word among the keywords included in the product dictionary, the original language notation keyword of the target keyword from the keyword list is used as a synonym of the target keyword using a transliteration model. A synonym extraction unit to extract; And
Synonym adding unit for additionally registering the synonym in the multi-graph
Including,
The synonym addition unit,
Adding a relationship between the target keyword and the synonym on the multi-graph
Computer system, characterized in that. The method of claim 15,
The synonym extracting unit,
After entering the target keyword as a search query, extracting at least one of a list of product names clicked, a list of other queries related to each click product, and a list of tags related to each click product
Computer system, characterized in that. The method of claim 15,
The synonym extracting unit,
Extracting the synonym using a transliteration model using the minimum unit of pronunciation
Computer system, characterized in that. The method of claim 17,
The synonym extracting unit,
Separating the target keyword into the smallest unit of pronunciation using the data format of LBG (linked bi-gramed)
Computer system, characterized in that. The method of claim 15,
The synonym extracting unit,
Extracting, as the synonym, a word matching the target keyword from among words extracted from the keyword list through the transliteration model
Computer system, characterized in that. The method of claim 19,
The synonym extracting unit,
After entering the target keyword as a search query, verifying the synonym using at least one of a product name list clicked, another query list related to each click product, and a tag list related to each click product
Computer system, characterized in that.

Images