KR102148451B1 - Method, server, and system for providing question and answer data set synchronization service for integration management and inkage of multi-shopping mall - Google Patents

Abstract

Provided is a method for providing a shopping mall question and answer synchronization service for integration management and linkage of a multi-shopping mall. The method comprises the steps of: receiving and storing at least one shopping mall page on which a product is registered and user information required to access the at least one shopping mall page from a seller terminal; collecting and preprocessing product data of products registered on at least one shopping mall page to analyze the product data so as to recommend a previously stored question and answer data set to the seller terminal; accessing the at least one shopping mall page as the user information when the recommended question and answer data set is selected and stored; and registering and synchronizing the question and answer data set stored in the accessed at least one shopping mall page.

Description

Shopping mall Q&A synchronization service provision method, server, and system for integrated management and interworking of multi shopping malls {METHOD, SERVER, AND SYSTEM FOR PROVIDING QUESTION AND ANSWER DATA SET SYNCHRONIZATION SERVICE FOR INTEGRATION MANAGEMENT AND INKAGE OF MULTI-SHOPPING MALL}

The present invention relates to a method for providing a shopping mall Q&A synchronization service for integrated management and interworking of multi shopping malls, and provides a method of synchronizing a recommended Q&A data set based on analysis of product data to be registered in a plurality of multi shopping malls. do.

Various types of online shopping malls are emerging due to the development of information and communication technology and logistics systems. In general, when it comes to online shopping malls, it is easy to think of only the structure of shopping malls in which sellers and buyers are connected one-to-one, but since sellers do not publish products only in one online shopping mall, sellers and multiple shopping malls are connected in a one-to-many form, and the transaction scale is also very big. Online shopping malls are composed of manufacturers that make products, companies that register in the open market to sell products, web service companies that provide online malls, and consumers, and the number of vendors to enter the online shopping malls of each company is increasing. In addition, since thousands of products are registered in the tens of thousands of products, companies that operate online shopping malls are seeking ways to efficiently manage vendors.

At this time, a method of integrated management of a plurality of shopping malls has been researched and developed. In this regard, Korean Patent Publication No. 2019-0073039 (published on June 26, 2019), which is a prior art, registers sales products of a plurality of shopping malls. Or, by receiving correction information, a configuration for registering or modifying sales products in each of at least one shopping mall and controlling the delivery management information of ordered products, access to the integrated management server through wired or wireless communication, and A configuration that performs registration or modification and delivery management information of ordered products collectively through an integrated management server, and a configuration that changes the change history of sales products of multiple shopping malls and management information on order delivery by processing the information provided. It is disclosed.

However, even if the above-described method is used, only product delivery management information can be managed in an integrated manner, and since F&Q or Q&A registered as Q&A for each shopping mall is not registered in an integrated manner, each seller sets one by one for each shopping mall. And input. In addition, when a seller uploads a product to the shopping mall, the above configuration can be used to upload it at the same time, but the background knowledge of the customers who access each shopping mall is different, and various organization members including individuals participate in the purchase process. If you answer each of the questions asked differently, the same question and answer will be repeated at a time difference. Although a systematic and efficient management technique is needed, unfortunately, domestic online shopping malls are not operating that way. In the end, if inefficient operation continues as in the present, the development of online shopping mall operation will be slow.

According to an embodiment of the present invention, when a seller who operates a multi-shopping mall selects and registers a Q&A data set in one shopping mall, interlocking and synchronization are performed so that the pages of other shopping malls are equally reflected, and Q&A When a Q&A that does not exist in the data set is newly registered, it synchronizes the registered Q&A and re-learns the data by inputting the registered Q&A so that the database can evolve and be enriched by itself by the seller and user. , It is possible to provide a method of providing a shopping mall Q&A synchronization service for integrated management and interworking of multi shopping malls. However, the technical problem to be achieved by the present embodiment is not limited to the technical problem as described above, and other technical problems may exist.

As a technical means for achieving the above-described technical problem, an embodiment of the present invention receives user information required to access at least one shopping mall page on which a product is registered and at least one shopping mall page from a seller terminal. Storing, collecting product data of products registered on at least one shopping mall page, performing preprocessing, analyzing product data, and recommending a previously stored Q&A data set to the seller terminal, Q&A recommended by the seller terminal When the data set is selected and stored, accessing the at least one shopping mall page as user information, and registering and synchronizing the Q&A data set stored in the accessed at least one shopping mall page.

In another embodiment of the present invention, at least one shopping mall page on which a product is registered from a seller terminal, a storage unit for receiving and storing user information required to access the at least one shopping mall page, and registering in at least one shopping mall page When the product data of the product is collected and pre-processed, the product data is analyzed, and a recommendation unit that recommends a previously stored Q&A data set to the seller terminal, and the Q&A data set recommended from the seller terminal is selected and stored. And an access unit for accessing one shopping mall page with user information, and a synchronization unit for registering and synchronizing a Q&A data set stored in the accessed at least one shopping mall page.

Another embodiment of the present invention is a seller terminal that transmits at least one shopping mall page in which a product is registered, user information required to access the at least one shopping mall page, and selects and synchronizes a recommended Q&A data set. , And, at least one shopping mall page on which a product is registered from a seller terminal, a storage unit for receiving and storing user information required to access at least one shopping mall page, and product data of products registered on at least one shopping mall page After collecting and performing pre-processing, a recommendation unit that analyzes product data and recommends the previously stored Q&A data set to the seller terminal, and when selecting and storing the Q&A data set recommended from the seller terminal, is stored in at least one shopping mall page. And a Q&A synchronization service providing server including an access unit for accessing user information, and a synchronization unit for registering and synchronizing Q&A data sets stored in the accessed at least one shopping mall page.

According to any one of the above-described problem solving means of the present invention, when a seller who operates a multi-shopping mall selects and registers a Q&A data set in one shopping mall, it is interlocked and synchronized so that it is equally reflected in pages of other shopping malls. If a Q&A that does not exist in the Q&A data set is newly registered, the registered Q&A is synchronized and the data is re-learned by inputting the registered Q&A so that the database is self It can evolve and be enriched.

1 is a diagram illustrating a system for providing a shopping mall Q&A synchronization service for integrated management and interworking of a multi-shopping mall according to an embodiment of the present invention.
FIG. 2 is a block diagram illustrating a Q&A synchronization service providing server included in the system of FIG. 1.
3 is a diagram for explaining an embodiment in which a shopping mall Q&A synchronization service for integrated management and interworking of a multi shopping mall is implemented according to an embodiment of the present invention.
FIG. 4 is a diagram illustrating a process of transmitting and receiving data between components included in the system for providing a shopping mall Q&A synchronization service for integrated management and interworking of the multi shopping mall of FIG. 1 according to an embodiment of the present invention.
5 is an operation flowchart illustrating a method of providing a shopping mall Q&A synchronization service for integrated management and interworking of a multi-shopping mall 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 so that those of ordinary skill in the art can easily implement the present invention. However, the present invention may be implemented in various different forms and is not limited to the embodiments described herein. In the drawings, parts irrelevant to the description are omitted in order to clearly describe the present invention, and similar reference numerals are assigned to similar parts throughout the specification.

Throughout the specification, when a part is said to be "connected" to another part, this includes not only "directly connected" but also "electrically connected" with another element interposed therebetween. . In addition, when a part "includes" a certain component, it means that other components may be further included, and one or more other features, not excluding other components, unless specifically stated to the contrary. It is to be understood that it does not preclude the presence or addition of any number, step, action, component, part, or combination thereof.

The terms "about", "substantially" and the like, as used throughout the specification, are used in or close to the numerical value when manufacturing and material tolerances specific to the stated meaning are presented, and are used in the sense of the present invention. To assist, accurate or absolute figures are used to prevent unfair use of the stated disclosure by unscrupulous infringers. As used throughout the specification of the present invention, the term "step (to)" or "step of" does not mean "step for".

In the present specification, the term "unit" includes a unit realized by hardware, a unit realized by software, and a unit realized using both. Further, one unit may be realized using two or more hardware, or two or more units may be realized using one hardware.

In this specification, some of the operations or functions described as being performed by the terminal, device, or device may be performed instead in a server connected to the terminal, device, or device. Likewise, some of the operations or functions described as being performed by the server may also be performed by a terminal, device, or device connected to the server.

In this specification, some of the operations or functions described as mapping or matching with the terminal means mapping or matching the unique number of the terminal or the identification information of the individual, which is the identification information of the terminal. Can be interpreted as.

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

1 is a diagram illustrating a system for providing a shopping mall Q&A synchronization service for integrated management and interworking of a multi-shopping mall according to an embodiment of the present invention. Referring to FIG. 1, a shopping mall Q&A synchronization service providing system 1 for integrated management and interworking of a multi shopping mall includes at least one seller terminal 100, a Q&A synchronization service providing server 300, and at least one shopping mall. It may include a server 400. However, since the system 1 for providing a shopping mall Q&A synchronization service for integrated management and interworking of a multi shopping mall of FIG. 1 is only an embodiment of the present invention, the present invention is not limitedly interpreted through FIG. 1.

In this case, each component of FIG. 1 is generally connected through a network 200. For example, as shown in FIG. 1, at least one seller terminal 100 may be connected to a Q&A synchronization service providing server 300 and at least one shopping mall server 400 through a network 200. In addition, the Q&A synchronization service providing server 300 may be connected to at least one seller terminal 100 and at least one shopping mall server 400 through the network 200. In addition, at least one shopping mall server 400 may be connected to at least one seller terminal 100 and a Q&A synchronization service providing server 300.

Here, the network refers to a connection structure in which information exchange is possible between respective nodes such as a plurality of terminals and servers, and examples of such networks include RF, 3rd Generation Partnership Project (3GPP) network, and Long Term (LTE). Evolution) network, 5GPP (5th Generation Partnership Project) network, WIMAX (World Interoperability for Microwave Access) network, Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network) , Personal Area Network (PAN), Bluetooth (Bluetooth) network, NFC network, satellite broadcasting network, analog broadcasting network, Digital Multimedia Broadcasting (DMB) network, and the like, but are not limited thereto.

In the following, the term “at least one” is defined as a term including the singular number and the plural number, and even if the term “at least one” does not exist, each component may exist in the singular or plural, and may mean the singular or plural. It will be self-evident. In addition, it will be possible to change according to the embodiment that each component is provided in a singular or plural.

At least one seller terminal 100 is a terminal that registers a Q&A data set using a web page, an app page, a program, or an application related to a shopping mall Q&A synchronization service for integrated management and interworking of a multi shopping mall. I can. To this end, the at least one seller terminal 100 may be a terminal for registering user information accessible to at least one shopping mall page and at least one shopping mall page with the Q&A synchronization service providing server 300. In addition, the at least one seller terminal 100 may be a terminal that selects a Q&A data set recommended from the Q&A synchronization service providing server 300. In addition, the at least one seller terminal 100 may be a terminal that modifies or edits the selected Q&A data set, and may be a terminal that stores for registration after modification or editing. In addition, the at least one seller terminal 100 may be a terminal that collectively registers the stored Q&A data set to at least one shopping mall server 400 through the Q&A synchronization service providing server 300.

Here, the at least one seller terminal 100 may be implemented as a computer capable of accessing a remote server or terminal through a network. Here, the computer may include, for example, a navigation system, a notebook equipped with a web browser, a desktop, a laptop, and the like. At this time, the at least one seller terminal 100 may be implemented as a terminal capable of accessing a remote server or terminal through a network. At least one vendor terminal 100 is, for example, a wireless communication device with guaranteed portability and mobility, and includes navigation, personal communication system (PCS), global system for mobile communications (GSM), personal digital cellular (PDC), PHS(Personal Handyphone System), PDA(Personal Digital Assistant), IMT(International Mobile Telecommunication)-2000, CDMA(Code Division Multiple Access)-2000, W-CDMA(W-Code Division Multiple Access), Wibro(Wireless Broadband Internet) ) All types of handheld-based wireless communication devices such as terminals, smartphones, smartpads, and tablet PCs may be included.

The Q&A synchronization service providing server 300 may be a server that provides a shopping mall Q&A synchronization service web page, an app page, a program, or an application for integrated management and interworking of a multi-shopping mall. In addition, the Q&A synchronization service providing server 300 may be a server that receives information on at least one shopping mall page in which a product is registered from at least one seller terminal 100 and user information for accessing the shopping mall page. . In addition, the Q&A synchronization service providing server 300 may be a server that accesses a registered shopping mall page to collect product data, and is mapped to a previously stored Q&A data set by analyzing text or images included in the product data. It may be a server that lists up a recommended Q&A data set according to a text or image and similarity. And, the Q&A synchronization service providing server 300 transmits the listed Q&A data set to the seller terminal 100, and when selecting any one Q&A data set from the seller terminal 100, at least It may be a server that synchronizes to be registered collectively in one shopping mall page. If, in the case of editing, such as modifying, deleting, or adding, the Q&A data set in the seller terminal 100, the Q&A data set edited to be mapped with the product data is modified, and the previously constructed Q&A data set and In addition, it may be a server that evolves a database by adding text such as product data mapped to a question-and-answer data set and performing relearning.

Here, the Q&A synchronization service providing server 300 may be implemented as a computer that can access a remote server or terminal through a network. Here, the computer may include, for example, a navigation system, a notebook equipped with a web browser, a desktop, a laptop, and the like.

The at least one shopping mall server 400 may be a server of a shopping mall using a shopping mall Q&A synchronization service related web page, an app page, a program, or an application for integrated management and interworking of a multi shopping mall. In this case, the at least one shopping mall server 400 may be a server that manages a shopping mall page of the seller terminal 100 and user information accessible to the shopping mall page. In addition, the at least one shopping mall server 400 may be a server that allows when accessing the account of the seller terminal 100 and accessing the shopping mall page from the Q&A synchronization service providing server 300, It may be a server that inputs and stores data registered in the Q&A synchronization service providing server 300 in the Q&A frame.

Here, the at least one shopping mall server 400 may be implemented as a computer that can access a remote server or terminal through a network. Here, the computer may include, for example, a navigation system, a notebook equipped with a web browser, a desktop, a laptop, and the like.

FIG. 2 is a block diagram illustrating a Q&A synchronization service providing server included in the system of FIG. 1, and FIG. 3 is a shopping mall Q&A synchronization service for integrated management and interworking of a multi-shopping mall according to an embodiment of the present invention. A diagram for explaining an embodiment in which is implemented.

Referring to FIG. 2, the Q&A synchronization service providing server 300 may include a storage unit 310, a recommendation unit 320, an access unit 330, and a synchronization unit 340.

The Q&A synchronization service providing server 300 according to an embodiment of the present invention or another server (not shown) interlocked with each other is a multi-shopping mall with at least one seller terminal 100 and at least one shopping mall server 400 In the case of transmitting a shopping mall Q&A synchronization service application, program, app page, web page, etc. for integrated management and interworking of, at least one seller terminal 100 and at least one shopping mall server 400 You can install or open a shopping mall Q&A synchronization service application, program, app page, web page, etc. for integrated management and interworking. In addition, a service program may be driven in at least one seller terminal 100 and at least one shopping mall server 400 by using a script executed in a web browser. Here, the web browser is a program that enables you to use the web (WWW: world wide web) service, and refers to a program that receives and displays hypertext described in HTML (hyper text mark-up language). For example, Netscape , Explorer, chrome, etc. In addition, the application means an application on the terminal, and includes, for example, an app that is executed on a mobile terminal (smartphone).

Referring to FIG. 2, the storage unit 310 may receive and store at least one shopping mall page on which products are registered and user information required to access the at least one shopping mall page from the seller terminal 100. In this case, the user information required for access may be login information including an ID and a password, but is not limited thereto. In addition, a method of temporarily receiving user information and completely deleting after use may be used only when the seller terminal 100 initially registers the Q&A data set or when a batch synchronization is requested by modifying. Among the compliance related to personal information protection that will be applied in May 2018, the GDPR (General Data Protection Regulation), which has very strict legal responsibilities such as a minimum penalty of KRW 20 billion, is applied, and preparation is required.Therefore, it is not uniquely identifiable information such as ID and password. When used with information, personally identifiable information should be used to a minimum. In addition, since the encryption target file is a violation of the Personal Information Protection Act just by storing or storing plain culture, the storage unit 310 does not store the received information forever, but only while performing batch synchronization with at least one shopping mall server 400. Encrypt and use, and complete deletion after use. At this time, since the seller terminal 100 stores all of its own account information, it is not necessary to input it every time, and if only access permission is obtained through the channel stored in the storage unit 310, the information can be automatically accessed. You can extract it.

In this case, the encryption method or the decryption method may use an ARIA method, a SEED method, or an AES method among block encryption algorithms using a block encryption algorithm, but is not limited thereto. The ARIA algorithm is an encryption algorithm led by the government and designated as a national standard (KS) by the Ministry of Knowledge Economy in accordance with the National Standard Basic Act in 2004. Most of the encryption support for the country uses SEED method. The SEED method is an encryption algorithm for private support. It is an algorithm created with the Korea Internet & Security Agency and domestic encryption experts who understand e-commerce, finance and wireless communication security. AES is the advanced encryption standard (AES) is an encryption method established by the National Institute of Standards and Technology (NIST). Complete deletion refers to permanent deletion in a method that cannot be restored in the case of an electromagnetic file, and a method that cannot be restored refers to the current technology in accordance with Article 10 (destruction method and procedure of personal information) standard guidelines of the Personal Information Protection Act. It refers to a method of taking measures to make it impossible to restore personal information destroyed at an appropriate cost according to social conventions. Complete deletion may be a method required by various laws, such as the destruction of personal information in Article 29 of the Information and Communication Network Act, and disposal of holdings when closing in Article 21 of the Credit Information Act. The complete deletion algorithm may use DoD 5220.22-M-ECE, DoD 5220.22-M, HMG IAS No.5 Higher Overwrite, etc., but is not limited thereto. And, the electromagnetic file containing personal information can be implemented using the algorithm of DoD 5220.22-M method, not the deletion method supported by the simple operating system. 0 is recorded in the entire sector and then 1 is recorded again. This can be done by verifying after filling in the letters of, and it can implement complete deletion of the electromagnetic archived file.

The recommendation unit 320 may collect product data of products registered on at least one shopping mall page, perform preprocessing, analyze product data, and recommend a previously stored Q&A data set to the seller terminal 100. Here, the question-and-answer data set may be a list of F&Q or Q&A in which items frequently asked about products are summarized. The recommendation unit 320 includes text included in product data, text included in product images included in product data, text previously mapped to and stored similar images similar to product images among product data, and product name text among product data. A pre-stored question and answer data set may be extracted based on at least one text extracted from the text. At this time, the recommendation unit 320 according to an embodiment of the present invention recommends a Q&A data set by driving four engines to be mutually linked, and the four engines are a text analysis engine, an OCR text engine, and a similar Includes image search engine and category analysis engine. The four engines will be described below, and those commonly applied will not be described repeatedly.

<Text Analysis Engine>

The recommendation unit 320 performs pre-processing of the text included in the product data, generates a word of the minimum unit after morpheme analysis through part of speech tagging, and uses a TF-IDF (Term Frequency Inverse Document Frequency) algorithm for the generated word. Can be used to analyze. At this time, when performing the pre-processing in the recommendation unit 320, it is possible to correct spaces errors, correct spelling errors, and remove stop words.The spacing error correction model deletes all spaces from the text included in the product data, and then machine learning. You can use the sequential labeling method to determine the space to write using the model. The sequential labeling method determines whether there is a space between the syllables and the syllables from the beginning to the end of the sentence and attaches a cover (0 or 1 in the figure).To determine the cover, n syllables before and after the current position are used as features. , As a machine learning method, CRFs (Conditional Random Fields) or Structural SVM (Support Vector Machine), which are known to reflect transition information well in both directions (from the beginning to the current position and from the last to the current position) can be used. However, it is not limited thereto. The spelling error correction model can mainly use dictionary or statistical information.If the spelling error correction using a dictionary is not found in the corpus or does not exist in a general dictionary, it can be assumed that it is a spelling error, and it is assumed as a spelling error. The resulting character string may be replaced with a dictionary word that is close to the corresponding word by using an edit-distance or metaphone algorithm. Since this dictionary-based method has to construct all candidate misspelling words in advance, probability models based on statistical information can also be used. For example, a noise channel model, which is a method of correcting spelling errors, can be used, and the noise channel model is a type of generation model that finds words that maximize the probability of generating input noise (spelling errors). It is calculated based on the edit distance in syllables (or alphabets) between the and error words. For example, it may be calculated using the probability that the error word'acress' is generated from the correct word'actress' (= the probability that'c' is deleted from'ct' or'ct' is replaced with'c'). .

At this time, Korean is a typical agglutinative language or affixing language in which research and endings are developed. Unlike English, morphemes rather than words play an important role in sentiment analysis or sentence structure analysis. A morpheme is defined as the smallest unit of speech with a meaning, and a word consists of one or more morphemes. For example, if'pretty' is a word, the morphemes for this are'pretty (= adjective)' and'go (= concatenated ending)'. Reflecting the importance of such morphemes, in text analysis targeting Korean, it is common to use morphemes as basic units rather than words, and morphemes vectors (not word vectors mainly used in English texts as input in text analysis) morpheme vector) can be used, but it does not exclude word vectors. A morpheme vector can be said to be a vector expression for a morpheme, and can be derived by applying a basic word vector derivation mechanism to a sentence expressed in morpheme units. In this case, the word vector is a word converted into a vector of tens to hundreds of dimensions of real number elements. The word vectors converted in this way contain more than just numerical values. In fact, words with similar semantic and syntactic properties are expressed as vectors with close Euclidean distance or cosine similarity distance in vector space. For example, the Word2Vec model internally learns the vector representation of words using a neural network structure called CBOW (Continuous Bag-Of-Words) or Skip-Gram. CBOW predicts the target word from context words, for example, in the sentence “This lipstick is beautiful in (_) and has a good sustainability.”, it corresponds to (_) from context words that precede and follow (_). It is a method of predicting target words such as'color'. The Skip-Gram structure inversely predicts context words from the target word.

For example, when there is a sentence "This lipstick has a beautiful color and has good lasting power", if you use words separated by word as input to CBOW, a word vector such as'Lipstick is' or'Color degree' is used. You will get However, in this way,'Lipstick을','Lipstick이','Color','Color', and'Color is' are all recognized as different words, which inefficiently increases the number of words to consider. By the way, if this sentence is morphologically analyzed, it will be slightly different depending on the selected morpheme analyzer or the degree of segmentation, but, “Lee (= spectral sentence) lipstick (= general noun) silver (= general noun) color (= general noun) also (= auxiliary) pretty =adjective) high (=connection ending) persistence (= general noun) degree (= auxiliary) good (= adjective) c (= ending ending) .(=period)”. From this, a sentence expression in morpheme units such as “this lipstick has a beautiful color and good lasting power” can be obtained, and by inputting this into CBOW, a morpheme vector in morpheme units can be obtained. Here, the morpheme vector may be used as an input of a CNN (Convolutional Neural Network) deep learning model.

CNN is a typical feedforward neural network, also called a multi-layer perceptron, in which units of adjacent layers are fully-connected. In addition to this fully-connected layer, CNN has a convolution layer and a pooling layer in which only specific units of an adjacent layer are connected. In the convolution layer, a convolution operation that calculates the sum after multiplying the corresponding elements element-wise while sequentially moving a filter in the form of a weight matrix smaller in size than the image in the weight region of the image expressed in the form of a weight matrix. This happens. In addition, the pooling layer performs an operation of sequentially extracting representative values such as a maximum value or an average value for each specific size area with respect to values obtained by convolution. The convolutional layer and the pooling layer may be paired with each other and repeated several times, or a structure in which the pooling layer comes after only the convolutional layer is repeated several times is possible. Useful features are hierarchically extracted from the input image through such an operation, and the extracted features are used to classify input data into target classes through one or more fully-connected layers. When CNN is applied to a weight matrix representing a sentence as a combination of word vectors, local n-gram features can be extracted according to the window size of a filter. And based on these qualities, it is possible to extract higher-level qualities again. The convolution operation applies a filter to a word window to extract a new feature, and the filter is applied to all possible windows of the sentence to create a feature map. The pooling operation obtains a global maximum value for this feature map or a local representative value for each specific size. If several filters with different window sizes are applied, features of various features can be extracted, and the features extracted in this way are connected to the pre-combination layer and used for class classification.

In addition, the recommendation unit 320 may perform a process of dividing a document into words, word embedding, and TF-IDF weighting steps after the spacing error, spelling error, stop word removal, etc. are performed. First, CNN requires input data in a two-dimensional form. Therefore, to satisfy this, the recommendation unit 320 may perform a word embedding of dividing the document into words and numbering each word. At this time, as the embedding method, a method of sequentially numbering the words according to the order in which they appear, or a word-level model that uses words as a basic unit or a character-level model that processes individual letters as basic units. May be. In addition, the recommendation unit 320 obtains a TF-IDF weight value to indicate how important the embedded words are in the corresponding document, attaches the weight value to the array of the embedded words, and adds the input value of one convolutional neural network. You can create and use it. Here, TF-IDF is a weight used in information retrieval and text mining, and is a statistical value indicating how important a word is in a specific document when there is a document group consisting of several documents. Word frequency (Term Frequency) is a value indicating how often a specific word appears in a document, and Inverse Document Frequency refers to a word that appears frequently in a document that is not many in other documents. The value of TF-IDF is used as the product of word frequency and reverse document frequency. For example, the word leather bag does not appear well in the one-piece Q&A data set, so the IDF value increases and can be a keyword for the document at the same time. Conversely, in the fashion category, the IDF value decreases because leather bags appear on many shopping mall sites. TF can be calculated using the total frequency of words in the document basically, but since the value can be continuously increased when the number of words increases, TF values are defined as 0 and 1 only by the occurrence of words using Boolean Frequency. can do. Alternatively, the size of the document can be resolved and the actual frequency can be reflected by taking a log of the TF value using a logarithmically scaled frequency. Finally, by using the Augmented Frequency, the maximum scale may not exceed 1 in a manner that indicates the relative frequency of words according to the document length. As a result, TF-IDF increases as the frequency of the word is high in a specific document and the number of documents containing the word among all documents decreases. This allows you to filter out words that appear frequently in all documents.

The recommender 320 may perform text analysis modeled using a classifier after forming the analyzed text into a matrix. In this case, the matrix may be a vector space, and a vector space model may be used. The vector space model is called a word space model in the field of text mining, and is an algebraic model representing a given text document as a vector of words. The vector space model can be used for information filtering, information search, indexing, and similarity ranking. If a word is included in a document, the word has a non-zero vector value. There may be various methods for calculating this value called the word weight, but in an embodiment of the present invention, the TF-IDF method described above is used. In this case, the classifier may be a document classification model based on a Convolution Neural Network (CNN), but is not limited thereto. CNN, as described above, is a kind of artificial intelligence deep neural network, has a structure suitable for learning two-dimensional data, can be trained through a back propagation algorithm, and is included in a pre-built database. It can be used to search the question-and-answer data set using the analyzed text and make recommendations to sellers based on similarity.

The recommendation unit 320 may extract a previously stored Q&A data set based on the analyzed text based on similarity and recommend it to the seller terminal 100. For example, if the product detail text for "Women's One Piece" was analyzed, and as a result of the analysis, three Q&A sets previously stored in "Women's One Piece" were extracted, and that the similarity was 90%, 80%, and 70%, respectively, Q&A sets can be sorted and recommended in the order of high similarity.

<OCR text engine>

After collecting product data describing the product, the recommendation unit 320 applies the product image to an OCR (Optical Character Recognition) text-based image analysis model when there is no text in the product data and only the product image exists. Can be converted to text. In most shopping mall pages, at least a few to dozens of product images are uploaded on one page, and only text is included in each product image, and text is rarely described separately. For example, 0061 (product number)-jasmine dress (brand name)-FREE (size)-beige (color) is simply uploaded as an image along with the product image as a text in the form of an image on the thumbnail, and even elasticity, thickness, and see-through , Lining, washing method, and material are often included in the imaged product image. Therefore, when the product image + product description text are all imaged, the recommendation unit 320 first performs an operation for reading the text from the image file.

OCR stands for optical character recognition and is a method of recognizing optically processed characters. Unlike online recognition, which recognizes characters expressed on a computer, optical recognition can be performed offline after recording or printing is complete. Hand-printed or printed characters can all be recognized, but accuracy depends on the quality of the entered document. It depends directly. By digitizing an analog document using an optical scanner, the text area can be extracted through the process of dividing the text area into each symbol, and the pre-processing process can proceed to remove the noise of the symbol to facilitate text extraction, After the text is extracted, each test is compared and the words and numbers of the original text are reconstructed. At this time, the OCR text-based image analysis model largely performs a process of text detection and text recognition. The former can use a deep learning CTPN (Connectionist Text Proposal Network) model that finds the location of text from an image. Here, the CTPN can be largely composed of a CNN model for distinguishing between text and non-text in an image, and a recurrent neural network (RNN) model that specifies the text estimation region, and the region estimated as text is divided into a plurality of pixels as a vertical segmentation plane. Through the split CNN model, it is possible to determine whether each split plane is close to the text, and check whether the split plane on both sides is connected text through the RNN. In an embodiment of the present invention, a structure of VGG16, Bidirectional Long Short-Term Memory (BLSTM), and FC may be used.

VCC16 is a program developed by Oxford University. The input is an RGB image, and the structure includes 8 to 16 convolutional layers and 3 fully-connected layers. It is not limited. At this time, VCC16 is used to extract features of an image, and BLSTM (Bidirectional Long Short-Term Memory) is an algorithm used to classify and predict time series data as a type of cyclic neural network (RNN). At this time, the BLSTM can perform the task by connecting two LSTM outputs that process a forward state sequence from left to right and a reverse state sequence from right to left. One-way LSTM is a context from a past time instance. While only information is considered, BLSTM can learn using both past and future contextual information transmitted in the forward and reverse directions. Here, LSTM is a kind of recurrent neural network with a structure to solve the gradient loss problem that occurs during learning in the RNN method, and maintains the potential long-term memory dependence. Thus, to classify, process, and predict time series signals, LSTMs can learn from history. The LSTM remembers the previous state and special memory cells that have their own connection to the cyclic concealment layer to maintain their state over time, and three gates are used to control the flow of information to the inputs and outputs of each layer. It can have structures (input gate, forget gate and output gate), and an LSTM with such a cyclic output layer can capture the contextual information of the input text together. In addition, a result of learning the temporal relationship of time series data in the BLSTM can be obtained through the above-described fully connected layer.

The latter, text recognition, may use a Convolution Recurrent Neural Network (CRNN) model that extracts text images as text. At this time, the CRNN is a deep learning structure that combines CNN and RNN, and can be used when extracting data where both spatial and temporal information are important, such as video. For example, after extracting feature information of a product image uploaded to a shopping mall page through a CNN, the extracted value may be used as an input of an RNN to predict and extract a text sequence of data. The CRNN according to an embodiment of the present invention may be composed of a convolution layer, a recurrent layer, and a transcription layer, and after extracting a feature sequence from an image input from the convolution layer, , In the circular layer, a value is predicted for each frame in the feature sequence, and the sequence of predicted values is rewritten as a correct answer label through the transfer layer to finally extract the text. And, unlike CTPN (Connectionist Text Proposal Network), since CRNN learns based on the image of the text area rather than the entire image, when extracting the text in the image, it is necessary to recognize the entity name by classifying it by field existing in the product image. However, BLSTM, CNNs, and CRF (Conditional Random Field) models can be further used, and learning can be performed based on characters and corresponding entity names.

Named Entity refers to a word that can identify an entity from a previously predefined set of other entities with similar properties, or a phrase that can no longer be resolved within a sentence. The entity name recognition process refers to the process of identifying entity names expressed in documents, and is a process that must precede the process of extracting information through the relationship between entity names. In the product page, the product number, product name, size, and color are generally described with predefined attributes at a certain location. Recognition of the entity name is classified as a kind of sequence labeling problem for information in a sentence. By learning which attribute has an entity name in a location, attribute information of each product can be obtained from a clothing or fashion page where dozens of products are registered on one page. If the CRF-based entity name recognition method of a linear linear chain is used, the performance of entity name recognition can be improved by maximizing the Log-Likelihood value, which is a conditional probability for string information. At this time, the deep learning technique that shows good performance in the continuous labeling problem, LSTM (Long-Short Term Unit), is used in both directions, and a BLSTM-CRF model in the form of attaching a conditional random field is constructed to recognize the entity name. You can also do it. In entity name recognition, entity name dictionary information including entity name information for each entity type is used as a major feature. Based on the BLSTM model mentioned in the text-based engine, the mapping information for the entity name dictionary can be configured in a binary vector format and merged into the input vector.

<Similar image search engine>

After collecting product data describing the product, the recommendation unit 320 uses at least one model learned by deep learning among the previously stored databases when there is no text in the product data and only the product image. A product using an algorithm or a clustering algorithm that extracts at least one feature of an image, expresses the extracted at least one feature in a vector space, and calculates the similarity distance. A similar image similar to the image may be searched, and the similar image unit 360 may extract text previously mapped to the searched similar image.

For deep learning, ImageNet may be used, an algorithm for calculating the similarity distance may be a K-Nearest Neighbor, and clustering may be a K-means clustering. Here, when extracting image features, transfer learning can be used, and DensNet, Resnet50, VGG16, Inceptionv3, Xception, and the like can be used. At this time, transfer learning is a method that uses deep learning as a feature extractor and learns another model with only the extracted features. When creating a new model using an existing model, it is a method of speeding up learning and improving prediction. With this completed model (Pre-Training Model), transfer learning is a method of learning by giving fine adjustments to the target learning, and this re-learning process of a neural network is called fine-tuning, and the actual CNN is constructed. In most cases, training is not performed from the beginning (random initialization), and large datasets such as ImageNet can be used. In summary, transfer learning is a learning method that overcomes the limitations of insufficient training data and efficiently extracts feature maps by initializing with values learned from a large amount of image data instead of random initialization of the convolutional layer.

The K-nearest neighbor algorithm means that the nearest neighbors are used. This method is a very intuitive way to classify unlabeled samples according to the similarity between samples in the training data set. In other words, if a sample without a label is given, it finds the sample with the closest label in the training data set and assigns it to the class with the most frequency in the subset. The K-nearest neighbor algorithm is simple because it only needs a constant, a sample of a labeled training data set, and a distance measure. At this time, the subset of data in the same class represented by each data in the classification problem can be calculated as a Euclidean distance, the middle of the distance having the largest distance value in the same class and the smallest distance value among other classes. You can determine the distance by value.

<Category Analysis Engine>

Category analysis is used when there is a case in a shopping mall where a seller does not provide text or an image in the detailed content of the product and there is no image of the product. In this case, the point is that the product name (product name) used in the shopping mall itself can extract the category of the product, and if the category can be extracted, the Q&A data set mapped to the corresponding category can be extracted. To this end, the recommendation unit 320 collects product data describing the product, and then, when text and images do not exist in the product data, and only a product name indicating a product exists, a stop word included in the product name (Stop- word) can be performed. Further, the recommendation unit 320 performs one-hot encoding on the text included in the product name on which the preprocessing has been performed, based on the syllable, and converts the text included in the product name changed into a vector as learning data of machine learning. It is possible to predict a category through a classification model based on the text input as and input as training data, and extract the stored text after being mapped to the predicted category. In this case, the classification model may be a deep neural network.

In this case, the deep neural network (DNN) is an artificial neural network (ANN) composed of a plurality of hidden layers between an input layer and an output layer. DNN can model a complex non-linear relationship like a general artificial neural network, and a neural network operation method may include a feed-forward network (FFN), backpropagation, and an RNN. For example, data includes numeric data and text data or categorical data. Since machine learning or deep learning algorithms can only understand numerical data, they transform the data into a form that machines can understand, while categorical data transforms it into a one-hot encoding form. One-hot encoding means that only one data is changed to 1 and the rest are filled with 0. For example, assuming that apples, pears, and persimmons are included in the fruit category, each fruit is apple and pear. Create a column with, persimmon, and mark only the fruits as 1, and mark the remaining fruits as 0. For example, assuming that entity name recognition is performed from a product name, there are many words that are difficult to say that it contains knowledge or meaning among the extracted entity names, and the dimension varies in the vectorization process due to the difference in the number of entity names extracted for each category. There is. Therefore, if you set a constraint that matches the dimension by limiting to the high-order number of entity names with high frequency of occurrence for each category, as a result, entity names extracted from a number of categories can be vectorized through one-hot encoding, and input for learning. The data is complete.

At this time, when test data is input to the trained DNN model, the behavior of neurons can be divided into two categories based on the result value output from the neurons of the neural network. The first is when the result value falls within the expected range and is said to be located in the main case area. The second is when the result value falls within the unexpected range, and it is said to be located in the corner case area. Here, the neuron of the DNN has a structure in which a result value is derived through an activation function using input data, weight, and bias. In the case of a trained DNN model, the weight and bias are fixed, and only the input data can affect the neuron's result value, and it is said that the neuron is activated when the neuron reacts by the input data, that is, when the result value changes. When activated on data, categories can be predicted. In other words, if it belongs to the section where the result value of the neuron is divided by the input data, it can be considered that the section is activated, so that the category can be predicted. Here, when neurons are not activated, that is, corner cases are excluded.

The four engines described above are organically connected while performing a complementary role, and regardless of what kind of product data the seller uploads, even if he just writes the product name, it can recommend appropriate Q&A, and the seller does not answer one by one. There is no need, and consumers do not have to try to find the answers they want from product inquiries and answers listed in a list type.

The access unit 330 may access at least one shopping mall page as user information when selecting and storing the recommended Q&A data set from the seller terminal 100. In this case, the access unit 330 may access a plurality of shopping mall servers 400 with one account information by introducing a single sign-on (SSO) integrated authentication system. SSO is a concept in which you have the authority to access all servers through one login. In particular, it can be implemented by using an API mechanism between an application program and an SSO server in a way that a certificate is received from a CA centering on a directory system, a Lightweight Directory Service Protocol (LDAP) server. In addition, the access unit 330 may use LDAP for fast authentication service. LDAP was developed based on ITU-T's X.500 as a communication protocol for storing and searching all types of directory-type data in a standardized way. LDAP is a communication protocol developed based on X.500, and X.500 is a directory that manages objects such as countries, organizations, people, machines, etc. and provides information so that it can be used anywhere in the world wherever Internet access is available. Service standard. Accordingly, the access unit 330 may use a method of storing data required for LDAP synchronization in a temporary table using an integrated user table, a site subscription table, and columns required for authentication and site access rights using an Oracle trigger. In the temporary table, data can be stored when a user or an administrator changes the integrated member information (membership registration, withdrawal, modification, etc.). You can do it in In order to synchronize user information to LDAP in real time, the service platform of the authentication server is described. When a seller signs up for a member, member information is stored and an oracle trigger operates to provide basic member information required for authentication. Save it. In web programming, the LDAP synchronization processing API is called after the user subscription processing is finished. In the LDAP synchronization processing API, the related data is retrieved from the table by the user ID, reflected in the LDAP, and then the data in the temporary table is deleted. However, it is not limited thereto.

The synchronization unit 340 may register and synchronize the Q&A data set stored in the accessed at least one shopping mall page. At this time, data synchronization is a specification for data synchronization between distributed data existing in the network, and a rule standardizing requirements required for message transmission using various network protocols may be used. If the seller terminal 100, which is a client, and at least one shopping mall server 400 are connected based on this rule, a logical connection between data generated by applications between the server and the client becomes possible. The synchronization agent may include a synchronization adapter, a synchronization engine, a synchronization process, and a synchronization client including a network process, a client agent, and a client database adapter, but is not limited thereto. The client agent may be a package for requesting synchronization from a client or transferring changed data to an embedded client, and the client database adapter may be a package for accessing and updating a client database. The network processing may be a package of functions necessary to support each protocol for receiving a synchronization document according to a network environment.

In addition, in an embodiment of the present invention, deep learning may be used when relearning a question and answer data set. Deep learning tends to train a better model as the size of the training model increases and the amount of training data increases. However, training a large model using a large data set requires a lot of computation. , Distributed deep learning technology that learns one global model through collaboration between multiple GPUs can be used. Distributed deep learning techniques are largely divided into model parallelization-based distributed deep learning and data parallelization-based distributed deep learning.In model parallelism-based distributed deep learning, one model is divided and stored in multiple nodes, and each part of the model is parallelized. In the distributed deep learning method based on data parallelization, each node has the same model, but a data set is divided and stored in multiple nodes, and each allocated data is learned in parallel.

Distributed deep learning based on data parallelization is a method of distributed deep learning in which a vast amount of training data is divided and stored in several nodes, and each node trains a model of the same type based on the data it is responsible for. In this method, since each node independently learns different data, a process of summing and synchronizing the learning results of each node is required at regular intervals. For this learning method, distributed deep learning can be performed by placing a parameter server that sums and manages the learning results of each node. Here, a worker is a node in charge of learning, and plays a role of learning the data stored in each. Each worker transmits the result of learning its own data to the parameter server, the parameter server updates the global parameter by summing the learning results received from the workers, and retransmits the updated parameter to each worker. At this time, various data parallelization-based distributed deep learning methods can be used according to the communication method between the parameter server and the worker.For example, in Bulk Synchronous Parallel (BSP), each worker learns batch data and calculates the learning result. When the parameter server is sent to the parameter server, and the parameter server updates the global parameter by summing the learning results of all workers, each worker receives the updated parameter from the parameter server and repeats the first process for the next batch data. That is, all workers learn based on the same parameters.

Asynchronous Parallel (ASP) For example, when each worker learns batch data and transmits the training result to the parameter server, the parameter server updates the global parameter with the learning result received from each worker and transmits the updated parameter to the worker. Then, the worker receiving the new parameter repeats the first process for the next batch of data. That is, each worker node learns without waiting for other worker nodes. Unlike BSP (Bulk Synchronous Parallel) and ASP (Asynchronous Parallel), SSP (Stalness Synchronous Parallel) is a method that allows learning using local parameters. For example, each worker performs learning based on local parameters. When batch processing data is learned and the training result is transmitted to the parameter server, the parameter server updates the global parameter with the training result received from the worker, and the worker is the parameter server only when its local parameter differs from the global parameter by more than a certain version. It receives the parameters from and updates the local parameters, and each worker repeats the first process. Of course, in addition to the above-described methods, various methods of distributing learning and synchronizing the results may be used, and are not limited to those described above.

Hereinafter, an operation process according to the configuration of the above-described Q&A synchronization service providing server of FIG. 2 will be described in detail with reference to FIG. 3 as an example. However, it will be apparent that the embodiment is only any one of various embodiments of the present invention, and is not limited thereto.

Referring to FIG. 3, (a) the Q&A synchronization service providing server 300 includes information such as URLs of pages of a plurality of shopping malls or multi shopping malls from at least one seller terminal 100, and access information accessible to each page, For example, account information can be registered, and (b) accordingly, the Q&A synchronization service providing server 300 collects at least one kind of product data posted on each page and analyzes the product data. Extracts the Q&A data set that is basically provided for the product data. And, (c) the Q&A synchronization service providing server 300 accesses at least one shopping mall server 400 using user information, and selects a product Q&A or F&Q frame of each shopping mall page from the seller terminal 100 The Q&A data set that has been selected or edited is automatically recorded, and the same contents are collectively synchronized so that the same contents are registered in the plurality of shopping mall servers 400. (d) In addition, in the case of changing or editing the Q&A data set in at least one seller terminal 100, it may be retrained to fit each product data based on this, and each learned data is not limited to clothing. It can be constructed in various ways.

For the information that is not described about the method of providing a shopping mall Q&A synchronization service for integrated management and interworking of the multi shopping malls of FIGS. 2 and 3 as described above, a shopping mall Q&A synchronization service for integrated management and interworking of the multi shopping malls Since the content of the providing method is the same as the description or can be easily inferred from the description, the following description will be omitted.

FIG. 4 is a diagram illustrating a process of transmitting and receiving data between components included in the system for providing a shopping mall Q&A synchronization service for integrated management and interworking of the multi shopping mall of FIG. 1 according to an embodiment of the present invention. Hereinafter, an example of a process in which data is transmitted/received between each component will be described through FIG. 4, but the present application is not limitedly interpreted as such an embodiment, and according to various embodiments described above, It is apparent to those skilled in the art that the process of transmitting and receiving data may be changed.

4, the Q&A synchronization service providing server 300 uploads a product from the seller terminal 100 to at least one shopping mall page 400 (S4100), and the Q&A synchronization service providing server 300 When registering information on a shopping mall page and information accessible to the page (S4200), product data is collected by accessing each page (S4300 and S4400). Then, the Q&A synchronization service providing server 300 analyzes the product data (S4500), checks whether text is included (S4600), and if there is no text, checks whether a product image exists (S4610), and When text exists in the image (S4630), text is extracted from the image using the OCR image text analysis model (S4670). In addition, when the product image exists but the text does not exist in the form of an image, the Q&A synchronization service providing server 300 searches for the most similar image among the previously established images using a similar image search model (S4650). In addition, the Q&A synchronization service providing server 300 performs category analysis on the product name when there is no product text and no product image, and thus no information to recommend Q&A exists (S4700).

On the other hand, if it is defined as containing text in step S4600, pre-processing is performed (S4730), parts of speech tagging and morpheme analysis are performed (S4750), and then text, images, and text analysis algorithms are used (S4770). This results in a step of extracting a Q&A dataset mapped to any one of the categories (S4800). In this way, if text is extracted from the imaged text or only the product name exists, the question-and-answer data set mapped to the data set exists (S4800) or the question-and-answer data set is searched based on similarity and extracted (S4860). ) It is recommended to the seller terminal 100 (S4810).

When the Q&A synchronization service providing server 300 selects any one of the Q&A data sets recommended from the seller terminal 100 (S4830), the selected Q&A data set is registered with at least one shopping mall server 400. Each page is accessed and registered (S4850). Then, the Q&A synchronization service providing server 300 checks whether each page is properly registered (S4900), and transmits the result to the seller terminal 100 (S4920).

The order between the above-described steps S4100 to S4920 is only an example and is not limited thereto. That is, the order of the above-described steps (S4100 to S4920) may be mutually changed, and some of the steps may be executed or deleted at the same time.

For the information that is not described about the method of providing the shopping mall Q&A synchronization service for integrated management and interworking of the multi shopping malls of FIG. 4, a shopping mall Q&A synchronization service for integrated management and interworking of the multi shopping malls through FIGS. 1 to 3 above. Since the content of the providing method is the same as the description or can be easily inferred from the description, the following description will be omitted.

5 is an operation flowchart illustrating a method of providing a shopping mall Q&A synchronization service for integrated management and interworking of a multi-shopping mall according to an embodiment of the present invention. Referring to FIG. 5, the server for providing a Q&A synchronization service receives and stores at least one shopping mall page on which a product is registered and user information required to access the at least one shopping mall page from a seller terminal (S5100).

The Q&A synchronization service providing server collects product data of products registered on at least one shopping mall page, performs preprocessing, analyzes the product data, and recommends a previously stored Q&A data set to the seller terminal (S5200), and the seller When the terminal selects and stores the recommended Q&A data set, at least one shopping mall page is accessed as user information (S5300).

Finally, the Q&A synchronization service providing server registers and synchronizes Q&A data sets stored in at least one accessed shopping mall page (S5400).

For the information that is not described about the method of providing a shopping mall Q&A synchronization service for integrated management and interworking of the multi shopping mall of FIG. 5, a shopping mall Q&A synchronization service for integrated management and interworking of a multi shopping mall through FIGS. 1 to 4 above. Since the content of the providing method is the same as the description or can be easily inferred from the description, the following description will be omitted.

The method for providing a shopping mall question-and-answer synchronization service for integrated management and interworking of a multi-shopping mall according to an embodiment described with reference to FIG. 5 is a record including an application executed by a computer or a computer-executable instruction such as a program module. It can also be implemented in the form of a medium. Computer-readable media can be any available media that can be accessed by a computer, and includes both volatile and nonvolatile media, removable and non-removable media. Further, the computer-readable medium may include all computer storage media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data.

In the above-described method for providing a shopping mall Q&A synchronization service for integrated management and interworking of a multi-shopping mall according to an embodiment of the present invention, an application basically installed in a terminal (this is a program included in a platform or operating system basically installed in the terminal). May be included), and may be executed by an application (that is, a program) directly installed on the master terminal through an application providing server such as an application store server, an application, or a web server related to the service. In this sense, the method for providing a shopping mall Q&A synchronization service for integrated management and interworking of a multi-shopping mall according to an embodiment of the present invention described above is implemented as an application (i.e., a program) installed in a terminal or directly installed by a user. It can be recorded on a computer-readable recording medium such as a terminal.

The above description of the present invention is for illustrative purposes only, and those of ordinary skill in the art to which the present invention pertains will be able to understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are illustrative and non-limiting in all respects. For example, each component described as a single type may be implemented in a distributed manner, and similarly, components described as being distributed may also be implemented in a combined form.

The scope of the present invention is indicated by the claims to be described later rather than the detailed description, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention. do.

Claims (7)

A seller terminal for transmitting at least one shopping mall page on which a product is registered, user information required to access the at least one shopping mall page, and selecting and synchronizing the recommended Q&A data set; And a storage unit for receiving and storing at least one shopping mall page on which a product is registered from the seller terminal, user information required to access the at least one shopping mall page, and a product of the product registered on the at least one shopping mall page. When collecting data and performing pre-processing, a recommendation unit that analyzes the product data and recommends a previously stored Q&A data set to the seller terminal, and when the seller terminal selects and stores the recommended Q&A data set, the A Q&A synchronization service providing server including an access unit for accessing at least one shopping mall page with the user information, and a synchronization unit for registering and synchronizing the stored Q&A data set on the accessed at least one shopping mall page; In the method for providing a Q&A synchronization service,
Receiving and storing at least one shopping mall page on which a product is registered and user information required to access the at least one shopping mall page from a seller terminal;
Collecting product data of products registered in the at least one shopping mall page, performing pre-processing, analyzing the product data, and recommending a previously stored Q&A data set to the seller terminal;
When the seller terminal selects and stores the recommended Q&A data set, accessing the at least one shopping mall page with the user information; And,
And registering and synchronizing the stored Q&A data set in the accessed at least one shopping mall page,
After collecting product data of products registered on the at least one shopping mall page, performing pre-processing, analyzing the product data and recommending a previously stored Q&A data set to the seller terminal,
Text included in the product data, text included in the product image included in the product data, a text pre-mapped to a previously stored similar image similar to the product image among the product data and stored, and extracted from the product name text among the product data Extracting the previously stored Q&A data set based on at least one text among texts; To provide a shopping mall Q&A synchronization service for integrated management and interworking of a multi-shopping mall.
delete The method of claim 1,
Extracting the previously stored Q&A data set based on the text included in the product data,
Pre-processing the text included in the product data, generating a word of the smallest unit after morpheme analysis through part-of-speech tagging, and analyzing the generated word;
Performing text analysis of modeling the analyzed text into a matrix and then modeling it with a classifier; And
Extracting a previously stored Q&A data set based on the analyzed text based on similarity and recommending it to the seller terminal;
A method for providing a shopping mall question-and-answer synchronization service for integrated management and interworking of a multi-shopping mall.
The method of claim 1,
Extracting the previously stored Q&A data set based on the text included in the product image included in the product data,
Converting the product image into text by applying an OCR (Optical Character Recognition) text-based image analysis model when there is no text in the product data and only a product image;
Is executed by performing
The OCR text-based image analysis model includes text detection for finding a location where text exists from an image and text recognition for extracting text images as text, integrated management of multi shopping malls, and Method of providing a shopping mall Q&A synchronization service for interworking.
The method of claim 1,
The step of extracting the pre-stored Q&A data set based on the pre-mapped and stored text to a pre-stored similar image similar to the product image among the product data,
Extracting at least one feature of the product image by using at least one model learned by deep learning from a database stored in advance when there is no text in the product data and only a product image;
Expressing the extracted at least one feature in a vector space and searching for a similar image similar to the product image using an algorithm or a clustering algorithm for calculating a similarity distance; And,
Extracting text previously mapped to the searched similar image;
A method for providing a shopping mall question-and-answer synchronization service for integrated management and interworking of a multi-shopping mall.
delete delete

Images