KR102464995B1 - Method for credit evaluation based on end-to-end data generated on process of purchase, sales, inventory, logistics, distribution and calculation on ECS(e-commerce solution) and apparatus for performing the method - Google Patents

Abstract

The present invention relates to a method for credit evaluation based on end-to-end data generated on the process of purchase, sales, inventory, logistics, distribution and calculation on an e-commerce solution (ECS) and an apparatus for performing the method. The method for credit evaluation based on end-to-end data generated on the process of purchase, sales, inventory, logistics, distribution and calculation on an ECS comprises the steps of: collecting, by a credit evaluation basic data collection unit, credit evaluation basic data for credit evaluation of a seller; preprocessing, by a credit evaluation basic data preprocessing unit, the collected credit evaluation basic data; generating, by a credit evaluation basic data learning unit, an artificial intelligence engine for the credit evaluation of the seller based on the preprocessed credit evaluation basic data; and determining, by a credit evaluation unit, credit evaluation data of the seller based on the artificial intelligence engine. According to the present invention, a customized financial service can be provided to the seller.

Description

A credit evaluation method based on end-to-end data generated in the process of purchase, sale, inventory, logistics, distribution, and settlement of e-commerce solutions, and a device for performing such method. process of purchase, sales, inventory, logistics, distribution and calculation on ECS(e-commerce solution) and apparatus for performing the method}

The present invention relates to a credit evaluation method based on end-to-end data generated in the process of purchase, sale, inventory, logistics, distribution, and settlement of an e-commerce solution, and an apparatus for performing such a method. More particularly, it relates to a method and apparatus for performing a credit evaluation on a seller based on data generated on an e-commerce solution.

As we enter the intelligent information society triggered by the 4th industrial revolution, the infinite possibility of using data is causing changes in the data industry. With the advent of the data era, the level of the future data industry will determine the difference in competitiveness between countries.

In particular, the construction of big data infrastructure in the financial market is not only urgent, but also has become an important asset enough to influence the direction of the country's data industry in the near future. The financial big data infrastructure consists of big data open systems, data exchanges, and data specialized institutions.

It is necessary to study new financial products based on such big data-based user financial data. Various risk analysis is possible through artificial intelligence-based learning of user financial products, and new financial services that have not been available before can be provided to users based on risk analysis.

Therefore, it is necessary to study a specific method for utilizing the user's financial data and providing various financial services based on the user's financial data.

An object of the present invention is to solve all of the above problems.

In addition, the present invention aims to provide a customized financial service to the seller through the credit evaluation based on the end-to-end data generated during the purchase, sale, inventory, logistics, distribution, and settlement process of the e-commerce solution.

In addition, an object of the present invention is to perform filtering to remove redundancy and grouping in consideration of reliability for each data on current data and predicted data generated on the ECS to utilize them as credit evaluation data.

A representative configuration of the present invention for achieving the above object is as follows.

According to an embodiment of the present invention, the credit evaluation method based on the end-to-end data generated in the process of purchase, sale, inventory, logistics, distribution, and settlement of the e-commerce solution is the credit evaluation basic data collection unit for credit evaluation of the seller. Collecting credit evaluation basic data, the credit evaluation basic data pre-processing unit pre-processing the collected credit evaluation basic data, the credit evaluation of the seller based on the credit evaluation basic data pre-processed by the credit evaluation basic data learning unit It may include generating an artificial intelligence engine for It can be processed by classifying into three credit evaluation basic data (production), credit evaluation basic data (distribution), and credit evaluation basic data (sales).

On the other hand, the first type data is data on a product that is only produced or purchased by the seller and before distribution and sale, and is data composed of only the credit evaluation basic data (production), and the second type data is produced by the seller and data on products that have been distributed but have not been sold, and are data consisting of only the credit evaluation basic data (production) and the credit evaluation basic data (distribution), and the third type data is data on products produced, distributed and sold by the seller. As data for the credit evaluation, the credit evaluation basic data (production), the credit evaluation basic data (distribution), and the credit evaluation basic data (sales) are data, and the credit evaluation data is the first type data and the second type data. and the credit evaluation basic data based on the third type data.

Also, the credit evaluation data may be determined using the credit evaluation basic data based on the first type data, the second type data, and the third type data.

According to another embodiment of the present invention, an end-to-end data-based credit evaluation device generated in the process of purchase, sale, inventory, logistics, distribution, and settlement of an e-commerce solution collects basic credit evaluation data for credit evaluation of a seller A credit evaluation basic data collection unit configured to do so, a credit evaluation basic data preprocessor configured to pre-process the collected basic credit evaluation data, an artificial intelligence engine for the credit evaluation of the seller based on the pre-processed credit evaluation basic data It may include a credit evaluation basic data learning unit implemented to generate a credit evaluation unit and a credit evaluation unit implemented to determine the credit evaluation data of the seller based on the artificial intelligence engine, wherein the evaluation unit is configured to determine the credit evaluation data of the seller based on the artificial intelligence engine. The credit evaluation basic data may be classified and processed into three credit evaluation basic data (production), credit evaluation basic data (distribution), and credit evaluation basic data (sales).

On the other hand, the first type data is data on a product that is only produced or purchased by the seller and before distribution and sale, and is data composed of only the credit evaluation basic data (production), and the second type data is produced by the seller and data on products that have been distributed but have not been sold, and are data consisting of only the credit evaluation basic data (production) and the credit evaluation basic data (distribution), and the third type data is data on products produced, distributed and sold by the seller. As data for the credit evaluation, the credit evaluation basic data (production), the credit evaluation basic data (distribution), and the credit evaluation basic data (sales) are data, and the credit evaluation data is the first type data and the second type data. and the credit evaluation basic data based on the third type data.

Also, the credit evaluation data may be determined using the credit evaluation basic data based on the first type data, the second type data, and the third type data.

According to the present invention, customized financial services can be provided to sellers through credit evaluation based on end-to-end data generated during purchase, sale, inventory, logistics, distribution, and settlement of e-commerce solutions.

In addition, according to the present invention, a customized financial service can be provided to a seller through filtering to eliminate redundancy with respect to current data and predicted data generated on the ECS and grouping in consideration of reliability for each data.

1 is a conceptual diagram illustrating a credit evaluation apparatus according to an embodiment of the present invention.
2 is a conceptual diagram illustrating a sales management platform and a method of collecting basic credit evaluation data through the sales management platform according to an embodiment of the present invention.
3 is a conceptual diagram illustrating an operation of a credit basic data preprocessor according to an embodiment of the present invention.
4 is a conceptual diagram illustrating a first pre-processing according to an embodiment of the present invention.
5 is a conceptual diagram illustrating a first pre-processing (seller) according to an embodiment of the present invention.
6 is a conceptual diagram illustrating a first preprocessing according to an embodiment of the present invention.
7 is a conceptual diagram illustrating a first preprocessing according to an embodiment of the present invention.
8 is a conceptual diagram illustrating a method of performing a first preprocessing according to an embodiment of the present invention.
9 is a conceptual diagram illustrating an operation of a credit evaluation unit according to an embodiment of the present invention.
10 is a conceptual diagram illustrating a scorecard determination method according to an embodiment of the present invention.
11 is a conceptual diagram illustrating a scorecard determination method according to an embodiment of the present invention.
12 is a conceptual diagram illustrating a scorecard determination method according to an embodiment of the present invention.
13 is a conceptual diagram illustrating a credit rating apparatus for determining a scorecard according to an embodiment of the present invention.
14 is a conceptual diagram illustrating a seller credit evaluation method based on credit evaluation basic data generated in OMS according to an embodiment of the present invention.
15 is a conceptual diagram illustrating a method for setting a scale of credit evaluation basic data according to an embodiment of the present invention.
16 is a conceptual diagram illustrating a seller credit evaluation method based on product sales data generated in OMS according to an embodiment of the present invention.
17 is a conceptual diagram illustrating a seller credit evaluation method based on credit evaluation basic data generated in the WMS according to an embodiment of the present invention.
18 is a method for generating WMS-based credit evaluation basic data according to an embodiment of the present invention is disclosed.
19 is a method for generating WMS-based credit evaluation basic data according to an embodiment of the present invention is disclosed.
20 is a conceptual diagram illustrating a seller credit evaluation method based on product sales cycle data generated in the WMS according to an embodiment of the present invention.
21 is a conceptual diagram illustrating a seller credit evaluation method based on credit evaluation basic data generated in ECS according to an embodiment of the present invention.
22 is a conceptual diagram illustrating a seller credit evaluation method based on credit evaluation basic data generated in ECS according to an embodiment of the present invention.
23 is a conceptual diagram illustrating a seller credit evaluation method based on credit evaluation basic data generated in ECS according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [0012] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [0014] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS [0016] Reference is made to the accompanying drawings, which show by way of illustration specific embodiments in which the present invention may be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the present invention. It should be understood that the various embodiments of the present invention are different but need not be mutually exclusive. For example, certain shapes, structures, and characteristics described herein may be implemented with changes from one embodiment to another without departing from the spirit and scope of the present invention. In addition, it should be understood that the location or arrangement of individual components within each embodiment may be changed without departing from the spirit and scope of the present invention. Accordingly, the detailed description given below is not to be taken in a limiting sense, and the scope of the present invention should be taken as encompassing the scope of the claims and all equivalents thereto. In the drawings, like reference numerals refer to the same or similar elements throughout the various aspects.

Hereinafter, various preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings in order to enable those of ordinary skill in the art to easily practice the present invention.

1 is a conceptual diagram illustrating a credit evaluation apparatus according to an embodiment of the present invention.

In FIG. 1, a credit evaluation device for performing a credit evaluation on a seller based on data collected through an e-commerce sales and distribution management system platform is disclosed.

Referring to FIG. 1 , the credit evaluation apparatus includes a credit evaluation basic data collection unit 110 , a credit evaluation basic data preprocessing unit 120 , a credit evaluation basic data learning unit 130 , a credit evaluation unit 140 , and a financial service unit. 150 and a processor 160 .

The credit evaluation basic data collection unit 110 may be implemented to collect basic credit evaluation data for credit evaluation of the seller. Sellers can sell products through various platforms for selling and distributing products through e-commerce. The e-commerce sales and distribution management system platform for managing various data related to the seller's product sales, product distribution, and product payment may be expressed in terms of the sales management platform 100 . The credit evaluation basic data collection unit 110 may be implemented to collect basic credit evaluation data for credit evaluation of the seller through various sales management platforms. A specific sales management platform 100 will be described later. The term "product" used in the present invention may be used to include a service provided by a seller as a single product.

The credit evaluation basic data preprocessor 120 may be implemented to pre-process the collected basic credit evaluation data. Credit evaluation basic data may be pre-processed and used for learning of an artificial intelligence engine for credit evaluation, or may be used for credit evaluation of sellers.

Credit evaluation basic data for learning of the artificial intelligence engine for credit evaluation may be transmitted to the credit evaluation basic data learning unit 130 through a first pre-processing. Credit evaluation basic data for learning of the artificial intelligence engine for credit evaluation of the seller may be transmitted to the credit evaluation unit 140 through the second pre-processing.

The credit evaluation basic data learning unit 130 may be implemented for learning artificial intelligence for credit evaluation of the seller. The credit evaluation basic data learning unit 130 includes a plurality of artificial intelligence engines for credit evaluation of the seller, and each of the plurality of artificial intelligence engines may be implemented to determine a lower credit rating factor for the credit evaluation of the seller. .

The credit evaluation unit 140 may be implemented to evaluate the seller's credit to determine the seller's credit evaluation data. The credit evaluation unit 140 may determine the credit evaluation data of the seller based on a plurality of sub-credit evaluation factors determined by each of the plurality of artificial intelligence engines of the credit evaluation basic data learning unit. In addition, the credit evaluation unit 140 may determine the credit evaluation data of the seller based on a separate algorithm rather than the artificial intelligence engine.

The financial service unit 150 may be implemented to provide a financial service to a seller based on credit evaluation data of the seller.

The processor 160 operates the credit evaluation basic data collection unit 110 , the credit evaluation basic data preprocessor 120 , the credit evaluation basic data learning unit 130 , the credit evaluation unit 140 , and the financial service unit 150 . can be implemented to control

2 is a conceptual diagram illustrating a sales management platform and a method of collecting basic credit evaluation data through the sales management platform according to an embodiment of the present invention.

2 discloses a sales management platform and credit evaluation basic data collected from the sales management platform.

2, the sales management platform is an order management system (OMS) 210, enterprise resource planning (ERP) 220, warehouse management system (WMS) 230, E-commerce solution (ECS) 240 and the like. The OMS 210 , the ERP 220 , the WMS 230 , and the ECS 240 are examples, and various other entities related to the seller's product sales may be a sales management platform.

The OMS 210 is a platform for managing a product order of a seller.

The OMS 210 is a computerized system that can integrally process a series of sales process tasks by a seller who sells products through a plurality of sales channels. The seller may check the status of products ordered on a plurality of sales channels through the OMS 210 and collectively handle payment confirmation, delivery, order cancellation, return, and the like.

Specifically, functions such as product batch registration correction, order collection, invoice registration and transmission, and inventory management may be provided on the OMS 210 . In addition, the OMS 210 may provide a function for managing payment information, sales information, sales settlement information, return information, refund information due to return, inventory information, and the like on a plurality of sales channels.

The ERP 220 may be a sales management platform for managing information such as product production (purchase), logistics, finance, accounting, sales, purchase, and inventory of a seller as enterprise resource planning.

WMS 230 is a warehouse management system, a sales management platform for supporting and optimizing warehouse or delivery center management. The WMS 230 may integrate and manage the logistics processes such as warehousing, piling, inventory, packing, and shipment of the seller's goods as a whole.

The ECS 240 may be a sales management platform for creating and managing an online mall for sales of sellers. The ECS 240 may be implemented to create an online shopping mall, manage data generated on the online shopping mall, and perform marketing for selling products.

The credit evaluation basic data collection unit may collect basic credit evaluation data in connection with a sales management platform such as the OMS 210 , the ERP 220 , the WMS 230 , and the ECS 240 .

For example, the credit evaluation basic data collection unit uses product registration information, inventory information, order information, return information, payment information, sales information, settlement information, refund information, etc. generated in the OMS 210 as the seller's credit evaluation basic data. can be collected

The credit evaluation basic data collection unit may collect product wearing information, product inventory information, product shipment information, product delivery information, and the like generated by the WMS 230 as basic credit evaluation data of the seller.

The credit evaluation basic data collection unit may collect product marketing information and the like generated by the ECS 240 as the credit evaluation basic data of the seller.

3 is a conceptual diagram illustrating an operation of a credit basic data preprocessor according to an embodiment of the present invention.

3 discloses a method of pre-processing credit evaluation basic data in the credit basic data pre-processing unit.

Referring to FIG. 3 , the credit evaluation basic data 300 may be transmitted to the credit evaluation basic data learning unit 360 as the first pre-processing credit evaluation basic data 320 through the first pre-processing 310 . Also, the credit evaluation basic data 300 may be transmitted to the credit evaluation unit 370 as the second pre-processing credit evaluation basic data 355 through the second pre-processing 350 .

The first preprocessing 310 may be a preprocessing for learning in the artificial intelligence engine.

The first pre-processing 310 may be performed in consideration of the characteristics of the sales management platform that generated the credit evaluation basic data 300 . In the present invention, since the financial service is provided in consideration of the seller characteristics and the supply chain characteristics, the first preprocessing 310 may be performed for learning of the artificial intelligence engine in consideration of the seller characteristics and the supply chain characteristics.

The preprocessing in consideration of the supply chain characteristics among the first preprocessing 310 may be performed in consideration of the supply chain stage corresponding to the credit evaluation basic data 300 . Among the first preprocessing 310 , the preprocessing in consideration of the supply chain characteristics may be expressed in terms of the first preprocessing (supply chain) 313 .

For example, when the supply chain is divided into a production (or purchase) stage, a distribution stage, and a sales stage, the credit evaluation basic data 300 is the primary credit evaluation basic data (production stage) based on the stage at which the data is acquired. ), credit evaluation basic data (distribution stage), and credit evaluation basic data (sales stage) may be generated as the first pre-processing credit evaluation basic data 320 .

In addition, the preprocessing in consideration of the seller characteristics among the first preprocessing 310 may be performed through seller data classification and seller data augmentation based on the seller characteristics. Among the first pre-processing, the pre-processing in consideration of the seller characteristics may be expressed in terms of the first pre-processing (seller) 316 .

The second preprocessing 350 may be performed for credit evaluation of the seller based on the artificial intelligence engine included in the credit evaluation unit. The second pre-processed credit rating basic data 355 may be input into the artificial intelligence engine and used to determine lower credit rating factors. Accordingly, the second preprocessing 350 may be performed in consideration of the input data format of the artificial intelligence engine. Prediction of different credit evaluation basic data is performed for each AI engine, and each AI engine may have a different data format. The second preprocessing 350 may be performed in consideration of at least one artificial intelligence engine that may be used for credit evaluation of the seller.

4 is a conceptual diagram illustrating a first pre-processing according to an embodiment of the present invention.

In FIG. 4 , a first preprocessing (seller) and a first preprocessing (supply chain) applied to credit evaluation basic data are disclosed.

Referring to FIG. 4 , the credit evaluation basic data classified for each seller through the first pre-processing (seller) 400 is pre-processed for each supply chain step through the first pre-processing (supply chain) 450, and the first pre-processing credit evaluation basis data 490 may be generated.

The first pre-processing (supply chain) 450 may generate the first pre-processing credit evaluation basic data 490 in consideration of the data format transmitted from the sales management platform and the sales management platform, which are data transmission subjects.

Credit evaluation basic data through the first pre-processing (supply chain) 450 that considers information included in the supply chain stage managed by the sales management platform and the credit evaluation basic data generated by the sales management platform, the credit evaluation basic data (production ) 460, credit evaluation basic data (distribution) 470, and credit evaluation basic data (sales) 480 may be generated as the first pre-processing credit evaluation basic data.

In addition, the first pre-processing (supply chain) 450 may generate the first pre-processing credit evaluation basic data 490 by performing redundant data processing on the basic credit evaluation data transmitted through the sales management platform. When credit evaluation basic data for the same product is duplicated in a plurality of sales management platforms, duplicate data processing may be performed. For example, when a seller purchases a specific product for sale, the product may be registered on the OMS, and the product may be stored on the WMS. That is, the seller purchases a specific product once, but the data on product registration and product stocking due to such purchase is generated for each sales management platform, which may lead to duplication of credit evaluation basic data.

The first pre-processing (supply chain) 450 determines the redundancy of the transmitted credit evaluation basic data in consideration of the data generation time of the credit evaluation basic data, information included in the credit evaluation basic data, and the credit evaluation basic data information transmitted later to generate the first pre-processing credit evaluation basic data 490 . When duplication of credit evaluation basic data occurs, only data of one sales management platform is used through the first preprocessing (supply chain) 450, or duplicated credit evaluation basic data is filtered out and duplicated credit Only other credit evaluation basic data including information included in the evaluation basic data may be used.

Also, the first pre-processing (supply chain) 450 may be a pre-processing of the credit evaluation basic data considering time.

The credit rating of the seller and the basic data of the credit rating of the seller may change over time. Therefore, the time scale setting for the credit evaluation basic data for learning can greatly affect the performance of the artificial intelligence engine. Accordingly, in the present invention, after setting a time scale for the obtained basic credit evaluation data, the first pre-processing credit basic data 490 may be generated by preprocessing the basic credit evaluation data in consideration of the time scale. A time scale for preprocessing may be set for each credit evaluation basic data.

5 is a conceptual diagram illustrating a first pre-processing (seller) according to an embodiment of the present invention.

In FIG. 5, a method for augmenting credit evaluation basic data for learning of an artificial intelligence engine through a first preprocessing (seller) is disclosed. In particular, among the data augmentation methods, a method in which credit evaluation basic data is divided into lower credit evaluation basic data and used as learning data is disclosed.

Referring to FIG. 5 , a method of performing more accurate artificial intelligence engine learning by augmenting specific credit evaluation basic data 500 is disclosed. For example, the credit evaluation basic data 500 may be seasonality, transaction size, delivery cycle, sales trend, return rate, products sold, inventory asset size, operation information, and the like. Specific credit rating basic data among the credit rating basic data 500 may be augmented and generated as a plurality of lower rating basic data 540 .

When the return rate data is increased, data such as the return size, the discard rate, the return rate average, the return rate fluctuation stability, the number of times the return rate MAX is exceeded, etc. may be generated as the lower credit rating basic data 540 .

In an embodiment of the present invention, when data augmentation is required to perform a more accurate credit evaluation, learning may be performed by augmenting the basic credit evaluation data 500 into the lower credit evaluation basic data 540 through a first pre-processing. The data augmentation of the first preprocessing (seller) may be expressed in terms of the lower data augmentation 520 .

6 is a conceptual diagram illustrating a first preprocessing according to an embodiment of the present invention.

6 discloses a method for augmenting credit evaluation basic data for learning of an artificial intelligence engine in a first preprocessing (seller). In particular, a method of augmenting data by analyzing credit evaluation basic data on a time scale among data augmentation methods, and a method of augmenting data through a statistical method are disclosed.

6 (a) is a method of augmenting data by analyzing the credit evaluation basic data 600 on a time scale among data augmentation methods. Among the data augmentation methods, a method of augmenting data by analyzing credit evaluation basic data on a time scale may be expressed in terms of time scale data augmentation 610 .

For example, when the credit evaluation basic data 600 is a return rate, data on the number of sellers with a monthly return rate of 5% or more for 36 months based on a monthly return rate of 5% or more may be augmented and generated. As another example, based on the average monthly return rate, data on the number of sellers with an average return rate for 36 months may be augmented and generated.

6 (b) is a method of augmenting data by statistically analyzing the credit evaluation basic data 650 among the data augmentation methods. Among the data augmentation methods, a method of augmenting data by statistically analyzing the credit evaluation basic data 650 may be expressed in terms of statistical data augmentation 660 . For example, when the credit evaluation basic data 650 is a return rate, data augmentation may be performed by increasing it in multiple ways through a statistical method such as the average, standard deviation, maximum, or more than a specific section of the return rate for each customer.

7 is a conceptual diagram illustrating a first preprocessing according to an embodiment of the present invention.

7 discloses a method for augmenting credit evaluation basic data for learning of an artificial intelligence engine in a first pre-processing (seller). In particular, among the data augmentation methods, a method of augmenting data by analyzing basic data of credit evaluation in two dimensions is disclosed.

Referring to FIG. 7 , a method of augmenting the basic credit evaluation data 700 as two-dimensional data among the data augmentation methods is disclosed. Among the data augmentation methods, the basic credit evaluation data 700 is divided into a plurality of dimensions to augment the data. The method may be expressed in terms of multidimensional data augmentation 710 .

For example, when the credit evaluation basic data 700 is a return rate, the total return rate may be increased by dividing it into two-dimensional data. The first dimension may be the average return rate over 36 months, and the second dimension may be the number of months in which the return rate exceeds 5% in the 36 month period. Through this two-dimensional analysis, it may be possible to more accurately evaluate the credit evaluation basic data 700 of the seller.

8 is a conceptual diagram illustrating a method of performing a first preprocessing according to an embodiment of the present invention.

In FIG. 8, a method of selectively performing a first pre-processing (seller) for learning of a credit evaluation basic data learning unit is disclosed.

Referring to FIG. 8 , the first pre-processing (seller) may use sub data augmentation 810 , time scale data augmentation 820 , statistical data augmentation 830 , and multidimensional data augmentation 840 .

A first pre-processing (seller) may be selectively performed for learning of a plurality of artificial intelligence engines included in the credit evaluation basic data learning unit.

For example, the lower data augmentation 810 may be used for learning of an artificial intelligence engine for generating a specialized result through specific analysis of characteristic credit evaluation data among artificial intelligence engines. For example, sub data augmentation on the return rate may be performed for learning of an artificial intelligence engine that generates credit evaluation data with more weight on the return rate.

The time scale data augmentation 820 may be used for learning of an artificial intelligence engine for predicting changes in credit rating data over time.

Statistical data augmentation 830 may be used for learning of an artificial intelligence engine for separately presetting specific criteria and predicting credit rating data according to the set criteria.

The multidimensional data augmentation 840 may be used for learning of an artificial intelligence engine for predicting credit rating data based on set criteria for two dimensions.

In an embodiment of the present invention, the first preprocessing (seller) may be variously performed and various artificial intelligence models may be generated according to the properties of the predicted credit evaluation data.

9 is a conceptual diagram illustrating an operation of a credit evaluation unit according to an embodiment of the present invention.

In FIG. 9, a method of generating credit evaluation data of a seller based on an artificial intelligence engine in a credit evaluation unit is disclosed.

Referring to FIG. 9 , the credit evaluation unit may generate credit evaluation data through credit evaluation of a seller based on at least one artificial intelligence engine.

The credit evaluation unit may generate credit evaluation data of the seller based on one artificial intelligence engine, but the credit evaluation unit determines an artificial intelligence engine adaptively applicable to the seller based on the seller characteristic information 900 , and the determined artificial intelligence engine Credit rating data 950 may be generated based on the intelligence engine.

For example, the target artificial intelligence engine 920 for credit evaluation most suitable for the seller characteristic information 900 based on seller information such as the seller's products sold, the seller's product sales platform, the seller's sales, the seller's net profit, etc. can be decided.

The credit evaluation unit may determine the reliability of the artificial intelligence engine according to the seller characteristic information based on the feedback information to each of the plurality of artificial intelligence engines. In addition, the credit evaluation unit may determine a reliability level for each seller characteristic information for each of the plurality of artificial intelligence engines. Specifically, the seller characteristic information may be vectorized and expressed in space based on each of the sub-seller characteristic information, a seller group may be formed based on the seller characteristic information through distance information between spaces, and credit evaluation data for each seller group The reliability of each seller group of the artificial intelligence engine can be determined by comparing the financial service result data with the financial service. The reliability level may be determined in consideration of the statistical characteristics of the reliability of the seller group for each artificial intelligence engine.

The credit evaluation unit may generate credit evaluation data 950 for the seller by determining an artificial intelligence engine having a relatively high reliability level as the target artificial intelligence engine 920 based on the seller characteristic information.

10 is a conceptual diagram illustrating a scorecard determination method according to an embodiment of the present invention.

10 discloses a method of determining a scorecard for determining credit rating data of a merchant.

Referring to FIG. 10 , the scorecard 1000 may be a card including a plurality of credit rating basic data for determining credit rating data of a seller. A plurality of credit evaluation basic data included in the scorecard 1000 may be pre-processed and then input to the artificial intelligence engine 1020 to determine credit evaluation data.

For example, the plurality of credit evaluation basic data included in the scorecard 1000 is information such as whether a person is a resident, a period of residence in the residence, a job, a period of maintaining a job, a bank record, a card use record, an existing loan record, and the like. can

The scorecard 1000 includes a combination of a plurality of credit evaluation basic data, and may be generated in various types according to a combination of a plurality of credit evaluation basic data. A combination of a plurality of credit rating basic data included in the scorecard 1000 may be pre-processed and input to the artificial intelligence engine 1020 for determining the seller's credit rating data.

In the embodiment of the present invention, there may be a plurality of scorecards 1000 composed of a combination of a plurality of various credit evaluation basic data, and also the scorecard 1000 including the same basic credit evaluation data is included in learning and It is defined as a different scorecard 1000 according to a weight applied to each of the plurality of credit evaluation basic data, a scale applied to each of the plurality of basic credit evaluation data, etc. to generate the credit evaluation data and input to the artificial intelligence engine 1020 have.

That is, in the present invention, a plurality of scorecards 1000 may be input to different artificial intelligence engines 1020 and learned, and may be implemented to generate different credit evaluation data 1040 accordingly. A specific scorecard 1000 among the plurality of scorecards 1000 may be selectively used. In the present invention, in order to determine the scorecard 1000 with the highest accuracy, various scorecard determination methods may be used.

11 is a conceptual diagram illustrating a scorecard determination method according to an embodiment of the present invention.

Referring to FIG. 11 , in order to determine a scorecard that generates high-accuracy credit rating data, a plurality of credit rating basic data is variously combined, and a plurality of primary candidate scorecards including a plurality of different credit rating basic data. (1100) may be determined.

After the determination of the plurality of primary candidate scorecards 1100, the credit evaluation basic data corresponding to the plurality of primary candidate scorecards 1100 is input to each of the plurality of artificial intelligence engines, and each of the plurality of artificial intelligence engines is trained. This can be done.

A plurality of artificial intelligence engines learned based on the credit evaluation basic data corresponding to the plurality of primary candidate scorecards 1100 may be expressed in terms of the primary candidate artificial intelligence engine 1120 .

When the reliability of the primary candidate AI engines 1120 is equal to or greater than the threshold reliability, the corresponding AI engine may be determined as the secondary candidate AI engine 1160 .

A plurality of credit evaluation basic data included in the primary candidate scorecard 1100 corresponding to the secondary candidate artificial intelligence engine 1160 having a threshold reliability or higher is weighted, scaled, or generated time adjustment is performed to obtain secondary candidates Scorecard 1140 may be included.

The credit evaluation basic data corresponding to the secondary candidate scorecard 1140 is input to the secondary candidate artificial intelligence engine 1160 , and the secondary candidate artificial intelligence engine 1160 may be trained.

Among the secondary candidate artificial intelligence engines 1160 , at least one artificial intelligence engine having a reliability equal to or greater than a threshold reliability or an artificial intelligence engine having the highest reliability may be finally determined as an artificial intelligence engine to be used in the credit evaluation unit. In addition, the secondary candidate scorecard 1140 used in the finally determined artificial intelligence engine may be determined as the final scorecard to be used.

The above scorecard and artificial intelligence engine determination operation may be performed for each seller group in consideration of seller characteristic information, an artificial intelligence engine may be determined for each seller group, and a scorecard to be used for each seller group may be determined. That is, the primary candidate scorecard 1000 , the primary candidate AI engine 1120 , the secondary candidate scorecard 1140 , and the secondary candidate AI engine 1160 may be determined for each seller group.

12 is a conceptual diagram illustrating a scorecard determination method according to an embodiment of the present invention.

12 discloses a method of determining a secondary candidate scorecard by performing weight adjustment, scaling, or generation time adjustment.

Referring to FIG. 12 , the weight adjustment 1200 may be set in consideration of the importance of each of the credit evaluation basic data included in the scorecard. Learning about the artificial intelligence engine can be performed by setting a larger weight for credit evaluation basic data that has a greater impact on actual financial service results.

The scaling 1210 may be an adjustment of a range scale for classifying the credit rating basic data. For example, in the case of the number of years of work, it can be classified into n categories, and the credit evaluation data of the artificial intelligence engine can reflect the actual financial service results depending on what scale and classification of the basic data for credit evaluation is performed. may affect whether Accordingly, optimal AI engine learning may be performed through scaling 1210 for each of the plurality of credit evaluation basic data included in the secondary candidate scorecard 1240 .

The generation time adjustment 1220 (or scoring time adjustment) may perform learning of the artificial intelligence engine in consideration of the generation time (or scoring time) of the credit evaluation basic data. A plurality of credit evaluation basic data included in the scorecard may be grouped and input to the artificial intelligence engine. Accordingly, it may be determined which learning data is generated through the setting of the generation time of the credit evaluation basic data. Therefore, more accurate artificial intelligence engine learning can be performed through the adjustment 1220 of the generation time of the credit evaluation basic data.

Also, in the generation time adjustment 1220 , a score error generated according to the generation time may be reduced by setting a plurality of generation time points. The credit evaluation basic data classifies the observation period 1250, the scoring time point 1260, and the operation period 1270 separately, and sets the observation period 1250 and the operation period 1270 differently to set a plurality of scoring points 1260. can be set. By setting a plurality of scoring time points 1260, score errors according to generation time points that may be generated depending on the seller's product such as seasonal changes can be reduced, and the score change according to such time can be reflected.

13 is a conceptual diagram illustrating a credit rating apparatus for determining a scorecard according to an embodiment of the present invention.

Referring to FIG. 13 , the credit rating apparatus may include a scorecard determining unit.

A scorecard determining unit may be implemented to determine a scorecard. The scorecard determiner includes a first candidate scorecard determiner 1310 , a first candidate artificial intelligence engine generator 1320 , a second candidate scorecard determiner 1330 , a second candidate artificial intelligence engine generator 1340 , and A scorecard determination unit 1350 may be included.

The primary candidate scorecard determination unit 1310 variously combines a plurality of credit evaluation basic data to determine a scorecard for determining high-accuracy credit evaluation data, and includes a plurality of different credit evaluation basic data. A primary candidate scorecard may be determined.

The first candidate artificial intelligence engine generator 1320 may be implemented to generate the first candidate artificial intelligence engine. After determining the plurality of primary candidate scorecards, credit evaluation basic data corresponding to the plurality of primary candidate scorecards is input to each of the plurality of artificial intelligence engines, each of the plurality of artificial intelligence engines is trained, and the first A candidate artificial intelligence engine may be determined.

The secondary candidate AI engine generating unit 1340 may determine the secondary candidate AI engine as the secondary candidate AI engine when the reliability of the primary candidate AI engines is equal to or greater than the threshold reliability.

The secondary candidate scorecard generation unit 1330 performs weight adjustment, scale adjustment, or generation time adjustment on a plurality of credit evaluation basic data included in the primary candidate scorecard corresponding to the secondary candidate artificial intelligence engine to obtain the secondary candidate scorecard. Candidate scorecards can be determined.

The scorecard determining unit 1350 may be implemented to determine a scorecard to be finally used among secondary candidate scorecards.

14 is a conceptual diagram illustrating a seller credit evaluation method based on credit evaluation basic data generated in OMS according to an embodiment of the present invention.

14 discloses a method of performing credit evaluation of a seller by processing credit evaluation basic data generated on the OMS.

Referring to FIG. 14 , the credit evaluation basic data generated on the OMS may include product registration management data 1410 , product order management data 1420 , and product inventory management data 1430 .

The product registration management data 1410 is registered for sale in an online shopping mall such as online shopping mall data in which a product is registered, product registration data registered in the online shopping mall, product sales data (sales quantity, selling price, cost, etc.), product out of stock data, etc. It may include data related to the product to be used.

The product order management data 1420 may include data related to ordering and distribution of products after product registration, such as product order amount data for each period, product delivery data, product refund data, and the like.

The product inventory management data 1430 may include product inventory-related data, such as product inventory data, product out-of-stock data, product safety inventory data, and the like.

Product registration management data 1410, product order management data 1420, and product inventory management data 1430, respectively, are credit evaluation basic data (production) 1415, credit evaluation basic data (distribution) 1425, as described above. , may be pre-processed as credit evaluation basic data (sales) 1435 and generated as first pre-processed credit evaluation data.

Credit evaluation basic data generated by OMS can be utilized for learning of artificial intelligence engine. The credit evaluation basic data can be processed through weight adjustment, scale adjustment, or creation time adjustment and utilized for artificial intelligence engine learning.

The present invention discloses a method for setting a time scale for credit evaluation basic data. A different time scale may be applied to each of the product registration management data 1410 , the product order management data 1420 , and the product inventory management data 1430 .

A time scale for generating credit evaluation basic data based on the product registration management data 1410 may be expressed in terms of a product registration time scale. A time scale for generating credit evaluation basic data based on the product order management data 1420 may be expressed in terms of a product order time scale. A time scale for generating credit evaluation basic data based on the product inventory management data 1430 may be expressed in terms of a product inventory time scale. A method of setting each of the product registration time scale, the product order time scale, and the product stock time scale will be described later.

15 is a conceptual diagram illustrating a method for setting a scale of credit evaluation basic data according to an embodiment of the present invention.

15 discloses a method of setting a product registration time scale, a product order time scale, and a product inventory time scale.

15, the product registration time scale 1515, the product order time scale 1525, and the product inventory time scale 1535 are product registration management data 1510 for generating credit evaluation data of the seller, product order, respectively. It may be a collection period of each of the management data 1520 and the product inventory management data 1530 .

Hereinafter, for convenience of explanation, product registration management data 1510, product order management data 1520, and product inventory management data 1530 are expressed in terms of OMS data, product registration time scale 1515, product order time scale ( 1525) and product inventory time scale 1535 may be expressed in terms of OMS data time scale. The OMS data time scale may be a time unit of OMS data that is collected for a change in the seller's credit rating data.

A seller registers a product, sells a product, and delivers a product constitute one cycle, and the OMS data time scale may be determined based on a product distribution cycle of product registration, sale, and delivery.

A plurality of determination units may be set in consideration of each product and the number of each product, and a plurality of distribution cycle data for registration, sale, and delivery of the plurality of determination units may be collected. For example, if the seller sells products 1 to 3, product 1 has 10 judgment units, product 2 has 20 judgment units, and product 3 has 30 judgment units, registration of judgment units set for each product , sales, and distribution cycle data can be collected. The judgment unit of the initial product can be adaptively adjusted in consideration of the seller characteristics, credit evaluation data feedback, and product characteristics. The number of circulation cycle data can be adaptively adjusted to be collected.

The first OMS data time scale 1550 may be determined based on the first distribution cycle data for the product 1, the distribution cycle data for the product 2, and the third distribution cycle data for the product 3 .

The first OMS data time scale 1550 may be set to be a multiple of the first distribution cycle data, the second distribution cycle data, and the third distribution cycle data.

Thereafter, the first OMS data time scale 1550 may be changed to the second OMS data scale 1560 according to the feedback on the credit evaluation result. In consideration of the feedback on the credit evaluation result of the seller, the first OMS data time scale 1550 may be adjusted so that the seller's credit evaluation data may have high reliability, and the second OMS data time scale 1560 may be determined. That is, the first OMS data time scale 1550 may be updated to the second OMS data time scale 1560 in consideration of the feedback on the credit evaluation data of the seller.

The first OMS data time scale 1550 may be adjusted according to the change of the product and the change of the distribution cycle of the product, and the second OMS data time scale 1560 is the change of the first OMS data time scale 1550 and the seller's It may be adjusted according to credit rating data.

16 is a conceptual diagram illustrating a seller credit evaluation method based on product sales data generated in OMS according to an embodiment of the present invention.

16 discloses a method of processing product sales data to generate credit rating data for a seller.

Referring to FIG. 16 , although other data may be considered, a method of preprocessing product sales data 1600 , which may be data having the highest correlation with credit evaluation data of a seller, for learning is disclosed.

The product sales data 1600 may be converted into predicted product sales revenue data 1620 . The product sales data may be converted into predicted product sales revenue data 1620 for each period in consideration of information on the selling price and cost of the product, or a sales margin ratio.

The predicted product sales revenue data 1620 may be converted into data for each period in consideration of a predetermined sales period such as day/week/month, a seller financial service period, and the like. A first data reliability 1640 of the predicted product sales revenue data 1620 may be determined based on the predicted product sales revenue data 1620 for each period. The first data reliability 1640 is the reliability of the repeatability of the pattern, and may have a higher value as the probability of similar predicted product sales revenue data 1620 occurring thereafter increases.

For example, the predicted product sales revenue data 1620 may be separated based on n time sections of different lengths for increasing predictability in consideration of time, and the n first data reliability 1640 is the highest. A combination of time intervals may be defined as one prediction cycle. Thereafter, the predicted product sales revenue data 1620 for the next n time sections may be predicted in consideration of n different time sections corresponding to the prediction cycle.

The n time intervals for sellers with no seasonality and constant sales can be divided almost uniformly. In the case of a seller having high seasonality of a product or a relatively large change in product sales, n time intervals may be divided into different lengths in consideration of seasonality and product sales change.

In this way, one prediction cycle for future prediction of the forecast product sales revenue data 1620 is divided into n time intervals, and learning and prediction are performed taking the n time intervals into consideration, so that more accurate prediction of the seller's credit rating data This may be possible. In addition, in the present invention, the second data reliability 1660 of the predicted product sales revenue data 1620 may be determined by comparing it with the actual revenue. According to the second data reliability 1660 , the weight given by the predicted product sales revenue data 1620 in determining the credit evaluation data of the seller may be determined.

The predicted product sales revenue data 1620 based on the product sales data 1600 is determined in consideration of the aforementioned distribution cycle. That is, after the generation of the product sales data 1600 , when a period in consideration of the distribution cycle passes, product sales revenue data in a range similar to the predicted product sales revenue data 1620 should be generated.

If the error between the predicted product sales revenue data 1620 and the product sales revenue data 1600 is relatively large, it may be determined that the second data reliability 1660 is relatively low. Conversely, as the error between the predicted product sales revenue data 1620 and the product sales revenue data 1600 is relatively small, it may be determined that the second data reliability 1660 is relatively high. In consideration of the second data reliability 1660, the weight that the predicted product sales revenue data 1620 has in determining the credit evaluation data of the seller may be determined.

17 is a conceptual diagram illustrating a seller credit evaluation method based on credit evaluation basic data generated in the WMS according to an embodiment of the present invention.

17 discloses a method of performing credit evaluation of a seller by processing credit evaluation basic data generated on the WMS.

Referring to FIG. 17 , as described above, WMS is a sales management platform for supporting and optimizing warehouse or distribution center management as a warehouse management system. WMS can integrate and manage logistics processes such as warehousing, warehousing, inventory, packing, and delivery of the seller's goods.

The credit evaluation basic data generated on the WMS may include product stock management data 1710 , product shipment management data 1720 , and product inventory management data 1730 .

The product stocking management data 1710 is data generated when a product is stored in a warehouse. The seller may stock goods for sale in the distribution warehouse, and product stock management data may be generated when the goods are received.

The product release management data 1720 is data generated when a received product is released to a buyer due to sales.

The product inventory management data 1730 is data on the product remaining in the current distribution warehouse in consideration of the stocked product and the outgoing product.

Product stock management data 1710, product shipment management data 1720, and product inventory management data 1730, respectively, as described above, credit evaluation basic data (distribution) 1740, credit evaluation basic data (sales) (1750) may be pre-processed and generated as the first pre-processed credit evaluation data.

The credit evaluation basic data generated in the WMS can be utilized for the learning of the artificial intelligence engine. Hereinafter, a method of generating credit evaluation basic data by processing each of the product wearing management data 1710 , the product shipment management data 1720 , and the product inventory management data 1730 is disclosed.

18 is a method for generating WMS-based credit evaluation basic data according to an embodiment of the present invention is disclosed.

In FIG. 18, a method of generating credit evaluation basic data by processing each of product stock management data, product shipment management data, and product inventory management data is disclosed.

Referring to FIG. 18 , product sales cycle data 1815 and product sales rate data 1825 may be determined based on product stocking management data 1810 and product shipment management data 1820 . Also, product inventory ratio data 1840 and product value data 1850 may be determined based on the product inventory management data 1830 .

Each of the product sales cycle data 1815 , the product sales rate data 1825 , the product inventory ratio data 1840 , and the product value data 1850 may be utilized as basic data for credit evaluation of the seller.

The product sales cycle data 1815 relates to a product sales cycle, and may be a cycle from when a product is received in a warehouse to a time point when it is shipped out. A sales cycle for each of the plurality of products sold by the seller may be different from each other, and a sales cycle for each of the plurality of products may be determined.

The product sales rate data 1825 is data related to the stocked amount of the product and the shipped out amount of the product, and may be data on how much sales the product has compared to the stocked amount. Such product sales rate data 1825 may also be determined for each of a plurality of products.

The product value data 1850 is data on the characteristics of the seller's product, and may include data on characteristics of the product, such as product price range, product sales target, product liquidity, product inventory characteristics, and the like. The product value data 1850 may be generated based on data such as a product sales target, product cashability, and product inventory characteristics generated through a product characteristic database associated with an external server.

Data on the consumer group targeted by product, data on whether a product can be converted into cash and the rate of conversion to cash when a product is converted into cash, data on costs incurred in storing product inventory and the period for which it can be stored, product sales by hour (day, month, year) ), product return rate, and the like, may be stored in a database, and product value data 1850 may be generated in conjunction with the database.

19 is a method for generating WMS-based credit evaluation basic data according to an embodiment of the present invention is disclosed.

19 discloses a method of generating credit evaluation data in consideration of product sales cycle data and product sales rate data.

Referring to FIG. 19 , product sales cycle data 1910 and product sales rate data 1920 may generate credit evaluation basic data (distribution) 1950 .

Based on the supply chain, the product sales cycle data 1910 and the product sales rate data 1920 may generate credit evaluation basic data (distribution) 1950 .

In the present invention, different weights are set for each of the basic credit evaluation data (production) (1960), the basic credit evaluation data (distribution) (1950), and the basic credit evaluation data (sales) (1970), and credit evaluation data for the seller is created, and a loan to value ratio (LTV) for the seller may be determined.

Unlike the existing credit evaluation, the amount and interest rate that can be loaned to the user may be determined by additionally considering the product existing in the warehouse and the sales cycle of the product existing in the warehouse and the value of the product.

More specifically, based on the product sales cycle data 1910 , information on products sold from the time of loan to the maturity date of the loan may be predicted, and the loanable amount and interest rate may be determined until the maturity date of the loan. The product value data may determine the loanable amount and interest rate until the loan maturity date in consideration of the possibility of converting the product into cash.

20 is a conceptual diagram illustrating a seller credit evaluation method based on product sales cycle data generated in the WMS according to an embodiment of the present invention.

20 discloses a method of processing product sales cycle data to generate credit rating data for a seller.

Referring to FIG. 20 , product sales cycle data 2010 may be converted into predicted product sales cycle data 2020 . The predicted product sales cycle data 2020 may be converted into the predicted product sales revenue data 2030 for each period in consideration of information on the selling price and cost of the product, or a sales margin ratio.

The predicted product sales revenue data 2030 may be converted into data for each period in consideration of a predetermined sales period such as day/week/month, a seller financial service period, and the like. The first data reliability 2050 of the predicted product sales revenue data 2030 may be determined based on period-by-period data of the predicted product sales revenue data 2030 . The first data reliability 2050 is the reliability of the repeatability of the pattern, and may have a higher value as the probability of similar predicted product sales revenue data occurring thereafter increases.

For example, the predicted product sales revenue data may be divided into n time intervals of different lengths to increase predictability in consideration of time, and the combination of the n time intervals with the highest first data reliability 2050 is It may be defined as one prediction cycle. Thereafter, the predicted product sales revenue data 2030 for the next n time sections may be predicted in consideration of n different time sections corresponding to the prediction cycle.

The n time intervals for sellers with no seasonality and constant sales can be divided almost uniformly. In the case of a seller having high seasonality of a product or a relatively large change in product sales, n time intervals may be divided into different lengths in consideration of seasonality and product sales change.

In this way, one prediction cycle for future prediction of the forecast product sales revenue data 2030 is divided into n time intervals, and learning and prediction are performed in consideration of the n time intervals, so that a more accurate seller's credit evaluation data prediction is It may be possible.

In addition, in the present invention, the second data reliability 2060 of the predicted product sales revenue data 2030 may be determined by comparing it with the actual revenue. According to the second data reliability 2060, the weight of the predicted product sales revenue data 2030 in determining the credit evaluation data of the seller may be determined.

The predicted product sales revenue data 2030 based on the product sales cycle data 2010 is determined in consideration of the aforementioned distribution cycle. That is, after the generation of the product sales cycle data 2030 , when a period in consideration of the distribution cycle elapses, product sales revenue data in a range similar to the predicted product sales revenue data 2030 should be generated.

If the error between the predicted product sales revenue data 2030 and the product sales revenue data is relatively large, it may be determined that the second data reliability 2060 is relatively low. Conversely, as the error between the predicted product sales revenue data 2030 and the product sales revenue data is relatively small, it may be determined that the second data reliability 2060 is relatively high. In consideration of the second data reliability 2060, the weight that the predicted product sales revenue data 2030 has in determining the seller's credit evaluation data may be determined.

In order to classify the source of data, the predicted product sales revenue data based on the product sales cycle data (2010) is defined by the term predicted product sales revenue data (product sales cycle data) 2070, and based on the product sales data One predicted product sales revenue data may be defined by the term predicted product sales revenue data (product sales data) 2080 .

Each of the predicted product sales revenue data (product sales cycle data) 2070 and the predicted product sales revenue data (product sales data) 2080 considers each of the aforementioned second data reliability 2060, the predicted product sales revenue data is the seller A weight applied in determining the credit rating data of may be determined.

21 is a conceptual diagram illustrating a seller credit evaluation method based on credit evaluation basic data generated in ECS according to an embodiment of the present invention.

In FIG. 21, a seller credit evaluation method based on end-to-end data generated in the process of purchase, sale, inventory, logistics, distribution, and settlement of an e-commerce solution is disclosed.

Referring to FIG. 21 , the e-commerce solution may generate credit evaluation basic data based on data generated in the seller's sales process.

In the e-commerce solution, data generated during the entire process can be classified and processed into three basic data for credit evaluation (production), basic data for credit evaluation (distribution), and basic data for credit evaluation (sales).

When creating credit evaluation basic data (production), credit evaluation basic data (distribution), credit evaluation basic data (sales), basic credit evaluation data (production), credit evaluation basic data (distribution), credit evaluation basic data for individual products (Sales) is created. Accordingly, data filtering for removing duplicate data through product-specific data identification may be performed.

In addition, credit evaluation basic data (production), credit evaluation basic data (distribution), and credit evaluation basic data (sales) may be configured as one set and utilized for learning.

A specific product may generate a data format of (x, y, z) = {Credit evaluation basic data (production), credit evaluation basic data (distribution), credit evaluation basic data (sales)}.

The first type data 2110 may be data that is only produced (or purchased) by the seller, and has only credit evaluation basic data (production) as data about a product before distribution and sale. The first type data 2110 may be composed of {credit evaluation basic data (production), 0, 0}.

The second type data 2120 may be data on a product produced and distributed by the seller but before being sold, and may be data having only basic credit evaluation data (production) and basic credit evaluation data (distribution). The second type data 2120 may include {credit evaluation basic data (production), credit evaluation basic data (distribution), 0}.

The third type data 2130 is data on products produced, distributed, and sold by the seller, and may be data consisting of basic credit evaluation data (production), basic credit evaluation data (distribution), and basic credit evaluation data (sales). have. The third type data 2130 may include {credit evaluation basic data (production), credit evaluation basic data (distribution), and credit evaluation basic data (sales)}.

In the present invention, credit evaluation basic data based on the first type data 2110, the second type data 2120, and the third type data 2130 is generated, and the credit evaluation data for determining the financial services available to the seller can be created.

As the first type data 2110, the second type data 2120, and the third type data 2130 go, it is data that can have a greater influence on the sales (or net profit) of the seller. Accordingly, the first type data 2110 , the second type data 2120 , and the third type data 2130 may be utilized to determine the seller's credit evaluation data with a higher weight.

In the present invention, based on credit evaluation data combining the first type data 2110, the second type data 2120, and the third type data 2130 on the time axis in consideration of the financial service period provided to the seller, it is possible to provide to the seller. A financial service (loan amount and loan interest rate) may be determined.

22 is a conceptual diagram illustrating a seller credit evaluation method based on credit evaluation basic data generated in ECS according to an embodiment of the present invention.

22 shows a method of determining a financial service (loan amount and loan interest rate) that can be provided to a seller based on credit evaluation data combining the first type data, the second type data, and the third type data on the time axis in consideration of the financial service period This is initiated.

Referring to FIG. 22 , first type data, second type data, and third type data of the current time may be determined according to the loan period of the loan period.

The first type data, the second type data, and the third type data of the current time based on the loan start time of the seller are the first type data (current) 2210 and the second type data (current) 2220 as current data. and third type data (current) 2230 .

Based on the loan termination time, the first type data (current) 2210 and the second type data (current) 2220 may be changed into other type data (second type data or third type data). Also, new first type data, second type data, and third type data may be generated based on the loan termination time.

The first type data, the second type data, and the third type data accumulated and predicted based on the loan termination time are prediction data, including first type data (prediction) 2260, second type data (prediction) 2270 and It may be expressed in terms of third type data (prediction) 2280 . Prediction data may be generated through prediction based on product sales cycle data and distribution cycle data for current data.

In the present invention, the first type data (current) 2210, the second type data (current) 2220, and the third type data (current) 2230, which are current data based on the loan start time, and the loan end time The first type data (prediction) 2260, the second type data (prediction) 2270, and the third type data (prediction) 2280, which are prediction data accumulated and predicted as a reference, are utilized as credit evaluation data of the seller and the seller financial services may be provided.

Current data, first type data (current) 2210, second type data (current) 2220, and third type data (current) 2230, and prediction data, first type data (prediction) 2260, The second type data (prediction) 2270 and the third type data (prediction) 2280 may be used as credit evaluation data by filtering to remove redundancy and grouping in consideration of reliability for each data.

23 is a conceptual diagram illustrating a seller credit evaluation method based on credit evaluation basic data generated in ECS according to an embodiment of the present invention.

23 discloses a filtering and grouping method for generating credit rating data.

Referring to FIG. 23 , first type data (current), second type data (current), and third type data (current), first type data (prediction), second type data (prediction), and third type data Filtering for (prediction) is performed, and duplicate data can be removed.

Filtering according to an embodiment of the present invention may be a process of removing data changed from current data to prediction data.

Among the first type data (current), second type data (current) and third type data (current), second type data (prediction) and third type data (prediction) and third type data ( Prediction) can be filtered out as redundant data and removed from the current data.

The filtered first type data (current), second type data (current), and third type data (current) and first type data (prediction), second type data (prediction) and third type data (prediction) are It can be divided into three data groups based on the loan start time and loan end time.

The first data group 2310 may include filtered and remaining current data. That is, the first type data (current), the second type data (current), and the third type data (current) that are not changed into the prediction data and remain filtered may be included in the first data group 2310 .

The second data group 2320 may include prediction data (first type data (prediction), second type data (prediction), and third type data (prediction)) changed from current data to prediction data. That is, the second data group 2320 may include prediction data predicted by changing current data.

The third data group 2330 includes prediction data (first type data (prediction), second type data (prediction), and third type data (prediction)) that is not based on current data but is directly generated as prediction data. can do. That is, the third data group may include prediction data generated without current data.

In the present invention, credit evaluation data for the seller may be determined by sequentially giving low weights in the order of the first data group 2310, the second data group 2320, and the third data group 2330. That is, the credit evaluation data for the seller may be determined by setting a relatively high weight for the data based on the current data.

The embodiments according to the present invention described above may be implemented in the form of program instructions that can be executed through various computer components and recorded in a computer-readable recording medium. The computer-readable recording medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the computer-readable recording medium may be specially designed and configured for the present invention, or may be known and available to those skilled in the art of computer software. Examples of the computer-readable recording medium include hard disks, magnetic media such as floppy disks and magnetic tapes, optical recording media such as CD-ROMs and DVDs, and magneto-optical media such as floppy disks. medium), and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. A hardware device may be converted into one or more software modules to perform processing in accordance with the present invention, and vice versa.

In the above, the present invention has been described with reference to specific matters such as specific components and limited embodiments and drawings, but these are provided to help a more general understanding of the present invention, and the present invention is not limited to the above embodiments, Those of ordinary skill in the art to which the invention pertains can make various modifications and changes from these descriptions.

Therefore, the spirit of the present invention should not be limited to the above-described embodiments, and the scope of the spirit of the present invention is not limited to the scope of the scope of the present invention. will be said to belong to

Claims (6)

The credit evaluation method based on end-to-end data that occurs in the process of purchasing, selling, inventory, logistics, distribution, and settlement of e-commerce solutions is:
collecting, by the credit evaluation basic data collection unit, basic credit evaluation data for credit evaluation of the seller;
pre-processing the collected credit evaluation basic data by a credit evaluation basic data pre-processing unit;
generating an artificial intelligence engine for the credit evaluation of the seller based on the credit evaluation basic data pre-processed by the credit evaluation basic data learning unit; and
Comprising the step of a credit rating unit determining the credit rating data of the seller based on the artificial intelligence engine,
In the e-commerce solution, the basic credit evaluation data generated in the entire process is classified and processed into three basic credit evaluation data (production), basic credit evaluation data (distribution), and basic credit evaluation data (sales),
The first type data is data on a product that is only produced or purchased by the seller and before distribution and sale, and is data consisting of only the credit evaluation basic data (production),
The second type data is data on a product produced and distributed by the seller but before it is sold, and is data consisting of only the credit evaluation basic data (production) and the credit evaluation basic data (distribution),
The third type data is data on products produced, distributed and sold by the seller, and is data consisting of the credit evaluation basic data (production), the credit evaluation basic data (distribution), and the credit evaluation basic data (sales), ,
the credit rating data is determined using the credit rating basic data based on the first type data, the second type data and the third type data;
the credit rating data is determined using the credit rating basic data based on the first type data, the second type data and the third type data;
Each of the first type data, the second type data, and the third type data is current data based on a loan start time, first type data (current), second type data (current), and third type data ( It includes first type data (prediction), second type data (prediction) and third type data (prediction), which are prediction data accumulated and predicted based on the current) and the loan termination time,
The first type data (current), the second type data (current) and the third type data (current) as the current data, and the first type data (prediction), the second type data (prediction) as the prediction data ( Prediction) and the third type data (prediction) are filtered to remove redundancy, and a grouped data group in consideration of reliability for each data is used as the credit evaluation data. According to claim 1,
The filtering is performed through the prediction in consideration of product sales cycle data and distribution cycle data among the first type data (current), the second type data (current), and the third type data (current), and the second type data (prediction) ) and the data that is changed to the third type data (prediction) is filtered as redundant data and removed from the current data,
the data group includes a first data group, a second data group and a third data group;
The first data group includes the filtered and remaining first type data (current), the second type data (current), and the third type data (current),
The second data group includes first type data (prediction), the second type data (prediction), and the third type data (prediction), which are prediction data predicted by changing the current data,
The third data group includes the first type data (prediction), the second type data (prediction), and the third type data (prediction), which are prediction data directly generated from prediction data without being based on the current data. including,
The credit evaluation data is a method characterized in that the first data group, the second data group, and the third data group is determined by sequentially giving a low weight in the order. 3. The method of claim 2,
The credit evaluation basic data includes product registration management data, product order management data, and product inventory management data,
The product stocking management data and the product shipment management data determine product sales cycle data and product sales rate data,
The product inventory management data determines product inventory rate data and product value data,
The product sales cycle data is converted into predicted product sales cycle data,
The predicted product sales cycle data is converted into forecast product sales revenue data for each period in consideration of information on the sales price and cost of the product or sales margin,
The forecast cycle of the forecast product sales revenue data is determined based on the first data reliability of the data for each period,
The first data reliability is the reliability of the repeatability of the pattern, and has a relatively high value as the probability of similar predicted product sales revenue data occurring thereafter is relatively high,
The prediction cycle is a combination of n time intervals having the highest reliability of the first data,
The n number of time sections are divided into different lengths based on seasonality of the product and changes in product sales. An end-to-end data-based credit evaluation device that occurs in the process of purchasing, selling, inventory, logistics, distribution, and settlement of e-commerce solutions,
a credit evaluation basic data collection unit configured to collect basic credit evaluation data for credit evaluation of the seller;
a credit evaluation basic data pre-processing unit implemented to pre-process the collected basic credit evaluation data;
a credit evaluation basic data learning unit implemented to generate an artificial intelligence engine for the credit evaluation of the seller based on the preprocessed credit evaluation basic data; and
Comprising a credit evaluation unit implemented to determine the credit evaluation data of the seller based on the artificial intelligence engine, the evaluation unit,
In the e-commerce solution, the basic credit evaluation data generated in the entire process is classified and processed into three basic credit evaluation data (production), basic credit evaluation data (distribution), and basic credit evaluation data (sales),
The first type data is data on a product that is only produced or purchased by the seller and before distribution and sale, and is data consisting of only the credit evaluation basic data (production),
The second type data is data on a product produced and distributed by the seller but before it is sold, and is data consisting of only the credit evaluation basic data (production) and the credit evaluation basic data (distribution),
The third type data is data on products produced, distributed and sold by the seller, and is data consisting of the credit evaluation basic data (production), the credit evaluation basic data (distribution), and the credit evaluation basic data (sales), ,
the credit rating data is determined using the credit rating basic data based on the first type data, the second type data and the third type data;
the credit rating data is determined using the credit rating basic data based on the first type data, the second type data and the third type data;
Each of the first type data, the second type data, and the third type data is current data based on a loan start time, first type data (current), second type data (current), and third type data ( It includes first type data (prediction), second type data (prediction) and third type data (prediction), which are prediction data accumulated and predicted based on the current) and the loan termination time,
The first type data (current), the second type data (current) and the third type data (current) as the current data, and the first type data (prediction), the second type data (prediction) as the prediction data ( Prediction) and the third type data (prediction) are filtered to remove redundancy, and a grouped data group in consideration of reliability for each data is utilized as the credit evaluation data. 5. The method of claim 4,
The filtering is performed through the prediction in consideration of product sales cycle data and distribution cycle data among the first type data (current), the second type data (current), and the third type data (current), and the second type data (prediction) ) and the data that is changed to the third type data (prediction) is filtered as redundant data and removed from the current data,
the data group includes a first data group, a second data group and a third data group;
The first data group includes the filtered and remaining first type data (current), the second type data (current), and the third type data (current),
The second data group includes first type data (prediction), the second type data (prediction), and the third type data (prediction), which are prediction data predicted by changing the current data,
The third data group includes the first type data (prediction), the second type data (prediction), and the third type data (prediction), which are prediction data directly generated from prediction data without being based on the current data. including,
The credit rating data is a credit rating apparatus, characterized in that the first data group, the second data group, and the third data group is determined by sequentially assigning a low weight in the order. 6. The method of claim 5,
The credit evaluation basic data includes product registration management data, product order management data, and product inventory management data,
The product stocking management data and the product shipment management data determine product sales cycle data and product sales rate data,
The product inventory management data determines product inventory rate data and product value data,
The product sales cycle data is converted into predicted product sales cycle data,
The predicted product sales cycle data is converted into forecast product sales revenue data for each period in consideration of information on the sales price and cost of the product or sales margin,
The forecast cycle of the forecast product sales revenue data is determined based on the first data reliability of the data for each period,
The first data reliability is the reliability of the repeatability of the pattern, and has a relatively high value as the probability of similar predicted product sales revenue data occurring thereafter is relatively high,
The prediction cycle is a combination of n time intervals having the highest reliability of the first data,
The n time sections are credit rating apparatus, characterized in that divided into different lengths based on seasonality of the product and changes in product sales.

Images