A financial server and method for providing customized marketing content to customers by using customer finalcial transaction information obtained from multiple finalcial institutions, and a system therefor

The financial server classifies and models customer transaction data to generate personalized marketing content, addressing the limitations of existing systems by providing tailored lifestyle-related content based on comprehensive transaction analysis.

KR1020260163643APending Publication Date: 2026-11-17K -BANK CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
KR1020250059604
Authority / Receiving Office
KR · KR
Patent Type
Applications
Current Assignee / Owner
Filing Date
2025-05-08
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing financial transaction systems lack the ability to effectively classify and utilize customer transaction information beyond deposit and withdrawal data, limiting the provision of customized lifestyle-related content and marketing.

Method used

A financial server that classifies customer transaction information from multiple financial institutions using payment codes, summary keywords, and industry codes, and applies model training to generate customized marketing content through a Text CNN process.

Benefits of technology

Enables differentiated financial services by providing personalized marketing content tailored to customers' lifestyles and transaction patterns, enhancing customer engagement and service relevance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure PAT00002_ABST
    Figure PAT00002_ABST
Patent Text Reader

Abstract

The present invention relates to a financial server, a method, and a system for providing customized marketing content to a customer using financial transaction information obtained from multiple financial institutions. To this end, the present invention obtains financial transaction information from each of a plurality of financial institutions with which a customer transacts, first classifies deposit and withdrawal transaction details among the obtained financial transaction information based on a payment code, secondly classifies the results of the first classification based on summary keywords, thirdly classifies check card payment details among the obtained financial transaction information based on an industry code, trains a model using the results obtained through the second classification and the third classification, and can provide customized marketing content to the customer's mobile terminal using the results of the model training.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The present invention relates to a financial server, a method, and a system for providing customized marketing content to a customer using financial transaction information obtained from multiple financial institutions. Background Technology

[0002] Generally, with the widespread adoption of mobile devices, banks are developing applications installed on these devices to provide financial transactions, and users can download and install these applications on their mobile devices. Users then receive financial services through these installed applications.

[0003] Furthermore, users can check financial payment information by viewing accounts opened at multiple banks in an integrated manner through applications provided by specific banks. While this financial payment information can be utilized for improvements such as the development of customized financial products or channel diversification, there were limitations in providing customized content to customers based solely on current deposit and withdrawal account transaction data.

[0004] In addition, conventionally, there was difficulty in identifying daily life transactions in addition to financial transactions.

[0005] Therefore, there is a need for technology that automatically classifies customers' financial transaction information and provides customized lifestyle-related messages and content to customers. Prior art literature

[0006] Korean Registered Patent Publication No. 10-2797925 (April 15, 2025) The problem to be solved

[0007] Accordingly, the present invention provides a financial server, a method, and a system for providing customized marketing content to a customer using financial transaction information of a customer obtained from a plurality of financial institutions.

[0008] In addition, the present invention provides a financial server and method that classifies customer deposit and withdrawal transaction history information, open banking inquiry information, check card payment information, etc., into a form that can be utilized in actual business, and provides customized marketing to customers, and a system for the same.

[0009] The objects of the present invention are not limited to those mentioned above, and other unmentioned objects and advantages of the present invention may be understood from the following description and will be more clearly understood by the embodiments of the present invention. Furthermore, it will be readily apparent that the objects and advantages of the present invention can be realized by the means and combinations thereof set forth in the claims. means of solving the problem

[0010] To achieve this purpose, a financial server that provides customized marketing content to a customer using financial transaction information obtained from a plurality of financial institutions according to an embodiment of the present invention comprises: a communication unit that transmits and receives data with at least one mobile terminal; a memory that stores financial transaction information of a customer; and a processor electrically connected to the communication unit and the memory. The processor obtains financial transaction information from each of the plurality of financial institutions with which the customer transacts, first classifies deposit and withdrawal transaction details among the obtained financial transaction information based on a payment code, secondly classifies the results of the first classification based on a summary keyword, thirdly classifies check card payment details among the obtained financial transaction information based on an industry code, and model-trains the results obtained through the second classification and the third classification, and can provide customized marketing content to the customer's mobile terminal using the results of the model-trained results.

[0011] In addition, a method performed on a financial server that provides customized marketing content to a customer using financial transaction information obtained from a plurality of financial institutions according to an embodiment of the present invention may include: a process of obtaining financial transaction information from each of the plurality of financial institutions with which the customer transacts; a process of first classifying deposit and withdrawal transaction details among the obtained financial transaction information based on a payment code; a process of second classifying the results of the first classification based on a summary keyword; a process of third classifying check card payment details among the obtained financial transaction information based on an industry code; a process of model training the results obtained through the second classification and the third classification; and a process of providing customized marketing content to the customer's mobile terminal using the results of the model training.

[0012] In addition, a system for providing customized marketing content to a customer using financial transaction information obtained from a plurality of financial institutions according to an embodiment of the present invention may include: a financial server that obtains financial transaction information from each of the plurality of financial institutions with which the customer transacts, first classifies deposit and withdrawal transaction details among the obtained financial transaction information based on a payment code, secondly classifies the results of the first classification based on a summary keyword, thirdly classifies check card payment details among the obtained financial transaction information based on an industry code, model-trains the results obtained through the second classification and the third classification, and provides customized marketing content to the customer's mobile terminal using the results of the model-trained results; and the mobile terminal that displays the customized marketing content provided from the financial server through an application. Effects of the invention

[0013] According to the present invention, financial transaction information obtained from each of a plurality of financial institutions is first classified based on a payment code, the results of the first classification are second classified based on summary keywords, check card payment history among the obtained financial transaction information is third classified based on an industry code, and customized marketing content can be provided to a customer's mobile terminal by model learning the results obtained through the second classification and the third classification.

[0014] In addition, the present invention can provide differentiated financial services by providing marketing content customized to customers.

[0015] In addition, the present invention can provide differentiated customized marketing to customers by estimating contracts or periodic transactions of other financial institutions.

[0016] In addition to the effects described above, the specific effects of the present invention are described together with the specific details for implementing the invention below. Brief explanation of the drawing

[0017] FIG. 1 is an exemplary diagram showing a system for providing customized marketing content to a customer using financial transaction information of a customer obtained from a plurality of financial institutions according to one embodiment of the present invention. FIG. 2 is a block diagram of a financial server that provides customized marketing content to a customer using financial transaction information obtained from a plurality of financial institutions according to one embodiment of the present invention. FIG. 3 is a flowchart illustrating the process of providing customized marketing content to a customer using financial transaction information obtained from a plurality of financial institutions according to one embodiment of the present invention. FIG. 4(a) is an example of a transaction history according to one embodiment of the present invention. Figure 4(b) is an example diagram showing financial transaction information classified according to one embodiment of the present invention. FIG. 5(a) is an example diagram of financial transaction information classified based on code according to one embodiment of the present invention. FIG. 5(b) is an example diagram of financial transaction information classified based on summary keywords according to one embodiment of the present invention. FIG. 6 is an example diagram of financial transaction information according to one embodiment of the present invention, in which check card payment details are classified by industry code. FIG. 7 is an example diagram showing data for generating customized marketing content by analyzing a customer's lifestyle pattern through model learning using financial transaction information according to one embodiment of the present invention. FIG. 8 is an example diagram illustrating the advancement of customized marketing content according to one embodiment of the present invention. FIG. 9 is an example diagram showing an application of a mobile terminal displaying customized marketing content according to one embodiment of the present invention. Specific details for implementing the invention

[0018] The aforementioned objectives, features, and advantages are described in detail below with reference to the attached drawings, thereby enabling those skilled in the art to easily implement the technical concept of the present invention. In describing the present invention, detailed descriptions of known technologies related to the present invention are omitted if it is determined that such descriptions would unnecessarily obscure the essence of the invention. Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the attached drawings. In the drawings, the same reference numerals are used to indicate the same or similar components.

[0019] Although terms such as "first," "second," etc., are used to describe various components, it goes without saying that these components are not limited by these terms. These terms are used merely to distinguish one component from another, and unless specifically stated otherwise, the first component may also be the second component.

[0020] In the following, the statement that any configuration is placed on the "upper (or lower)" of a component or on the "upper (or lower)" of a component may mean not only that any configuration is placed in contact with the upper (or lower) surface of said component, but also that another configuration may be interposed between said component and any configuration placed on (or below) said component.

[0021] In addition, where it is stated that one component is "connected," "combined," or "connected" to another component, it should be understood that while the components may be directly connected or connected to each other, another component may be "interposed" between each component, or each component may be "connected," "combined," or "connected" through another component.

[0022] Throughout the specification, unless specifically stated otherwise, each component may be singular or plural.

[0023] Singular expressions used in this specification include plural expressions unless the context clearly indicates otherwise. In this application, terms such as "composed of" or "comprising" should not be interpreted as necessarily including all of the various components or steps described in the specification, and should be interpreted as meaning that some of the components or steps may be omitted or additional components or steps may be included.

[0024] Throughout the specification, "A and / or B" means A, B, or A and B unless specifically stated otherwise, and "C to D" means C or more and D or less, unless specifically stated otherwise.

[0025] Hereinafter, a financial server, a method, and a system for providing customized marketing content to a customer using financial transaction information obtained from a plurality of financial institutions according to several embodiments of the present invention will be described.

[0026] FIG. 1 is an exemplary diagram showing a system for providing customized marketing content to a customer using financial transaction information of a customer obtained from a plurality of financial institutions according to one embodiment of the present invention.

[0027] Referring to FIG. 1, a system (100) for providing customized marketing content to a customer using financial transaction information (e.g., MyData) obtained from a plurality of financial institutions according to one embodiment of the present invention includes a mobile terminal (110) and a financial server (120), and the transmission and reception of data between the mobile terminal (110) and the financial server (120) is performed through a network (101).

[0028] According to one embodiment, the portable terminal (110) includes an electronic device that can be carried by an individual, such as a smartphone or a laptop, and has or can have an application (e.g., K-Bank application) installed that provides financial transactions (e.g., deposits, savings, loans, etc.).

[0029] According to one embodiment, a mobile terminal (110) provides various financial products and marketing content regarding each financial product to a customer through an application (e.g., K-Bank application), and the customer can use financial services (e.g., deposits, installments, funds, loans, consultations, etc.) provided through the application (e.g., K-Bank application).

[0030] According to one embodiment, the financial server (120) may obtain financial transaction information from each of the multiple financial institutions with which the customer transacts. The financial transaction information may include the customer's deposit and withdrawal transaction history, open banking transaction history, credit card transaction history, and debit card transaction history obtained from each of the multiple financial institutions.

[0031] According to one embodiment, the financial server (120) can classify the deposit and withdrawal transaction details among the acquired financial transaction information in a first stage based on a payment code. The financial server (120) can classify the acquired financial transaction information in a first stage using a transaction summary classification code and a network classification code.

[0032] According to one embodiment, the financial server (120) can classify the first classified results a second time using a summary keyword. The financial server (120) can classify the first classified results a second time using a summary value, a deposit passbook display content, and a withdrawal passbook display content.

[0033] According to one embodiment, the financial server (120) can classify check card payment details among the acquired financial transaction information into a third category using an industry code. The financial server (120) can classify the acquired financial transaction information into a third category using a merchant industry code.

[0034] According to one embodiment, the financial server (120) can model-learn the results obtained through the second classification and the third classification. The financial server (120) can perform a tokenization process to subdivide the text included in the acquired financial transaction information into words, and a stop word removal process to remove articles, prepositions, and particles from the text. Additionally, the financial server (120) can perform a normalization process to check spelling and adjust spacing in the text, perform an embedding process to convert the text into machine language and classify it by the same transaction, and generate customized marketing content by learning the results classified through the tokenization process, the stop word removal process, the normalization process, and the embedding process through a Text CNN (Convolutional Neural Network).

[0035] According to one embodiment, the financial server (120) can provide customized marketing content generated using the model learned results to the customer's mobile terminal (110) via the network (101). The customized marketing content can be displayed through an application installed on the customer's mobile terminal (110).

[0036] According to one embodiment, the application server (130) can support dynamic content generation, application logic, and integration with various resources. Additionally, the application server (130) can provide a runtime environment capable of executing application code and interacting with other software components, such as a messaging system and a database.

[0037] These application servers (130) mostly provide dynamic content, which is content that changes depending on how the user interacts with the content, and may include dynamically generated reports, customized data representations, personalized UI (User Interface), database results, and processed HTML (HyperText Markup Language).

[0038] According to one embodiment, the application server (130) can provide marketing content customized to each customer to the customer based on a connection using an application on the customer's mobile terminal (110) under the control of the financial server (120).

[0039] This application server (130) may be mounted on the financial server (120) or configured as a separate device.

[0040] FIG. 2 is a block diagram of a financial server that provides customized marketing content to a customer using financial transaction information obtained from a plurality of financial institutions according to one embodiment of the present invention.

[0041] Referring to FIG. 2, a financial server (120) that provides customized marketing content to a customer using financial transaction information obtained from a plurality of financial institutions according to one embodiment of the present invention may include a communication unit (210), a memory (220), and a processor (230) electrically connected to the communication unit (210) and the memory (220).

[0042] The configuration of the financial server (120) shown in FIG. 2 is according to one embodiment, and the components of the financial server (120) are not limited to the embodiment shown in FIG. 2, and some components may be added, changed, or deleted as needed.

[0043] According to one embodiment, the communication unit (210) may include at least one circuit capable of transmitting and receiving at least one signal, information, and data with at least one mobile terminal (110) and an application server (130).

[0044] Additionally, the communication unit (210) can be connected to at least one mobile terminal (110) via a wired or wireless connection to transmit and receive signals or data in real time. Through this, the financial server (120) can provide marketing content customized for the customer to the application server (130) based on the connection of the mobile terminal.

[0045] According to one embodiment, the communication unit (210) can receive various financial information through an application of the mobile terminal (110) and store the received information in the memory (220).

[0046] According to one embodiment, the memory (220) stores control logic or instructions that enable operation in conjunction with the mobile terminal (110) and the application server (130). The memory (220) may include volatile memory or non-volatile memory. For example, the memory (220) may store information, data, programs, applications, etc., necessary for the operation of the financial server (120). Accordingly, the processor (230) can perform the control operation described below by referring to the information stored in the memory (220).

[0047] According to one embodiment, the memory (220) may store various platforms. The memory (220) may include at least one type of storage medium among, for example, a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (e.g., SD or XD memory, etc.), RAM, ROM (EEPROM, etc.), and a USB memory. This memory (220) may store information and data obtained through the communication unit (210).

[0048] According to one embodiment, the memory (220) can store data about all customers, information about various financial products provided by the bank (e.g., K-Bank), information about preferred products, and marketing content about each product.

[0049] According to one embodiment, the memory (220) can store customer basic information, customer fund product holding information, customer internal fund product information, credit card payment history information, debit card payment history information, cash receipt processing history, and customer external fund product information for various customers. The memory (220) can store payment-related keywords for each customer extracted from customer basic information, customer fund product holding information, customer internal fund product information, credit card payment history information, debit card payment history information, cash receipt processing history, and customer external fund product information.

[0050] Additionally, the memory (220) can store fund product-specific characteristic information extracted from customer internal fund product information and customer external fund product information.

[0051] Additionally, the memory (220) can store an artificial intelligence algorithm. For example, the memory (220) can store an algorithm that provides customized marketing content through machine learning.

[0052] Additionally, memory (220) can store instructions or programs capable of performing the described operations.

[0053] According to one embodiment, the memory (220) can store MyData including deposit and withdrawal history of various banks, deposits, savings, IRP (Individual Retirement Pension), ISA (Individual Savings Account), credit loan amounts and product information held.

[0054] Such MyData may include investment product items and valuation amounts from securities firms, spending history and monthly billing amounts from credit card companies, and insurance premiums and the number of insurance products from insurance companies, and may also include telecommunications information.

[0055] In addition, MyData may include quantified information regarding the subscriber's MyData service activity history, such as MyData service login information, the number of registered institutions, and whether periodic transmission is requested.

[0056] According to one embodiment, the memory (220) can store an algorithm that exposes customized marketing content to the customer by reflecting it in the customer's financial transaction information.

[0057] According to one embodiment, the processor (230) may be implemented as at least one physical element among ASICs (application specific integrated circuits), DSPs (digital signal processors), DSPDs (digital signal processing devices), PLDs (programmable logic devices), FPGAs (field programmable gate arrays), processors, microcontrollers, and microprocessors.

[0058] According to one embodiment, the processor (230) may have an artificial intelligence algorithm embedded therein. The algorithm for such artificial intelligence may be implemented by the processor (230). The artificial intelligence may support a deep learning neural network algorithm that analyzes, recognizes, infers, and judges various data on its own as a program that mimics the neural network of the human brain.

[0059] According to one embodiment, the processor (230) can load commands or data received from the communication unit (210) into memory (220) for processing, and store the processed data back into memory (220). The processor (230) can control the components of the financial server (120) overall (e.g., communication unit (210), and memory (220)). Additionally, the processor (230) may include a financial transaction extraction unit (231), a classification unit (232), a learning unit (233), and a marketing provision unit (234).

[0060] According to one embodiment, the financial transaction extraction unit (231) can obtain financial transaction information (e.g., MyData) from each of the multiple financial institutions with which the customer transacts. The financial transaction information may include the customer's deposit and withdrawal transaction history, transaction history through open banking, credit card transaction history, and debit card transaction history obtained from each of the multiple financial institutions.

[0061] According to one embodiment, the classification unit (232) can classify the deposit and withdrawal transaction details among the acquired financial transaction information in a first stage based on the payment code. The classification unit (232) can classify the acquired financial transaction information in a first stage using a transaction summary classification code and a network classification code.

[0062] Additionally, the classification unit (232) can classify the first classified result a second time using a summary keyword. The classification unit (232) can classify the first classified result a second time using a summary value, a deposit passbook display content, and a withdrawal passbook display content.

[0063] Additionally, the classification unit (232) can classify check card payment details among the acquired financial transaction information into a third category using an industry code. The classification unit (232) can classify the acquired financial transaction information into a third category using a merchant industry code.

[0064] According to one embodiment, the classification unit (232) may perform a tokenization process to subdivide the text included in the acquired financial transaction information into word units, and may perform a stop word removal process to remove articles, prepositions, and particles from the text. The classification unit (232) may divide the text included in the financial transaction information into meaningful word units. Additionally, the classification unit (232) may remove articles, prepositions, particles, etc. from the text included in the financial transaction information to resolve the complexity of the model caused by words that do not have any particular meaning.

[0065] The classification unit (232) can perform additional refinement based on AI (Artificial Intelligence), and this additional refinement means further subdividing transaction history and automatically classifying it when new description values ​​are received.

[0066] According to one embodiment, the classification unit (232) may perform a normalization process to check spelling and adjust spacing in the text, and may perform an embedding process to convert the text into machine language and classify it by the same transaction. The classification unit (232) may check the spelling of the text included in the financial transaction information and adjust the spacing. In addition, the classification unit (232) may convert the text included in the financial transaction information into a vector, which is a sequence of characters (e.g., machine language numbers) that a machine can understand, and classify similar memo records into the same transaction.

[0067] According to one embodiment, the classification unit (232) can generate customized marketing content by learning the results classified through the tokenization process, stop word removal process, normalization process and embedding process through a Text CNN (Convolutional Neural Network).

[0068] Text CNNs can be used to classify sentences or words by analyzing local features of text data.

[0069] In this way, the classification unit (232) can perform a tokenization process that subdivides the text included in the acquired financial transaction information into words, and a stop word removal process that removes articles, prepositions, and particles from the text.

[0070] According to one embodiment, the learning unit (233) can train a model with the results obtained through the second classification and the third classification. The learning unit (233) can train the classification results to be customized for customers to generate customized marketing content.

[0071] According to one embodiment, the marketing provider (234) may provide customized marketing content generated using the model learned results to the customer's mobile terminal (110). The customized marketing content may be displayed through an application installed on the customer's mobile terminal (110).

[0072] FIG. 3 is a flowchart illustrating a process of providing customized marketing content to a customer using financial transaction information obtained from a plurality of financial institutions according to an embodiment of the present invention. FIG. 4(a) is an example diagram of transaction history according to an embodiment of the present invention. FIG. 4(b) is an example diagram of financial transaction information classified according to an embodiment of the present invention. FIG. 5(a) is an example diagram of financial transaction information classified based on code according to an embodiment of the present invention. FIG. 5(b) is an example diagram of financial transaction information classified based on summary keywords according to an embodiment of the present invention. FIG. 6 is an example diagram of check card payment history among financial transaction information classified by industry code according to an embodiment of the present invention. FIG. 7 is an example diagram showing data for generating customized marketing content by analyzing a customer's lifestyle pattern through model learning using financial transaction information according to an embodiment of the present invention. FIG. 8 is an example diagram of enhanced customized marketing content according to an embodiment of the present invention. FIG. 9 is an example diagram showing an application of a mobile terminal displaying customized marketing content according to one embodiment of the present invention.

[0073] Hereinafter, with reference to FIGS. 3 to 9, the process of providing customized marketing content to a customer using financial transaction information of a customer obtained from a plurality of financial institutions according to one embodiment of the present invention will be described in detail as follows.

[0074] According to one embodiment, a financial server (120) (e.g., processor (230)) can obtain financial transaction information from financial institutions with which a customer transacts (S310). The financial server (120) (e.g., processor (230)) can obtain financial transaction information regarding the customer from each of the multiple financial institutions with which the customer transacts. This financial transaction information may include the customer's deposit and withdrawal transaction history, transaction history through open banking, credit card transaction history, and debit card transaction history obtained from each of the multiple financial institutions.

[0075] Additionally, the financial server (120) (e.g., processor (230)) can obtain MyData including customer basic information, customer fund product holding information, customer internal fund product information, credit card payment history information, debit card payment history information, cash receipt processing history, and customer external fund product information.

[0076] Referring to Figure 4(a), financial transaction information may include information regarding the date, time, bank name, deposit / withdrawal type, summary description code, detailed description, and deposit / withdrawal amount.

[0077] Referring to Fig. 4(b), financial transaction information can be classified into 26 major categories and 47 medium categories through a first classification based on payment codes, a second classification based on summary keywords of the first classified results, and a third classification based on industry codes and model learning of check card payment history.

[0078] This financial transaction information can be classified based on payment codes, description keywords, industry codes, etc., and utilized to provide customized marketing content to the relevant customers. This marketing content can be used not only for customer expenditure analysis services, personalized financial transaction services, and recurring transaction notifications, but also to estimate the customer's lifestyle patterns or life stage.

[0079] According to one embodiment, a financial server (120) (e.g., processor (230)) can classify the acquired financial transaction information in a first manner through a code basis (S312). The financial server (120) (e.g., processor (230)) can classify the acquired financial transaction information in a first manner using a transaction summary classification code and a network classification code.

[0080] Referring to FIG. 5(a), a financial server (120) (e.g., processor (230)) can classify financial transaction information according to a description code and a network classification code by applying a code value as a rule to the financial transaction information. For example, if the description code is cash deposit and the network classification code is electronic finance, CMS, the transaction information can be classified as 1st financial_personal interbank account transfer.

[0081] According to one embodiment, the financial server (120) (e.g., processor (230)) can classify the first classified results into a second classification using keywords (S314). The financial server (120) (e.g., processor (230)) can classify the first classified results into a second classification using description values, deposit passbook display content, and withdrawal passbook display content.

[0082] Referring to FIG. 5(b), the financial server (120) (e.g., processor (230)) can classify financial transactions that are not classified in detail through code into text by referring to existing transaction history classifications. For example, if the detailed description is “DB Insurance”, “Hanwha Insurance 002 cases” and “Meritz”, the information classification can be classified as 1 Financial_Insurance.

[0083] According to one embodiment, a financial server (120) (e.g., processor (230)) can classify the acquired financial transaction information into a third category using an industry code (S316). The financial server (120) (e.g., processor (230)) can classify the acquired financial transaction information into a third category using a merchant industry code.

[0084] Referring to FIG. 6, a financial server (120) (e.g., processor (230)) can classify card transactions by merchant industry code. For example, the financial server (120) (e.g., processor (230)) can classify financial transactions based on the merchant industry code loaded when making a check card transaction with the bank (e.g., K-Bank). This classification is categorized into food, shopping, refueling, transportation, travel, beauty, medical health, education, culture, lifestyle, etc., according to the type of transaction (i.e., first level), and can be further classified into detailed items (i.e., second level) according to each type of transaction.

[0085] According to one embodiment, a financial server (120) (e.g., processor (230)) can train a model of the classified results (S318). The financial server (120) (e.g., processor (230)) can train a model of the data obtained by classifying financial transaction information from the first to the third level. The financial server (120) (e.g., processor (230)) can perform a tokenization process to subdivide the text included in the acquired financial transaction information into words, and can perform a stop word removal process to remove articles, prepositions, and particles from the text. Additionally, the financial server (120) can perform a normalization process to check spelling and adjust spacing in the text, perform an embedding process to convert the text into machine language and classify it by the same transaction, and generate customized marketing content by training the results classified through the tokenization process, the stop word removal process, the normalization process, and the embedding process using a Text CNN (Convolutional Neural Network).

[0086] That is, a financial server (120) (e.g., processor (230)) can generate customized marketing content by performing a preprocessing process (i.e., tokenization process, stop word removal process, normalization process and embedding process) on financial transaction information and learning the classified results through a Text CNN.

[0087] Referring to FIG. 7, a financial server (120) (e.g., processor (230)) can estimate a customer's lifestyle pattern or life stage from result data obtained by learning through Text CNN and generate a customized product for the customer.

[0088] That is, the financial server (120) (e.g., processor (230)) can subdivide the acquired result data by major life events of the customer. Additionally, the financial server (120) (e.g., processor (230)) can further subdivide the acquired result data by estimating the customer's lifestyle preference information. The financial server (120) (e.g., processor (230)) can determine (or generate) customized marketing content through these processes (i.e., the first advancement process).

[0089] Referring to FIG. 8, a financial server (120) (e.g., processor (230)) can perform a first enhancement process on the classified results obtained by performing a preprocessing process (i.e., tokenization, stop word removal, normalization, and embedding) on ​​financial transaction information, and then perform a second enhancement process to identify financial needs based on expenditure classification and provide customized financial transactions.

[0090] Through this second advanced process, the financial server (120) (e.g., processor (230)) can identify financial needs based on the customer's expenditure classification and provide customized marketing content to the customer. In addition, the financial server (120) (e.g., processor (230)) can provide customized financial services to the customer by providing customized messages to the customer.

[0091] According to one embodiment, a financial server (120) (e.g., processor (230)) can provide customized marketing content to a customer using the results of a model learned (S320). The financial server (120) (e.g., processor (230)) can provide the customized marketing content generated using the results of the model learned to the customer's mobile terminal (110) via a network (101). The customized marketing content can be displayed through an application installed on the customer's mobile terminal (110).

[0092] Alternatively, the financial server (120) (e.g., processor (230)) provides the generated customized marketing content to the application server (130), so that it can be displayed on the mobile terminal (110) as the customer runs the application through the mobile terminal (110).

[0093] Referring to FIG. 9, a mobile terminal (110) can display customized marketing content (911) through the screen (910) of an application. For example, when the date on which a customer periodically visits a mart is identified, a financial server (120) (e.g., processor (230)) generates customized marketing content regarding mart prices and provides it to the mobile terminal (110), thereby allowing the mobile terminal (110) to display customized marketing content (911) through the screen (910) of an application.

[0094] As described above, the present invention can provide customized marketing content to a customer's mobile terminal (110) by first classifying financial transaction information obtained from each of a plurality of financial institutions based on a payment code, secondly classifying the results of the first classification based on summary keywords, thirdly classifying check card payment history among the obtained financial transaction information based on an industry code, and model learning of the results obtained through the second classification and the third classification.

[0095] Each step in each of the flowcharts described above may be operated independently of the illustrated order or may be performed simultaneously. Additionally, at least one component of the present invention and at least one operation performed by said at least one component may be implemented in hardware and / or software.

[0096] Although the present invention has been described above with reference to the illustrated drawings, the present invention is not limited by the embodiments and drawings disclosed in this specification, and it is obvious that various modifications can be made by a person skilled in the art within the scope of the technical concept of the present invention. Furthermore, even if the effects of the configuration according to the present invention were not explicitly described while explaining the embodiments of the present invention above, it is natural to acknowledge that the effects predictable by said configuration should also be recognized. Explanation of the symbols

[0097] 100: System 110: Mobile terminal 120: Financial Server 130: Application Server 210: Communications Department 220: Memory 230: Processor

Claims

Claim 1 A financial server configured to provide customized marketing content to a customer using financial transaction information of a customer obtained from multiple financial institutions, comprising: a communication unit that transmits and receives data with at least one mobile terminal; a memory that stores financial transaction information of a customer; and a processor electrically connected to the communication unit and the memory, wherein the processor obtains financial transaction information from each of the multiple financial institutions with which the customer transacts, first classifies deposit and withdrawal transaction details among the obtained financial transaction information based on a payment code, secondly classifies the results of the first classification based on a summary keyword, thirdly classifies check card payment details among the obtained financial transaction information based on an industry code, model-trains the results obtained through the second classification and the third classification, and provides customized marketing content to the customer's mobile terminal using the results of the model-trained financial server. Claim 2 In claim 1, the financial transaction information comprises a financial server including a customer's deposit and withdrawal transaction history, open banking transaction history, credit card transaction history, and debit card transaction history obtained from each of the plurality of financial institutions. Claim 3 In claim 1, the processor is a financial server configured to first classify the acquired financial transaction information using a transaction summary classification code and a shared network classification code. Claim 4 In claim 3, the processor is a financial server configured to secondarily classify the first classified result using the summary value, deposit passbook display content and withdrawal passbook display content. Claim 5 In claim 1, the processor is a financial server configured to classify the acquired financial transaction information to a third level using a merchant industry code. Claim 6 A financial server configured to generate customized marketing content by the processor, wherein the processor performs a tokenization process to subdivide the text included in the acquired financial transaction information into word units, performs a stop word removal process to remove articles, prepositions, and particles from the text, performs a normalization process to check spelling and adjust spacing in the text, performs an embedding process to convert the text into machine language and classify it by the same transaction, and learns the results classified through the tokenization process, the stop word removal process, the normalization process, and the embedding process through a Text CNN (Convolutional Neural Network). Claim 7 In claim 1, the customized marketing content is a financial server displayed through an application installed on the customer's mobile terminal. Claim 8 A method performed on a financial server that provides customized marketing content to a customer using financial transaction information obtained from multiple financial institutions, comprising: a process of obtaining financial transaction information from each of the multiple financial institutions with which the customer transacts; a process of first classifying deposit and withdrawal transaction details among the obtained financial transaction information based on a payment code; a process of second classifying the results of the first classification based on a summary keyword; a process of third classifying check card payment details among the obtained financial transaction information based on an industry code; a process of model training the results obtained through the second classification and the third classification; and a process of providing customized marketing content to the customer's mobile terminal using the results of the model training. Claim 9 In claim 8, the process of model learning the acquired results comprises: a process of performing a tokenization process that subdivides the text included in the acquired financial transaction information into word units; a process of performing a stop word removal process that removes articles, prepositions, and particles from the text; a process of performing a normalization process that checks spelling and adjusts spacing in the text; a process of performing an embedding process that converts the text into machine language and classifies it by the same transaction; and a process of generating customized marketing content by learning the results classified through the tokenization process, the stop word removal process, the normalization process, and the embedding process through a Text CNN (Convolutional Neural Network). Claim 10 A system for providing customized marketing content to a customer using financial transaction information obtained from multiple financial institutions, comprising: a financial server that obtains financial transaction information from each of the multiple financial institutions with which the customer transacts, first classifies deposit and withdrawal transaction details among the obtained financial transaction information based on a payment code, secondly classifies the results of the first classification based on a summary keyword, thirdly classifies check card payment details among the obtained financial transaction information based on an industry code, and model-trains the results obtained through the second classification and the third classification, and provides customized marketing content to the customer's mobile terminal using the results of the model-trained results; and a mobile terminal that displays the customized marketing content provided from the financial server through an application.