Apparatus and method for managing data by means of artificial intelligence-based order sheet analysis

Abstract

The present disclosure relates to an apparatus and a method for managing data by means of artificial intelligence-based order sheet analysis, the apparatus comprising: a communication module for communicating with an external device; a storage module for storing at least one process for managing data by means of artificial intelligence-based order sheet analysis; and a control module for performing an operation for managing data by means of the artificial intelligence-based order sheet analysis on the basis of the at least one process. The control module may: when an order sheet including order information is received, analyze header information and a repeating unit on the basis of the order sheet; determine a repetition rule according to an analysis result so as to identify a mapped template among at least one pre-stored template; and extract data corresponding to each attribute included in the identified template from the order information on the basis of a database.

Description

Data management device and method using artificial intelligence-based order book analysis

[0001] The present disclosure relates to a data management device and method, and more particularly, to a data management device and method through artificial intelligence-based order analysis.

[0002] Luxury brands rarely sell directly to B2B (Business to Business) buyers (parallel importers, retailers, etc.) in each country because they directly enter overseas markets to manage their brands.

[0003] Accordingly, buyers import luxury goods through boutiques (e.g., luxury select shops) in the respective countries that deal directly with luxury brands. Because boutiques and buyers have different tastes and coordination is difficult due to factors such as the availability of free trade agreements between countries, intermediaries, called agencies, assist with import and export operations. These agencies can be categorized into overseas-based agencies, which visit small boutiques in real time to assess inventory and facilitate domestic imports, and domestic-based agencies, which primarily connect with domestic buyers.

[0004] For luxury fashion products, each buyer (local seller) creates and submits an order form based on the order form (including product information) issued by the boutique. Once the order is confirmed, payment and delivery are processed.

[0005] Because there's no standardized format for these order forms, they vary by brand, boutique, and overseas agency. Furthermore, boutiques sometimes reformat and distribute brand-issued order forms in their own formats, contributing to the expansion of order form templates.

[0006] As a result, order forms written in a variety of non-standardized templates are distributed worldwide, resulting in a correspondingly diverse number of purchase orders being created and submitted. Consequently, global B2B transaction information is difficult to standardize by humans. Buyers must spend significant time reviewing and ordering from different templates, and boutiques must spend significant time editing these order forms and collating product information. Furthermore, poor accuracy can lead to incorrect orders, resulting in financial losses.

[0007] In reality, a significant portion of the agency's work time is spent double-checking invoice information, delivery quantities, and other details after order sheets and boutique confirmation. Furthermore, the information in order forms and purchase orders cannot be analyzed or utilized as data.

[0008] Therefore, there is a need to develop a technology that can automatically analyze order forms containing product information based on artificial intelligence, generate purchase orders, and store them in a database.

[0009] The present disclosure provides a data management device and method through artificial intelligence-based order analysis, which can maximize work efficiency by analyzing an order including order information based on artificial intelligence, automatically generating a purchase order, and storing the same in a database, and can utilize the analysis results for demand forecasting as well as real-time order status analysis.

[0010] The problems to be solved by the present disclosure are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

[0011] According to one aspect of the present disclosure, a data management device through artificial intelligence-based order analysis for achieving the above-described technical task comprises: a communication module for performing communication with an external device; a storage module for storing at least one process for managing data through artificial intelligence-based order analysis;And a processor for performing an operation for managing data through the artificial intelligence-based order form analysis based on the at least one process, wherein when an order form including order information is received, the processor analyzes header information and repetition units based on the order form, determines a repetition rule based on the analysis result to check a mapped template among at least one pre-stored template, and extracts data corresponding to each attribute included in the checked template from the order information based on a database, wherein the header information includes at least one of an image, a brand, an item code, a category, a quantity, a size, and a price for each of at least one product included in the order form, and when extracting the data, the processor extracts a classification criterion corresponding to the checked template from among at least one classification criterion set in advance through an artificial intelligence-based dictionary learning model, and automatically classifies data corresponding to each attribute included in the checked template from the order information based on the extracted classification criterion, and the at least one classification criterion includes a criterion for a major classification and a minor classification determined according to a structure of each of the at least one pre-stored templates, and the major classification is classified according to the presence or absence of inventory quantity information and a size notation method for each product, and The subcategories are distinguished according to the format in which information of each product is expressed, and the processor first classifies data by the presence or absence of the stock quantity information during the major classification, and then performs a second classification according to the size notation method, and during the second classification, a table indicating the size format by country is displayed, and a first format in which the size information of the product and the stock quantity for each size are displayed in the lower row of the table, and a second format in which the size information of the product and the stock quantity for each size are displayed in different rows, can be extracted by distinguishing them using different classification criteria.

[0012] In addition, a data management method through artificial intelligence-based order form analysis performed by a processor of a device according to one aspect of the present disclosure comprises the steps of: when an order form including order information is received, analyzing header information and a repetition unit based on the order form; determining a repetition rule based on the analysis result to confirm a template to be mapped among at least one template stored in advance; and extracting data corresponding to each attribute included in the confirmed template from the order information based on a database, wherein the header information includes at least one of an image, a brand, an item code, a category, a quantity, a size, and a price for each of at least one product included in the order form, and the step of extracting the data comprises the steps of: extracting a classification criterion corresponding to the confirmed template from at least one classification criterion set in advance through an artificial intelligence-based pre-learning model; And a step of automatically classifying data corresponding to each attribute included in the confirmed template in the order information based on the extracted classification criteria, wherein the at least one classification criterion includes criteria for major and minor classifications determined according to the structure of each of the at least one template stored in advance, the major classification being classified according to the presence or absence of stock quantity information and the size notation method for each product, and the minor classification being classified according to the format in which information of each product is expressed, and the step of extracting the classification criteria is such that, in the major classification, data is first classified according to the presence or absence of stock quantity information, and then a second classification is performed according to the size notation method, and in the second classification, a table indicating size formats by country is displayed, and a first format in which the size information of the product and the stock quantity for each size are displayed in a lower row of the table, and a second format in which the size information of the product and the stock quantity for each size are displayed in different rows, can be extracted by distinguishing them using different classification criteria.

[0013] In addition, a computer program stored in a computer-readable recording medium for executing a method for implementing the present disclosure may be further provided.

[0014] In addition, a computer-readable recording medium recording a computer program for executing a method for implementing the present disclosure may be further provided.

[0015] According to the aforementioned problem solving means of the present disclosure, by analyzing an order form containing order information based on artificial intelligence to automatically generate a purchase order form and store it in a database, work efficiency can be maximized, and the analysis results can be utilized for demand forecasting as well as real-time order status analysis.

[0016] The effects of the present disclosure are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

[0017] FIG. 1 is a diagram showing a network structure of a data management system according to one embodiment of the present disclosure.

[0018] FIG. 2 is a flowchart showing the operation of a data management system according to one embodiment of the present disclosure.

[0019] FIG. 3 is a diagram showing the configuration of a data management device according to one embodiment of the present disclosure.

[0020] FIG. 4 is a diagram illustrating a data management method according to one embodiment of the present disclosure.

[0021] FIG. 5 is a diagram showing a specific operation of extracting data according to one embodiment of the present disclosure.

[0022] FIG. 6 is a diagram showing a specific operation of automatically classifying data based on a ruleset according to one embodiment of the present disclosure.

[0023] FIG. 7 is a diagram showing an example of automatically classifying data based on a ruleset according to one embodiment of the present disclosure.

[0024] FIG. 8 is a diagram showing a specific operation of automatically classifying data based on ML according to one embodiment of the present disclosure.

[0025] FIG. 9 is a diagram showing a specific operation of creating a purchase order using data extracted based on a ruleset according to one embodiment of the present disclosure.

[0026] The advantages and features of the present disclosure, and methods for achieving them, will become clearer with reference to the embodiments described below in detail with the accompanying drawings. However, the present disclosure is not limited to the embodiments disclosed below and may be implemented in various different forms. These embodiments are provided solely to ensure that the present disclosure is complete and to fully inform those skilled in the art of the scope of the present disclosure, and the present disclosure is defined solely by the scope of the claims.

[0027] The terminology used herein is for the purpose of describing embodiments only and is not intended to limit the present disclosure. In this specification, the singular also includes the plural unless specifically stated otherwise. As used herein, the terms "comprises" and / or "comprising" do not exclude the presence or addition of one or more other components in addition to the mentioned components. Like reference numerals refer to like components throughout the specification, and "and / or" includes each and any combination of one or more of the mentioned components. Although "first", "second", etc. are used to describe various components, these components are not limited by these terms. These terms are only used to distinguish one component from another. Therefore, it should be understood that a first component mentioned below may also be a second component within the technical spirit of the present disclosure.

[0028] Unless otherwise defined, all terms (including technical and scientific terms) used herein may be used in their common sense to those of ordinary skill in the art to which this disclosure pertains. Furthermore, terms defined in commonly used dictionaries are not to be interpreted ideally or excessively unless explicitly and specifically defined otherwise.

[0029] Throughout this disclosure, the same reference numerals denote the same components. This disclosure does not describe all elements of the embodiments, and general contents in the technical field to which this disclosure belongs or contents overlapping between embodiments are omitted. The term "part" or "module" as used in the specification means a hardware component such as a software or FPGA or ASIC, and the "part" or "module" performs certain roles. However, the "part" or "module" is not limited to software or hardware. The "part" or "module" may be configured to be in an addressable storage medium and may be configured to play one or more processors. Thus, as an example, the "part" or "module" includes components such as software components, object-oriented software components, class components, and task components, processes, functions, attributes, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables. The functionality provided within the components and “sub-components” or “modules” may be combined into a smaller number of components and “sub-components” or “modules” or further separated into additional components and “sub-components” or “modules”.

[0030] Throughout the specification, when a part is said to be "connected" to another part, this includes not only direct connection but also indirect connection, and indirect connection includes connection via a wireless communication network.

[0031] Additionally, when a part is said to "include" a component, this does not mean that it excludes other components, but rather that it may include other components, unless otherwise specifically stated.

[0032] Throughout the specification, when we say that an element is "on" another element, this includes not only cases where the element is in contact with the other element, but also cases where another element exists between the two elements.

[0033] The terms first, second, etc. are used to distinguish one component from another, and the components are not limited by the aforementioned terms.

[0034] Singular expressions include plural expressions unless the context clearly indicates otherwise.

[0035] The identification codes for each step are used for convenience of explanation and do not describe the order of each step. Each step may be performed in a different order than specified unless the context clearly indicates a specific order.

[0036] The terms used in the following explanation are defined as follows.

[0037] As used herein, an "order form" includes order information for at least one product, which may be provided in list form. These order forms can be broadly categorized into two types, depending on the ordering method.

[0038] First, a Season-Order (Pre-Order) is a format where customers review pre-production catalogs and place orders. Orders are placed approximately six months before the arrival of the fashion season (SS / FW). The order form consists of product information, manufactured and distributed by the brand, excluding quantity information. In this case, there is no inventory information, and customers can order the desired quantity of each product. A Stock-Order is a format where customers order products already in stock at a boutique. The order form is manufactured and distributed by the boutique, and contains product inventory information, including quantity information.

[0039] Meanwhile, based on the information in this order form, each buyer creates and submits a purchase order containing order information. Some buyers may use the distributed order form as is, while others may create and use their own format.

[0040] Although the description herein is limited to a "data management device (100)", this is a device for managing data through an artificial intelligence-based order form, and may include various devices capable of performing computational processing. That is, the data management device (100) may further include a server, a computer, a server, and / or a portable terminal, or may be in the form of any one of them, but is not limited thereto.

[0041] Here, the computer may include, for example, a notebook, desktop, laptop, tablet PC, slate PC, etc. equipped with a web browser.

[0042] The above server is a server that processes information by communicating with external devices, and may include an application server, a computing server, a database server, a file server, a game server, a mail server, a proxy server, and a web server.

[0043] The above portable terminal may include, for example, a wireless communication device that ensures portability and mobility, and may include all kinds of handheld-based wireless communication devices such as a PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), 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) terminal, a smart phone, and a wearable device such as a watch, a ring, a bracelet, an anklet, a necklace, glasses, contact lenses, or a head-mounted device (HMD).

[0044] In this specification, an artificial intelligence-based pre-learning model can be used to recognize or analyze keywords or commands input to a data management device (100). At this time, as an example of an artificial intelligence algorithm, a Recurrent Neural Network (RNN) or a transformer can be applied, but is not limited thereto, and other artificial intelligence algorithms can also be applied.

[0045] The operating principle and embodiments of the present disclosure are described below with reference to the attached drawings.

[0046] FIG. 1 is a diagram showing a network structure of a data management system according to one embodiment of the present disclosure.

[0047] Referring to FIG. 1, a data management system (10) according to one embodiment of the present disclosure may be configured to include a data management device (100), a user terminal (200), and at least one orderer terminal (300).

[0048] First, the data management device (100) provides a data management service through artificial intelligence-based order analysis based on a separate platform (application) or web page.

[0049] Specifically, when the data management device (100) receives request information requesting registration or execution of a project from the first user terminal (200), it registers or executes the project based on the request information. Here, if the project is new, registration is performed, and if the project is already registered, execution is performed.

[0050] Meanwhile, when an order form containing order information is received from a user terminal (200), the data management device (100) verifies and analyzes the order form using an artificial intelligence-based pre-learning model, extracts data as analysis results, and automatically generates a purchase order form using the extracted data. The order form may be, for example, a sheet created using either PDF or Excel.

[0051] To this end, the data management device (100) can be used by storing at least one template according to the format (structure) of an existing order form (at least one order form already stored) in a database, and for a new order form, a template is created through the analysis result and stored in the database.

[0052] At this time, the purchase order corresponds to one of the ordering parties, i.e., the target company, brand, line, or boutique, and can be set for each ordering party or confirmed through data linkage.

[0053] Meanwhile, the data management device (100) can classify data based on a ruleset or based on ML (Machine Learning) when extracting data, and this will be described in detail below based on FIGS. 6 to 8.

[0054] Additionally, the data management device (100) can transmit the generated purchase order to at least one purchaser terminal (300) registered for the corresponding purchaser.

[0055] The user terminal (200) is a terminal possessed by a seller selling a product, and the user receives a data management service (hereinafter referred to as “data management service”) through artificial intelligence-based order analysis from a data management device (100) through a separate platform (application) or web page.

[0056] Specifically, the user terminal (200) can create and upload (transmit) an order form by entering order information for at least one product to be ordered (placed) by executing a separate platform (application) or web page.

[0057] Here, the order information may include at least one of the product name, item, quantity, size, specification brand, and orderer of at least one product.

[0058] Meanwhile, at least one ordering terminal (300) is a terminal equipped at the ordering terminal, or carried by a person in charge of the ordering terminal, that receives an order for at least one product and performs the product provision (delivery) procedure to the user based on the order. Like the user terminal (200), this at least one ordering terminal (300) may receive data management services through a separate platform (application) or web page, or may be configured to receive only order forms from the data management device (100).

[0059] In addition, the user terminal (200) and at least one ordering party terminal (300) may be a computer, an Ultra Mobile PC (UMPC), a workstation, a netbook, a Personal Digital Assistants (PDA), a portable computer, a web tablet, a wireless phone, a mobile phone, a smart phone, a pad, a smart watch, a wearable terminal, an e-book, a portable multimedia player (PMP), a portable game console, a navigation device, a black box, a digital camera, or other mobile communication terminals, etc., on which a user and / or at least one ordering party (ordering party official / manager / person in charge) can install and execute multiple application programs (i.e., applications) desired by the user. That is, the user terminal (200) and at least one ordering terminal (300) may be provided in various forms, and are not limited thereto.

[0060] FIG. 2 is a flowchart illustrating the operation of a data management system according to one embodiment of the present disclosure.

[0061] Referring to FIG. 2, the data management device (100) collects at least one order including order information for a preset period of time and trains the original model using a learning data set generated to build (generate) a pre-learning model (S101).

[0062] Accordingly, the data management device (100) can initiate a data management service, and when registration request information is received from a user terminal (200) that wishes to receive the data management service (S103), registration for the user is performed based on this (S105). In addition, when registration request information is received from a client terminal (300) that wishes to receive the data management service (S107), registration for the user is performed based on this (S109).

[0063] However, steps S103 to S109 are not mandatory operations and can be omitted.

[0064] Next, the user terminal (200) executes the application (S111), creates an order form by entering order information for at least one product to be ordered through the application (S113), and uploads the created order form to transmit it to the data management device (100) (S115).

[0065] Next, the data management device (100) verifies and analyzes the order received by step S115 (S117), automatically generates an order based on the analysis result (S119), and then transmits the order to the corresponding orderer terminal (300) (S121).

[0066] Accordingly, the ordering terminal (300) can confirm order information for the user terminal (200) based on the order form received by step S121 and perform a provision procedure for at least one product (S123).

[0067] However, although FIG. 2 illustrates one user terminal (200) and one ordering terminal (300) each, this is only for convenience of explanation, and all of them may be configured with at least one or more, and this is not limited thereto.

[0068] FIG. 3 is a diagram showing the configuration of a data management device according to one embodiment of the present disclosure.

[0069] Referring to FIG. 3, a data management device (100) according to one embodiment of the present disclosure may be configured to include a communication module (110), a storage module (120), and a control module (130).

[0070] The communication module (110) transmits and receives at least one piece of information or data with at least one device / terminal. Here, the at least one device / terminal may be a device / terminal that provides a data management service, or another device / terminal that receives the data management service, and its type and form are not limited.

[0071] In addition, this communication module (110) can also perform communication with other devices, and transmit and receive wireless signals in a communication network according to wireless Internet technologies.

[0072] Wireless Internet technologies include, for example, WLAN (Wireless LAN), Wi-Fi (Wireless-Fidelity), Wi-Fi (Wireless Fidelity) Direct, DLNA (Digital Living Network Alliance), WiBro (Wireless Broadband), WiMAX (World Interoperability for Microwave Access), HSDPA (High Speed ​​Downlink Packet Access), HSUPA (High Speed ​​Uplink Packet Access), LTE (Long Term Evolution), LTE-A (Long Term Evolution-Advanced), etc., and the data management device (100) transmits and receives data according to at least one wireless Internet technology, including Internet technologies not listed above.

[0073] As for short range communication, short range communication can be supported using at least one of Bluetooth™, RFID (Radio Frequency Identification), Infrared Data Association (IrDA), UWB (Ultra Wideband), ZigBee, NFC (Near Field Communication), Wi-Fi (Wireless-Fidelity), Wi-Fi Direct, and Wireless USB (Wireless Universal Serial Bus) technologies. Such short range wireless communication networks (Wireless Area Networks) can support wireless communication between a data management device (100), a user terminal (200), and at least one orderer terminal (300). At this time, the short range wireless communication network can be a short range wireless personal area network (Wireless Personal Area Network).

[0074] The storage module (120) can store data for at least one process (algorithm) for managing data through AI-based order analysis, or a program that reproduces that process. Furthermore, the storage module (120) can further store processes for performing other operations, but is not limited thereto.

[0075] The storage module (120) can store various information / data required for managing data through artificial intelligence-based order analysis, as well as various other data supporting various functions of the data management device (100). The storage module (120) can store a plurality of application programs (or applications) driven by the data management device (100), data for the operation of the data management device (100), and commands. At least some of these application programs can be downloaded from an external server via wireless communication. Meanwhile, the application programs can be stored in at least one memory provided in the storage module (120), installed on the data management device (100), and driven to perform operations (or functions) by at least one processor stored in the storage module (120) through the control module (130).

[0076] Meanwhile, at least one memory may include at least one type of storage medium among a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (e.g., SD or XD memory, etc.), a random access memory (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, and an optical disk. In addition, the memory may store information temporarily, permanently, or semi-permanently, and may be provided as a built-in or removable type.

[0077] In addition, the storage module (120) may build a database that stores various information necessary for managing data through artificial intelligence-based order analysis, or may be linked to a separate external server (including a cloud server).

[0078] Meanwhile, the control module (130) controls all components within the data management device (100) based on at least one processor in addition to operations related to the application program, to process input or output signals, data, information, etc., or executes commands, algorithms, and application programs stored in at least one memory to perform various processes, and can provide or process appropriate information or functions for managing data through artificial intelligence-based order analysis.

[0079] Specifically, when an order form including order information is received, the control module (130) analyzes header information and repetition units based on the order form, determines a repetition rule based on the analysis result, checks a template to be mapped among at least one template stored in advance, and then extracts data corresponding to each attribute included in the previously checked template from the order information included in the order form based on a database.

[0080] Here, the header information may include at least one or a combination of two or more of the following: product image, brand, item code, item name, gender, category, color, size, price, and quantity for each product included in the order form. The categories may be subdivided into main (or parent) categories and sub (or child) categories. Furthermore, the prices may be subdivided into supply price, cost price, and consumer price. Furthermore, the quantity may only be entered for products already in stock, and may be omitted for products manufactured to order, such as custom orders.

[0081] At this time, the control module (130) analyzes the repetition unit based on the header information and determines the repetition rule based on the analysis result.

[0082] Meanwhile, when extracting data, the control module (130) identifies a classification criterion corresponding to a previously identified template from at least one pre-established classification criterion through an artificial intelligence-based pre-learning model, and automatically classifies data corresponding to each attribute included in the previously identified template from the order information based on the identified classification criterion. This data extraction operation is performed repeatedly by the control module (130) by moving and substituting rows or columns based on the identified classification criterion until no more data is extracted.

[0083] For example, the control module (130) extracts each product information, quantity information, etc. from the order form as data and then loads the data into a database. Since the data within each order form may have different column or row locations, the location of the first product from which all information is extracted is found by substituting the identified classification criteria while moving the rows / columns one by one. Thereafter, the row is moved according to the identified classification criteria and the extraction of information is repeated until no more information can be extracted.

[0084] At this time, at least one of the preset classification criteria can be configured based on a ruleset or machine learning (ML) system. For example, in the case of a ruleset, criteria for major and minor categories can be established based on the structure of each of at least one template pre-stored in the database. Major categories can be distinguished based on the presence or absence of inventory quantity information and the size notation method for each product, while minor categories can be distinguished based on the format in which information for each product is expressed.

[0085] Meanwhile, if the control module (130) does not identify a template to be mapped among at least one of the pre-stored templates, it uploads the order form as a new order form to the database, and determines an appropriate classification criterion by inputting at least one of the pre-set classification criteria into the order information of the order form through an artificial intelligence-based pre-learning model, and then automatically classifies the data corresponding to each attribute in the order information based on the determined classification criterion. If there is no appropriate criterion, the control module (130) separately classifies it with a new classification criterion and then updates at least one of the pre-set classification criteria.

[0086] In addition, when configuring a database, the control module (130) may design terms that express the characteristics of each of a plurality of products in a variety of ways (for example, when Italian and English are used in two order forms for the characteristics of one product, or when the characteristics of one product are expressed as 'color' in one order form and expressed in detail as 'main color' and 'sub color' in another order form), or may configure terms that express the characteristics of the products in a unified manner, or may configure them as a union of the characteristics of the products.

[0087] Meanwhile, the control module (130) can load the extracted data into the database as data corresponding to the order form, and compare the number of rows of the order form and the number of loaded data to determine whether it is operating normally.

[0088] In addition, the specific operation of the control module (130) will be described below based on each drawing.

[0089] FIG. 4 is a diagram illustrating a data management method according to one embodiment of the present disclosure.

[0090] Referring to FIG. 4, when an order form including order information is received from a user terminal (200), the data management device (100) analyzes header information and repetition units based on the order form (S210).

[0091] Next, the data management device (100) determines a repetition rule based on the analysis result by step S210 to check a template to be mapped among at least one template stored in advance (S220), and extracts data corresponding to each attribute included in the template checked by step S220 from the order information based on the database (S230).

[0092] Here, the database is standardized, as described above, by unifying the terminology as a union of characteristics representing the characteristics of each of a plurality of products in each of at least one template stored in advance based on at least one accumulated and stored order.

[0093] FIG. 5 is a drawing showing a specific operation of extracting data according to one embodiment of the present disclosure, specifically illustrating step S230 of FIG. 4.

[0094] Referring to FIG. 5, the data management device (100) confirms the classification criterion corresponding to the confirmed template from at least one classification criterion set in advance through an artificial intelligence-based pre-learning model (S231).

[0095] Next, the data management device (100) automatically classifies data corresponding to each attribute included in the template confirmed by step S220 by extracting the data from the order information based on the classification criteria confirmed by step S231.

[0096] At this time, at least one of the preset classification criteria can be configured based on a ruleset or ML (Machine Learning).

[0097] FIG. 6 is a diagram illustrating a specific operation of automatically classifying data based on a ruleset according to one embodiment of the present disclosure.

[0098] First, if at least one of the preset classification criteria is configured based on a ruleset, criteria for major and minor classifications determined according to the structure of each of at least one of the preset templates can be set.

[0099] Referring to FIG. 6, the data management device (100) confirms the major and minor classifications based on the classification criteria of the template confirmed by step S220 (S311), and automatically classifies data based on the confirmed major and minor classification criteria (S312).

[0100] FIG. 7 is a diagram illustrating an example of automatically classifying data based on a ruleset according to one embodiment of the present disclosure.

[0101] First, the major classifications can be distinguished according to the presence or absence of inventory quantity information and the method of indicating the size of each product. If the order form is related to Season-Order, it can be written in a format where the quantity to be ordered is entered without inventory quantity information, as in (a). After the data is first classified according to the presence or absence of inventory quantity information, a second classification (reclassification) can be performed according to the size notation method.

[0102] Meanwhile, as in (b), the order form includes size information at the top (a table that organizes country-specific size formats such as A: SS, S, M, L, … or B: 36, 37, 38, …), and the row containing information on each product can be classified into a format in which there is a column (A, B, etc.) expressing the size information and a column (numbers such as 1, 2, etc.) expressing the stock quantity for each size.

[0103] In addition, as in (c), the row expressing information about a product is composed of two rows, and can be classified into a format in which the size type is expressed in the upper row and the stock quantity is expressed in the lower row.

[0104] In addition, subcategories can be classified according to the format in which information about each product is expressed. For example, (d) can be classified into a format in which product characteristics are expressed in each column in one row, or (e) can be classified into a format in which product characteristics are expressed in each column but two rows are used, quantity information is expressed in two rows, and characteristics of each product are expressed by combining the two rows.

[0105] Additionally, it can be classified into a format that uses multiple rows, such as (f), to express product information separated by spaces in a single column. This (f) shows a state where product information above the size is combined into a single cell.

[0106] FIG. 8 is a diagram illustrating a specific operation of automatically classifying data based on ML according to one embodiment of the present disclosure.

[0107] Referring to FIG. 8, the data management device (100) identifies the meaning of each cell in the order received by step S210 (S321), and automatically classifies data by identifying the classification criteria corresponding to the meaning of each cell identified by step S321 based on the classification criteria corresponding to the template identified by step S220 (S322).

[0108] Comparing the rule-set-based configuration and the ML-based configuration described above, unlike the rule-set-based configuration where the user sets the classification criteria in a top-down manner, the ML-based configuration classifies the meaning of each cell's coordinates in a bottom-up manner, sets criteria for classification by cell, and then collects the classifications of each cell to create a single classification criterion. For example, if cell A1 is classified as 'product category' and cell B1 is 'price', and a single classification criterion is created where cell A1 is the product category and cell B1 is the price, a much more diverse range of criteria can be set in detail compared to the rule-set-based configuration defined directly by the user, and not only are the categories that can be responded to detailed, but new classification criteria can also be automatically set when a new type of order is entered.

[0109] Meanwhile, when a new order form is entered, the system searches for an appropriate criterion among the established classification criteria. For example, it searches for the criterion with the highest similarity based on the similarity between the generated categories and the meaning of each cell. Subsequent operations are identical to ruleset-based configuration.

[0110] Meanwhile, in FIG. 3, the control module (130) is depicted as performing the overall operation, but for the rule set-based configuration described above, an order classification module and / or an order information loading module may be separately provided, and for the ML-based configuration, a classification criterion creation module and / or a criterion-based classification module may be separately provided, and the types and numbers of the modules are not limited.

[0111] FIG. 9 is a diagram illustrating a specific operation of creating a purchase order using data extracted based on a ruleset according to one embodiment of the present disclosure.

[0112] Referring to FIG. 9, the data management device (100) verifies the orderer based on the order form received by step S210 (S240), and automatically generates an order form corresponding to the orderer, i.e., one of the target company, brand, line, and boutique, using the data extracted by step S230 (S250).

[0113] Next, the data management device (100) transmits the purchase order automatically generated by step S250 to at least one purchaser terminal (300) corresponding to the purchaser (S260).

[0114] Meanwhile, the present disclosure performs prediction (inference) for a predetermined purpose using a model implemented in an artificial neural network manner. In the following, the artificial neural network will be examined.

[0115] A model in this specification may refer to any form of computer program that operates based on a network function, an artificial neural network, and / or a neural network. Throughout this specification, the terms model, neural network, network function, and neural network may be used interchangeably. A neural network is a network in which one or more nodes are interconnected through one or more links to form input node and output node relationships within the neural network. The characteristics of a neural network can be determined based on the number of nodes and links within the neural network, the correlation between the nodes and links, and the weight value assigned to each link. A neural network may be composed of a set of one or more nodes. A subset of the nodes constituting the neural network may constitute a layer.

[0116] A deep neural network (DNN) can refer to a neural network that includes multiple hidden layers in addition to the input layer and the output layer, and the hidden layer in the middle is composed of one or more, preferably two or more, layers in the deep neural network.

[0117] These deep neural networks may include convolutional neural networks (CNNs), vision transformers, recurrent neural networks (RNNs), long short-term memory (LSTM) networks, generative pre-trained transformers (GPTs), autoencoders, generative adversarial networks (GANs), restricted boltzmann machines (RBMs), deep belief networks (DBNs), Q-networks, U-networks, Siamese networks, generative adversarial networks (GANs), transformers, etc.

[0118] Alternatively, depending on the embodiment, the deep neural network may be a model trained using transfer learning. Here, transfer learning refers to a learning method in which a large amount of unlabeled training data is pre-trained using a semi-supervised learning or self-learning method to obtain a pre-trained model (or base part) having a first task using techniques such as MLM and NSP as shown in FIGS. 11 and 12, and then the pre-trained model is trained using labeled training data using a supervised learning method to fine-tune it to be suitable for a second task, thereby implementing a target model. One of the models trained using this transfer learning method includes, but is not limited to, BERT (Bidirectional Encoder Representations from Transformers).

[0119] The description of the aforementioned deep neural network is merely an example, and the present disclosure is not limited thereto. In the case of the aforementioned convolutional neural network, it is composed of a feature extraction unit (feature learning) that extracts features from an image, and a classification unit (classification) that performs classification using the extracted features. The feature extraction unit may include, but is not limited to, a convolution layer that extracts features from an image using a kernel, a ReLU layer as one of the activation functions, and a pooling layer for reducing the dimensionality of the data. In addition, the classification unit may include, but is not limited to, a flatten layer that lines up the features extracted from the feature extraction unit, and a fully connected layer and a softmax function that actually perform classification.

[0120] Neural networks can learn through at least one of the following methods: supervised learning, unsupervised learning, semi-supervised learning, self-supervised learning, or reinforcement learning. Neural network learning can be the process of applying knowledge to the neural network to perform a specific action.

[0121] Neural networks can be trained to minimize output errors. This process involves repeatedly inputting training data into the neural network, calculating the neural network output and target error for the training data, and backpropagating the neural network error from the output layer to the input layer to update the weights of each node in the neural network to reduce the error. In supervised learning, labeled data is used for each training data, while unsupervised learning uses unlabeled data. The amount of change in the connection weights of each updated node can be determined by the learning rate. The neural network's calculation of input data and backpropagation of errors can constitute a learning cycle (epoch). The learning rate can vary depending on the number of iterations in the neural network's training cycle. Additionally, to prevent overfitting, methods such as increasing the learning data, regularization, dropout that disables some nodes, and batch normalization layers can be applied.

[0122] Meanwhile, the model disclosed in one embodiment may borrow at least a portion of a transformer. The transformer may be composed of an encoder that encodes embedded data and a decoder that decodes the encoded data. The transformer may have a structure that receives a series of data, performs encoding and decoding steps, and outputs a series of data of different types. In one embodiment, the series of data may be processed into a form operable by the transformer. The process of processing the series of data into a form operable by the transformer may include an embedding process. Expressions such as data tokens, embedding vectors, and embedding tokens may refer to data embedded in a form operable by the transformer.

[0123] To encode and decode a series of data, a transformer can utilize an attention algorithm to process the encoders and decoders within the transformer. An attention algorithm can refer to an algorithm that, for a given query, calculates the similarity for one or more keys, reflects this similarity in the values ​​corresponding to each key, and then weights and sums the values ​​to which the similarity is reflected to calculate an attention value.

[0124] Depending on how the query, key, and value are configured, various types of attention algorithms can be categorized. For example, if attention is obtained by setting the query, key, and value all to the same value, this could be a self-attention algorithm. If attention is obtained by reducing the dimensionality of the embedding vector to process a series of input data in parallel and then generating individual attention heads for each segmented embedding vector, this could be a multi-head attention algorithm.

[0125] In one embodiment, the transformer may be composed of modules that perform multiple multi-head self-attention algorithms or multi-head encoder-decoder algorithms. In one embodiment, the transformer may also include additional components other than attention algorithms, such as embedding, normalization, and softmax. Methods for constructing a transformer using attention algorithms may include methods disclosed in Vaswani et al., Attention Is All You Need, 2017 NIPS, which is incorporated herein by reference.

[0126] A transformer can be applied to various data domains, such as embedded natural language, segmented image data, and audio waveforms, to transform a series of input data into a series of output data. To transform data with various data domains into a series of data that can be input to a transformer, the transformer can embed the data. The transformer can process additional data that expresses the relative positional relationship or phase relationship between the series of input data. Alternatively, vectors expressing the relative positional relationship or phase relationship between the input data can be additionally reflected in the series of input data to embed the series of input data. In one example, the relative positional relationship between the series of input data may include, but is not limited to, word order within a natural language sentence, the relative positional relationship between each segmented image, and the time order of segmented audio waveforms. The process of adding information expressing the relative positional relationship or phase relationship between the series of input data may be referred to as positional encoding.

[0127] The above-described program may include codes coded in a computer language, such as C, C++, JAVA, or machine language, that can be read by the processor (CPU) of the computer through the device interface of the computer, so that the computer reads the program and executes the methods implemented as a program. Such codes may include functional codes related to functions that define functions necessary for executing the methods, and may include control codes related to execution procedures necessary for the processor of the computer to execute the functions according to a predetermined procedure. In addition, such codes may further include memory reference-related codes regarding which location (address address) of the internal or external memory of the computer should reference additional information or media necessary for the processor of the computer to execute the functions. In addition, if the processor of the computer needs to communicate with any other computer or server located remotely in order to execute the functions, the code may further include communication-related code regarding how to communicate with any other computer or server located remotely using the communication module of the computer, and what information or media to send and receive during communication.

[0128] The above storage medium refers to a medium that stores data semi-permanently and can be read by a device, rather than a medium that stores data for a short period of time, such as a register, cache, or memory. Specifically, examples of the storage medium include, but are not limited to, ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage device. That is, the program can be stored in various recording media on various servers that the computer can access or in various recording media on the user's computer. In addition, the medium can be distributed across network-connected computer systems, so that computer-readable code can be stored in a distributed manner.

[0129] The steps of a method or algorithm described in connection with the embodiments of the present disclosure may be implemented directly in hardware, implemented as a software module executed by hardware, or implemented by a combination thereof. The software module may reside in a random access memory (RAM), a read only memory (ROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a flash memory, a hard disk, a removable disk, a CD-ROM, or any other form of computer-readable recording medium well known in the art to which the present disclosure pertains.

[0130] While the embodiments of the present disclosure have been described above with reference to the attached drawings, those skilled in the art will appreciate that the present disclosure can be implemented in other specific forms without altering the technical spirit or essential features thereof. Therefore, the embodiments described above should be understood to be illustrative in all respects and not restrictive.

Claims

1. Communication module that communicates with external devices; A storage module storing at least one process for managing data through artificial intelligence-based order book analysis; and A processor that performs an operation for managing data through the artificial intelligence-based order analysis based on at least one of the above processes, The above processor, when an order form including order information is received, analyzes header information and repetition units based on the order form, determines a repetition rule based on the analysis result, checks a template to be mapped among at least one template stored in advance, and extracts data corresponding to each attribute included in the checked template from the order information based on a database. The above header information includes at least one of an image, brand, item code, category, quantity, size, and price for each of at least one product included in the order form. The processor, when extracting the data, extracts a classification criterion corresponding to the confirmed template from among at least one classification criterion set in advance through an artificial intelligence-based pre-learning model, and automatically classifies data corresponding to each attribute included in the confirmed template in the order information based on the extracted classification criterion. The at least one classification criterion includes criteria for major and minor classifications determined according to the structure of each of the at least one template stored in advance, The above major categories are classified according to the presence or absence of inventory quantity information and the size notation method for each product. The above subcategories are classified according to the format in which information about each product is expressed. The processor is characterized in that, after first classifying data by the presence or absence of the stock quantity information during the major classification, it performs a second classification according to the size notation method, and in the second classification, a table showing the size format by country is displayed, and a first format in which the size information of the product and the stock quantity for each size are displayed in the lower row of the table, and a second format in which the size information of the product and the stock quantity for each size are displayed in different rows are extracted by distinguishing them using different classification criteria. Data management device through AI-based order book analysis.

2. In paragraph 1, The above processor, Using the above extracted data, an order form corresponding to one of the target companies, brands, lines, and boutiques is automatically created and sent, The above order form, Characterized in that it is preset or confirmed through linkage in response to any one of the above target companies, brands, lines and boutiques. Data management device through AI-based order book analysis.

3. In paragraph 1, The above processor, When extracting the above data, the data is repeatedly extracted by moving and substituting rows or columns based on the confirmed classification criteria until no more data is extracted. Data management device through AI-based order book analysis.

4. In paragraph 1, The above database is, It is characterized in that the characteristics of each of the multiple products are composed of a union of characteristics expressed in each of the at least one template. Data management device through AI-based order book analysis.

5. In paragraph 1, The above processor, Characterized in that the extracted data is loaded into the database as data corresponding to the order form, and the number of rows of the order form and the number of loaded data are compared to determine whether normal operation is performed. Data management device through AI-based order book analysis.

6. In a data management method through artificial intelligence-based order analysis performed by the processor of the device, When an order form including order information is received, a step of analyzing header information and repeating units based on the order form; A step of determining a repetition rule based on the analysis result and checking a template to be mapped among at least one template stored in advance; and A step of extracting data corresponding to each attribute included in the confirmed template from the order information based on the database, The above header information includes at least one of an image, brand, item code, category, quantity, size, and price for each of at least one product included in the order form. The steps for extracting the above data are: A step of extracting a classification criterion corresponding to the identified template from among at least one classification criterion set in advance through an artificial intelligence-based pre-learning model; and A step of automatically classifying data corresponding to each attribute included in the confirmed template in the order information based on the extracted classification criteria is included. The at least one classification criterion includes criteria for major and minor classifications determined according to the structure of each of the at least one template stored in advance, The above major categories are classified according to the presence or absence of inventory quantity information and the size notation method for each product. The above subcategories are classified according to the format in which information about each product is expressed. The step of extracting the above classification criteria is: In the above classification, the data is first classified based on the presence or absence of the above inventory quantity information, and then a second classification is performed based on the above size notation method. In the second classification above, a table showing the size format by country is displayed, and the first format in which the size information of the product and the stock quantity for each size are displayed in the lower row of the table, and the second format in which the size information of the product and the stock quantity for each size are displayed in different rows are extracted by distinguishing them using different classification criteria. Data management method using artificial intelligence-based order book analysis.

7. In paragraph 6, The above processor, Using the above extracted data, an order form corresponding to one of the target companies, brands, lines, and boutiques is automatically created and sent, The above order form, Characterized in that it is preset or confirmed through linkage in response to any one of the above target companies, brands, lines and boutiques. Data management method using artificial intelligence-based order book analysis.

8. In paragraph 6, The above processor, When extracting the above data, the data is repeatedly extracted by moving and substituting rows or columns based on the confirmed classification criteria until no more data is extracted. Data management method using artificial intelligence-based order book analysis.

9. In paragraph 6, The above database is, It is characterized in that the characteristics of each of the multiple products are composed of a union of characteristics expressed in each of the at least one template. Data management method using artificial intelligence-based order book analysis.

10. In paragraph 6, The above processor, Characterized in that the extracted data is loaded into the database as data corresponding to the order form, and the number of rows of the order form and the number of loaded data are compared to determine whether normal operation is performed. Data management method using artificial intelligence-based order book analysis.

Images