Method, electronic apparatus and recording medium for predicting sales volume of product in sales platform

Abstract

The present disclosure relates to a method for predicting sales volume on a sales platform, an electronic device, and a recording medium. A method according to one embodiment of the present disclosure may include: a step of determining a total sales volume prediction value of a target item corresponding to a specific period based on a first sales volume prediction algorithm that predicts sales volume based on past sales volume of a target item; a step of determining a total sales volume prediction value of an item group corresponding to a specific period based on a second sales volume prediction algorithm that predicts sales volume based on past sales volume of an item group including a target item; a step of determining a correction coefficient based on the total sales volume prediction value of an item group corresponding to a specific period; a step of determining a corrected total sales volume prediction value of a target item by applying a correction coefficient to the total sales volume prediction value of the target item; and a step of determining a daily sales volume prediction value of a target item corresponding to a specific date included in the specific period by applying a date-based division coefficient to the corrected total sales volume prediction value of the target item.

Description

Technology Field

[0001] The present disclosure relates to a technology for predicting the sales volume of products on a sales platform. Background Technology

[0002] With the recent proliferation of e-commerce, the importance of sales forecasting across various sales platforms is increasing. Sales forecasting plays a crucial role in various business functions, such as supply chain management, inventory optimization, and the formulation of marketing strategies. However, existing sales forecasting algorithms have struggled to simultaneously improve the forecasting accuracy of specific items as well as the overall total. Therefore, there was a need to develop a forecasting algorithm capable of achieving high accuracy for both specific items and the entire product range. The problem to be solved

[0003] The present disclosure provides a technology for predicting the sales volume of products on a sales platform. means of solving the problem

[0004] A method for predicting sales volume on a sales platform, performed by an electronic device according to one embodiment of the present disclosure, may include: a step of determining a total sales volume prediction value of said target item corresponding to a specific period based on a first sales volume prediction algorithm that predicts sales volume based on past sales volume of said target item; a step of determining a total sales volume prediction value of said item group corresponding to the specific period based on a second sales volume prediction algorithm that predicts sales volume based on past sales volume of said item group including said target item; a step of determining a correction coefficient based on the total sales volume prediction value of said item group corresponding to the specific period; a step of determining a corrected total sales volume prediction value of said target item by applying the correction coefficient to the total sales volume prediction value of said target item; and a step of determining a daily sales volume prediction value of said target item corresponding to a specific date included in said specific period by applying a date-based division coefficient to the corrected total sales volume prediction value of said target item.

[0005] In one embodiment, the item group including the target item may be a set of all items sold on the sales platform.

[0006] In one embodiment, the step of determining the correction coefficient may include: determining the total sales volume prediction value of each item included in the item group corresponding to the specific period based on the first sales volume prediction algorithm; and determining the correction coefficient based on the sum of the total sales volume prediction values ​​of each item and the total sales volume prediction value of the item group.

[0007] In one embodiment, the correction factor may be determined based on the value obtained by dividing the total sales volume prediction value of the item group by the sum of the total sales volume prediction values ​​of each of the items.

[0008] In one embodiment, the second sales volume prediction algorithm may include logic for determining a ratio value corresponding to each date based on the attributes of each date from the current day to a specific future date.

[0009] In one embodiment, the attribute may be based on at least one of the day of the week, the date, and the date of a specific event.

[0010] In one embodiment, the ratio value may be determined by an artificial neural network-based learning model that has been trained based on Light GBM (Light Gradient Boosting Machine).

[0011] In one embodiment, the date-by-date division factor may be determined based on the ratio value of each date included in the specific period.

[0012] In one embodiment, the daily division factor may be proportional to the ratio value of each of the dates included in the specific period.

[0013] In one embodiment, the second sales volume prediction algorithm may include logic for determining a predicted total sales volume of the item group for a specific period based on the total sales volume of the item group from a specific past date to the present, the sum of the ratio values ​​of each day from the specific past date to the present, and the sum of the ratio values ​​of each day of the specific period.

[0014] In one embodiment, the total sales volume prediction value of the item group for a specific period determined by the second sales volume prediction algorithm may be determined based on the value obtained by dividing the total sales volume of the item group from the past specific date to the present by the sum of the ratio values ​​of each day from the past specific date to the present and multiplying it by the sum of the ratio values ​​of each day of the specific period.

[0015] In one embodiment, the corrected total sales volume prediction value of the target item may be determined based on the value obtained by multiplying the total sales volume prediction value of the target item by the correction coefficient.

[0016] In one embodiment, the daily sales volume forecast value of the target item corresponding to a date included in the specific period may be determined based on the value obtained by dividing the corrected total sales volume forecast value of the target item by the sum of the daily division coefficients for each date included in the specific period and multiplying it by the daily division coefficient corresponding to the specific date.

[0017] An electronic device according to one embodiment of the present disclosure comprises one or more processors; and one or more memories in which instructions to be executed by the one or more processors are stored, and when the instructions are executed, the one or more processors may be configured to execute a method according to the present disclosure.

[0018] A non-transient computer-readable recording medium according to one embodiment of the present disclosure records instructions to be executed by one or more processors, and said instructions may be configured to cause said one or more processors to perform a method according to the present disclosure when said instructions are executed. Effects of the invention

[0019] According to one embodiment of the present disclosure, both the prediction accuracy of a specific item and the prediction accuracy of all items can be increased.

[0020] The effects according to the technical concept of the present disclosure are not limited to those mentioned above, and other unmentioned effects will be clearly understood by a person skilled in the art from the description in the specification. Brief explanation of the drawing

[0021] FIG. 1 illustrates an environment in which an electronic device according to one embodiment of the present disclosure can be applied. FIG. 2 is a block diagram of an electronic device according to one embodiment of the present disclosure. FIG. 3 is a flowchart illustrating a method according to one embodiment of the present disclosure. FIG. 4 is a drawing showing ratio values ​​corresponding to each date determined by an electronic device according to one embodiment of the present disclosure. FIG. 5 is a diagram showing an electronic device according to one embodiment of the present disclosure determining a proportional value corresponding to a specific period. FIG. 6 is a drawing showing a proportional value corresponding to a specific period determined by an electronic device according to one embodiment of the present disclosure. FIG. 7 is a diagram showing an electronic device according to one embodiment of the present disclosure determining a correction factor. FIG. 8 is a diagram showing an electronic device according to one embodiment of the present disclosure determining a daily sales volume forecast value of a target item. Specific details for implementing the invention

[0022] The various embodiments described in this disclosure are illustrative for the purpose of clearly explaining the technical concept of this disclosure and are not intended to limit it to specific embodiments. The technical concept of this disclosure includes various modifications, equivalents, alternatives, and embodiments selectively combined from all or part of each embodiment described in this disclosure. Furthermore, the scope of the technical concept of this disclosure is not limited to the various embodiments presented below or the specific descriptions thereof.

[0023] Terms used in this disclosure, including technical or scientific terms, may have the meaning generally understood by those skilled in the art to which this disclosure pertains, unless otherwise defined.

[0024] Expressions used in this disclosure, such as “comprising,” “may compose,” “possessing,” “possessing,” “having,” and “possessing,” mean that the subject feature (e.g., function, operation, or component, etc.) exists and do not exclude the existence of other additional features. That is, such expressions should be understood as open-ended terms implying the possibility of including other embodiments.

[0025] Singular expressions used in this disclosure may include the meaning of the plural form unless otherwise indicated by the context, and this applies likewise to singular expressions described in the claims.

[0026] Expressions such as "first," "second," or "first," "second," etc., used in this disclosure are used to distinguish one object from another when referring to a plurality of objects of the same kind, unless otherwise indicated by the context, and do not limit the order or importance of said objects.

[0027] Expressions used in the present disclosure, such as “A, B, and C,” “A, B, or C,” “A, B, and / or C,” or “at least one of A, B, and C,” “at least one of A, B, or C,” “at least one of A, B, and / or C,” “at least one selected from A, B, and C,” “at least one selected from A, B, or C,” or “at least one selected from A, B, and / or C,” may mean each of the listed items or all possible combinations of the listed items. For example, “at least one selected from A and B” may refer to (1) A, (2) at least one of A, (3) B, (4) at least one of B, (5) at least one of A and at least one of B, (6) at least one of A and B, (7) at least one of B and A, and (8) all of A and B.

[0028] The expression “based on” as used in this disclosure is used to describe one or more factors affecting an act or action of a decision or judgment described in the phrase or sentence containing this expression, and this expression does not exclude additional factors affecting said act or action of a decision or judgment.

[0029] As used in the present disclosure, the expression that a certain component (e.g., a first component) is "connected" or "connected" to another component (e.g., a second component) may mean that the certain component is not only directly connected or connected to the other component, but is also connected or connected through a new other component (e.g., a third component).

[0030] As used in this disclosure, the expression "configured to" may have meanings such as "set to," "capable of," "modified to," "made to," or "capable of." This expression is not limited to the meaning of "specifically designed in hardware." For example, a processor configured to perform a specific operation may mean a generic-purpose processor capable of performing that specific operation by executing software, or a special-purpose computer structured through programming to perform that specific operation.

[0031] Hereinafter, various embodiments described in this disclosure will be explained with reference to the attached drawings. In the attached drawings and the description thereof, identical or substantially equivalent components may be given the same reference numerals. Furthermore, in the description of the various embodiments below, the description of identical or corresponding components may be omitted, but this does not mean that such components are not included in the embodiments.

[0032] FIG. 1 illustrates an environment in which an electronic device (110) according to one embodiment of the present disclosure may be applied. The electronic device (110) and the database (120) are connected via a network and can communicate with each other. Here, communication between the electronic device (110) and the database (120) may mean that the electronic device (110) communicates with a database management system (DBMS) that manages the database (120). That is, the electronic device (110) can access the database (120) via a network.

[0033] The electronic device (110) may be a device for predicting the sales volume of a product on a sales platform. The sales platform may include, for example, a sales platform of an e-commerce service that sells one or more products online. The electronic device (110) may include, for example, a server device that manages the sales platform or an electronic device that manages inventory on the sales platform. As another example, the electronic device (110) may be implemented as a terminal capable of transmitting and receiving various information to and from a database (120) via a network. For example, the electronic device (110) may be one of a computer, a laptop, a portable communication terminal (smartphone, etc.), a portable multimedia device, a wearable device, or an HMD. However, the type of the electronic device (110) is not limited thereto, and the electronic device (110) may be any device capable of receiving information from a user or outputting information to a user, and communicating with a database (120) or other devices via a network. The electronic device (110) can utilize information stored in the database (120) to predict how much of the product will be sold.

[0034] The electronic device (110) can provide various information regarding sales volume forecasting to the user and can receive input regarding sales volume forecasting from the user. Specifically, the electronic device (110) can obtain a command regarding sales volume forecasting from the user through an input / output interface, perform sales volume forecasting in response to the obtained command, and output the result of the sales volume forecasting to the user, store it within the electronic device (110), or transmit it to another device. The input obtained from the user may include various forms of input, such as clicking using a mouse, touching using a touch pad or touch screen, voice recognition, other electronic inputs, or voice input through a microphone. The output of the electronic device (110) may include various forms of output, such as a visual form through a display, projector, hologram, etc., or an auditory form through a speaker, etc.

[0035] The database (120) can store information that serves as the basis for predicting the sales volume of a product on a sales platform. For example, the database (120) can store sales logs for each product handled on the sales platform. Additionally, for example, the sales logs for each product in the database (120) may be updated in real time, or may be updated repeatedly at a periodic or non-periodic rate. The electronic device (110) can obtain information regarding the daily sales volume of any product through the database (120). For example, it can obtain information regarding the daily sales volume of a specific product, all products handled on the sales platform, or products satisfying specific conditions among all products handled on the sales platform. Furthermore, the electronic device (110) can obtain information regarding the sales volume of any product by time of day, day, week, month, quarter, year, or specific period through the database (120).

[0036] The network can serve to connect the electronic device (110) with the database (120) or other external devices. For example, the network can provide a connection path so that the electronic device (110) can be connected to the database (120) and transmit and receive packet data with the database (120). The network can be implemented as any kind of wired or wireless network, such as, for example, a Local Area Network (LAN), a Wide Area Network (WAN), a Mobile Radio Communication Network, or a Wibro (Wireless Broadband Internet).

[0037] In an environment where an electronic device according to another embodiment of the present disclosure may be applied and in a form not illustrated, the electronic device (110) may include a database. In this case, information stored in the database can be obtained without going through the network illustrated in FIG. 1. In this case, the electronic device (110) may also include a database management system (DBMS) that manages the database.

[0038] FIG. 2 is a block diagram of an electronic device (110) according to one embodiment of the present disclosure. The electronic device (110) can process information regarding the prediction of sales volume of a product. In one embodiment, the electronic device (110) may include one or more processors (210), one or more memories (220), and a communication circuit (230) as components. In one embodiment, at least one of the components of the electronic device (110) may be omitted, or another component may be added to the electronic device (110). In one embodiment, additionally or alternatively, some components may be implemented as an integrated unit or as a singular or plural entity. In the present disclosure, one or more processors (210) may be referred to as processors (210). Unless the context clearly indicates otherwise, the expression processors (210) may mean a set of one or more processors. In the present disclosure, one or more memories (220) may be referred to as memories (220). The expression "memory" (220) may mean a set of one or more memories unless the context clearly indicates otherwise. In one embodiment, at least some of the components inside and outside the electronic device (110) may be connected to each other via a bus, GPIO (General Purpose Input / Output), SPI (Serial Peripheral Interface), or MIPI (Mobile Industry Processor Interface), etc., to exchange information (data, signals, etc.).

[0039] The processor (210) can control at least one component of an electronic device (110) connected to the processor (210) by running software (e.g., instructions, programs, etc.). Additionally, the processor (210) can perform various operations related to the present disclosure, such as computation, processing, data generation, and processing. Furthermore, the processor (210) can load data, etc. from memory (220) or store it in memory (220). Moreover, the processor (210) can transmit and receive various information with the database (120) through a communication circuit (230). In one embodiment, the processor (210) can control the communication circuit (230) to request various information regarding sales volume prediction from the database (120) and receive various information regarding sales volume prediction from the database (120).

[0040] The memory (220) can store various information (data). The information stored in the memory (220) is information acquired, processed, or used by at least one component of the electronic device (110), and may include software (e.g., instructions, programs, etc.). The memory (220) may include volatile and / or non-volatile memory. In the present disclosure, instructions or programs are software stored in the memory (220) and may include an operating system for controlling the resources of the electronic device (110), an application, and / or middleware that provides various functions to the application so that the application can utilize the resources of the electronic device (110). In one embodiment, the memory (220) may store instructions that cause the processor (210) to perform calculations when executed by the processor (210). The memory (220) may store at least a portion of information received from a database through a communication circuit (230) and / or information transmitted to a database through a communication circuit (230). Specifically, the memory (220) can store information regarding the sales volume prediction of a product (item) and instructions executed by the processor (210).

[0041] A communication circuit (230) can perform wireless or wired communication between an electronic device (110) and a database or other external electronic device. For example, the communication circuit (230) can perform wireless communication according to methods such as eMBB (enhanced Mobile Broadband), URLLC (Ultra Reliable Low-Latency Communications), MMTC (Massive Machine Type Communications), LTE (Long-Term Evolution), LTE-A (LTE Advance), NR (New Radio), UMTS (Universal Mobile Telecommunications System), GSM (Global System for Mobile communications), CDMA (Code Division Multiple Access), WCDMA (Wideband CDMA), WiBro (Wireless Broadband), WiFi (Wireless Fidelity), Bluetooth, NFC (Near Field Communication), GPS (Global Positioning System), or GNSS (Global Navigation Satellite System). For example, the communication circuit (230) can perform wired communication according to methods such as USB (Universal Serial Bus), HDMI (High Definition Multimedia Interface), RS-232 (Recommended Standard-232), or POTS (Plain Old Telephone Service). In one embodiment, the electronic device (110) may be implemented by integrating it with another device. In this case, the communication circuit (230) may function as a connection circuit or interface connecting the electronic device (110) and the other device.

[0042] Hereinafter, the operation described as being performed by the electronic device (110) in FIGS. 3 to 8 can be understood as being performed by the processor (210) of the electronic device (110) described in FIG. 2.

[0043] FIG. 3 is a flowchart illustrating a method (300) according to an embodiment of the present disclosure. A method (300) for predicting sales volume on a sales platform may include: a step (S310) of determining a total sales volume prediction value of a target item corresponding to a specific period based on a first sales volume prediction algorithm that predicts sales volume based on past sales volume of a target item; a step (S320) of determining a total sales volume prediction value of an item group corresponding to a specific period based on a second sales volume prediction algorithm that predicts sales volume based on past sales volume of an item group including a target item; a step (S330) of determining a correction coefficient based on the total sales volume prediction value of an item group corresponding to a specific period; a step (S340) of determining a corrected total sales volume prediction value of a target item by applying a correction coefficient to the total sales volume prediction value of the target item; and a step (S350) of determining a daily sales volume prediction value of a target item corresponding to a specific date included in a specific period by applying a date-based division coefficient to the corrected total sales volume prediction value of the target item.

[0044] The electronic device (110) can determine a total sales volume prediction value of a target item corresponding to a specific period based on a first sales volume prediction algorithm that predicts sales volume based on the past sales volume of the target item. Here, the total sales volume prediction value of the target item corresponding to a specific period may refer to the sales volume of the target item predicted to be sold throughout the specific period. Here, the first sales volume prediction algorithm may be an algorithm that determines a total sales volume prediction value of the target item corresponding to a specific period based on the past sales volume of the target item, but not based on the past sales volume of a separate item group including the target item. For example, past sales volume may include past daily sales volume. The first sales volume prediction algorithm may be an algorithm that predicts the sales volume of the target item on a specific date or for a specific period based on past daily sales volume information of the target item. For example, the first sales volume prediction algorithm may be any known sales volume prediction algorithm based on the past sales volume of the target item.

[0045] The sales volume prediction value determined by the first sales volume prediction algorithm may provide relatively high prediction accuracy for the sales volume of the item in question. However, the sales volume prediction value for all items sold on the sales platform or the sales volume prediction value for a specific item group, which is the sum of the sales volume prediction values ​​for each item determined by this algorithm, may be inaccurate. Therefore, it is necessary to provide a more accurate sales volume prediction for the target item and / or a more accurate sales volume prediction value for all items or a specific item group through the correction of the sales volume prediction value. To this end, the present disclosure proposes a correction based on the past sales volume of an item group including the target item. In one embodiment, the item group including the target item may be a set of all items sold on the sales platform. Alternatively, the item group including the target item may be, for example, a set of one or more items belonging to the same category as the target item.

[0046] The electronic device (110) can determine the predicted total sales volume of an item group corresponding to a specific period based on a second sales volume prediction algorithm that predicts sales volume based on the past sales volume of an item group including a target item. The past sales volume of an item group may be the sum of the past sales volumes of each item included in the item group. The past sales volume of an item group may include the past daily sales volume of the entire item group. The second sales volume prediction algorithm can determine the predicted total sales volume of an item group corresponding to a specific period by analyzing the past sales volume trends of all items included in the item group.

[0047] In one embodiment, the second sales volume prediction algorithm may include logic for determining a ratio value corresponding to each date based on the attributes of each date from the current day to a specific future date. The ratio value corresponding to each date will be explained in detail later through FIG. 4.

[0048] In one embodiment, the second sales volume prediction algorithm may include logic for determining a predicted total sales volume of an item group for a specific period based on the total sales volume of an item group from a specific past date to the present, the sum of ratio values ​​for each day from a specific past date to the present, and the sum of ratio values ​​for each day of a specific period. In one embodiment, the predicted total sales volume of an item group for a specific period determined by the second sales volume prediction algorithm may be determined based on the value obtained by dividing the total sales volume of an item group from a specific past date to the present by the sum of ratio values ​​for each day from a specific past date to the present and multiplying by the sum of ratio values ​​for each day of a specific period. For example, the predicted total sales volume of an item group for a specific period determined by the second sales volume prediction algorithm may be determined by the value obtained by dividing the total sales volume of an item group from a specific past date to the present by the sum of ratio values ​​for each day from a specific past date to the present and multiplying by the sum of ratio values ​​for each day of a specific period, or by applying additional adjustments to this value.

[0049] The electronic device (110) can determine a correction coefficient based on the total sales volume prediction value of an item group corresponding to a specific period. In one embodiment, the electronic device (110) can determine the total sales volume prediction value of each item included in the item group corresponding to a specific period based on a first sales volume prediction algorithm. Additionally, the electronic device (110) can determine a correction coefficient based on the sum of the total sales volume prediction values ​​of each item included in the item group corresponding to a specific period determined based on the first sales volume prediction algorithm and the total sales volume prediction value of the item group corresponding to a specific period determined based on a second sales volume prediction algorithm. In one embodiment, the electronic device (110) can determine a correction coefficient based on the value obtained by dividing the total sales volume prediction value of the item group corresponding to a specific period determined based on the second sales volume prediction algorithm by the sum of the total sales volume prediction values ​​of each item included in the item group corresponding to a specific period determined based on the first sales volume prediction algorithm. For example, the electronic device (110) may determine a correction factor by dividing the total sales volume prediction value of an item group corresponding to a specific period determined based on a second sales volume prediction algorithm by the sum of the total sales volume prediction values ​​of each item included in an item group corresponding to a specific period determined based on a first sales volume prediction algorithm, or by applying additional adjustments to this value. That is, the correction factor may be determined corresponding to a specific period.

[0050] In one embodiment, additional adjustments to the correction factor may be applied according to the attributes of a specific period. For example, the attributes of a specific period may include whether the specific period is the beginning of the month. Here, the beginning of the month may mean a period that falls within the range from the 1st day of the month to a predetermined specific day. That is, the electronic device (110) may determine whether the specific period is the beginning of the month based on whether the specific period falls within the range from the 1st day of the month to a predetermined specific day. Subsequently, if the specific period is the beginning of the month, the electronic device (110) may perform an adjustment to apply a lower limit of the correction factor. Specifically, if the specific period is the beginning of the month, if the determined correction factor is 1 or greater, the determined correction factor may be used as is, and if it is less than 1, the determined correction factor may be adjusted to 1 or a preset value. By making such adjustments, the sales volume at the beginning of the month may not be underestimated.

[0051] In one embodiment, the electronic device (110) may determine a corrected total sales volume prediction value of a target item corresponding to a specific period based on a value obtained by multiplying a correction factor by a value obtained by multiplying a total sales volume prediction value of a target item corresponding to a specific period. For example, the corrected total sales volume prediction value of a target item corresponding to a specific period may be determined by multiplying a correction factor by a value obtained by multiplying a total sales volume prediction value of a target item corresponding to a specific period, or by applying additional adjustments to this value.

[0052] The electronic device (110) can determine the daily sales volume prediction value of a target item corresponding to a specific date included in a specific period by applying a daily division factor to the corrected total sales volume prediction value of a target item corresponding to a specific period. In one embodiment, the electronic device (110) can determine the daily sales volume prediction value of a target item corresponding to a specific period based on the value obtained by dividing the corrected total sales volume prediction value of a target item corresponding to a specific period by the sum of the daily division factors for each date included in the specific period and multiplying it by the daily division factor corresponding to the specific date.

[0053] FIG. 4 is a diagram showing ratio values ​​corresponding to each date determined by an electronic device (110) according to one embodiment of the present disclosure. In one embodiment, the electronic device (110) may determine ratio values ​​corresponding to each date based on the attributes of each date from the current date to a specific future date. Referring to a table (400) showing ratio values ​​corresponding to each date, ratio values ​​corresponding to each date in the period from 2023-09-18 to 2023-10-29 determined by the electronic device (110) are shown when a prediction is executed on each of the d-7 date (2023-09-25) and the d-7 date (2023-09-18). For example, the specific future date may be 34 days after the execution date. Specifically, if a prediction is executed on 2023-09-25, ratio values ​​corresponding to each day of the period from the execution date, 2023-09-25, to 2023-10-29, which is 34 days after the execution date (a total of 35 days) can be determined. Additionally, if a prediction is executed on 2023-09-18, ratio values ​​corresponding to each day of the period from the execution date, 2023-09-18, to 2023-10-22, which is 34 days after the execution date (a total of 35 days) can be determined.

[0054] In one embodiment, the electronic device (110) can determine a ratio value corresponding to each date based on an attribute based on at least one of a day of the week, a date, and a specific event date, corresponding to each date from the current date to a specific future date. That is, the attribute of each date in the period from the current date to a specific future date can be based on at least one of a day of the week, a date, and a specific event date. In one embodiment, the ratio value can be determined by an artificial neural network-based learning model that has been trained based on a Light Gradient Boosting Machine (Light GBM).

[0055] Specifically, a learning model based on a previously trained artificial neural network may be a learning model that takes an attribute based on at least one of a day of the week, a date, and a specific event date as an input value and takes a ratio value as an output value. Herein, an attribute based on at least one of a day of the week, a date, and a specific event date comprises: information on which day of the week the date is; information on which month the date is; information on which month the date is of the year; information on whether the date is a specific event date; information on how many days before / after the date is relative to the specific event date; whether the date is a public holiday; whether the date is a substitute public holiday; whether the day before the date is a public holiday or a weekend; whether the day before the date is a working day with a public holiday or a weekend; whether the date is the day before the first working day of the month; whether the date is the first working day of the month; whether the date is a Monday or Tuesday and simultaneously a public holiday or a weekend; whether the date is a Monday or Tuesday and simultaneously not the first day of the month and simultaneously a public holiday or a weekend; whether the date is a Monday of the first week of the month and simultaneously a public holiday or a weekend; whether the date is a Monday of the week not the first week of the month and simultaneously a public holiday or a weekend; whether the date is a Monday and simultaneously a public holiday or a weekend; and whether the date is a Monday and simultaneously a public holiday or a weekend. It may include at least one of whether the date is the first working day of the week and is not a Monday, whether the date is the first working day of the week and is a Monday, whether the date is the first working day of the week, whether the date is the first Monday of the month and is not a Monday and is not the first day of the month, whether the date is the first working day of the month and is not a Monday and is not the first day of the month, and whether the date is the first working day of the month and is not a Monday and is not the first day of the month.Here, specific event dates may include specific public holidays such as Lunar New Year, Chuseok, Buddha's Birthday, and Christmas, which are officially designated as days off every year. Alternatively, specific event dates may include days recognized by the public as special occasions, such as Halloween, Valentine's Day, and White Day. Furthermore, when training an artificial neural network-based learning model, the model can be trained so that ratio values ​​are based on sales volume and the sum of the ratio values ​​for each day within the period from the current day to a specific future date equals 1. Through this training, ratio values ​​can be derived that reflect attributes based on various variables. In other words, for example, while there may be a trend of high daily sales volume on Mondays, this trend may not be reflected on Mondays that are also public holidays; however, according to the aforementioned training, ratio values ​​that reflect this tendency can be derived. Accordingly, the sales volume prediction value determined based on ratio values ​​can be more accurate than conventional values.

[0056] That is, the training dataset may be a dataset of proportional ratio values ​​of attributes and sales volume for each day included in a specific period (a period from the current day to a specific future date, for example, a period of 35 days). Here, sales volume may refer to the daily sales volume of an item group. Accordingly, the ratio value for each day may be a value that reflects the daily sales volume trend of the item group for that day (i.e., proportional to the expected daily sales volume). Additionally, the determined ratio value may be stored in memory (220). The electronic device (110) may also utilize the ratio values ​​for each day of the period from 2023-09-18 to 2023-09-24, which were determined on 2023-09-18 and stored in memory (220), in calculations executed on 2023-09-25.

[0057] FIG. 5 is a diagram showing that an electronic device (110) according to one embodiment of the present disclosure determines a proportional value corresponding to a specific period. According to the example of the illustrated table (500), there may be five specific periods designated as fw1, fw2, fw3, fw4, and fw5. For example, fw1 may be a period from day d to day d+6 (6 days after d), fw2 may be a period from day d+7 to day d+13, fw3 may be a period from day d+14 to day d+20, fw4 may be a period from day d+21 to day d+27, and fw5 may be a period from day d+28 to day d+34. That is, each specific period may be a period of 7 days.

[0058] The electronic device (110) can determine the ratio of the sum of the ratio values ​​of each specific period (fw1, fw2, fw3, fw4, fw5) to the sum of the ratio values ​​of each specific period (pw1) from d as a proportional value corresponding to each specific period. Specifically, on d-7, the electronic device (110) can determine the ratio value of each day in the period from d-7 to d+27, and on d, the electronic device (110) can determine the ratio value of each day in the period from d-7 to d+27. For example, the electronic device (110) may utilize the ratio values ​​of each day in the period from day d-7 to day d+27 determined on day d-7 to calculate the ratio value of each day in the period (fw1, fw2, fw3, fw4) relative to the sum of the ratio values ​​of each day in the period from day d-7 to day d+27 determined on day d-7. As another example, the electronic device (110) may utilize the ratio values ​​of each day in the period from day d-7 to day d-1 determined on day d-7 and the ratio values ​​of each day in the period from day d to day d+27 determined on day d to calculate the ratio value of each day in the period (fw1, fw2, fw3, fw4) relative to the sum of the ratio values ​​of each day in the period from day d-7 to day d-1 determined on day d to calculate the ratio value of each day in the period from day d to day d+27-7 to day d+27 determined on day d to calculate the ratio value of each day in the period from day d to day d+27 determined on day d.However, in order to calculate the proportional value of the sum of the ratio values ​​of each day included in the last specific period (fw5) relative to the sum of the ratio values ​​of each day included in the past specific period (pw1) from d, the electronic device (110) may use the ratio values ​​of each day of the period from d-7 to d-1 determined on d-7 and the ratio values ​​of each day of the period from d+28 to d+24 determined on d (since the ratio values ​​of each day of the period from d+28 to d+24 are not determined on d-7). In one embodiment, the electronic device (110) may calculate a proportional value of the sum of the ratio values ​​of each day included in the last specific period (fw5) relative to the sum of the ratio values ​​of each day included in the past specific period (pw1) determined on d-7, by multiplying the sum of the ratio values ​​of each day in the period from d-7 to d+6 determined on d-7 (fw1 / pw1) by the sum of the ratio values ​​of each day in the period from d+28 to d+34 determined on d determined on d divided by the sum of the ratio values ​​of each day in the period from d to d+6 determined on d determined on d (fw5 / fw1). Through these various calculations, it is possible to enable a more accurate sales volume forecast to be implemented in the present disclosure.

[0059] FIG. 6 is a diagram showing proportional values ​​corresponding to specific periods determined by an electronic device (110) according to one embodiment of the present disclosure. For example, a proportional value (over1_fw1, over1_fw2, over1_fw3, over1_fw4, over1_fw5) corresponding to each specific period (fw1, fw2, fw3, fw4, fw5) (i.e., the sum of the ratio values ​​of each day of each specific period (fw1, fw2, fw3, f4, f5) relative to the sum of the ratio values ​​of each day of each specific period (fw1, fw2, fw3, f4, f5)) is displayed in a table (600). For convenience of explanation, the ratio values ​​of each day of the period from day d-7 to day d+27 determined on day d-7 may be denoted as r1_-7 to r1_27, and the ratio values ​​of each day of the period from day d to day d+35 may be denoted as r2_0 to r1_34. In this case, over1_fw1 can be calculated as {(r1_0) + (r1_1) + (r1_2) + (r1_3) + (r1_4) + (r1_5) + (r1_6)} / {(r1_-7) + (r1_-6) + (r1_-5) + (r1_-4) + (r1_-3) + (r1_-2) + (r1_-1)}. Additionally, over1_fw2 can be calculated as {(r1_7) + (r1_8) + (r1_9) + (r1_10) + (r1_11) + (r1_12) + (r1_13)} / {(r1_-7) + (r1_-6) + (r1_-5) + (r1_-4) + (r1_-3) + (r1_-2) + (r1_-1)}. Additionally, over1_fw3 can be calculated as {(r1_14) + (r1_15) + (r1_16) + (r1_17) + (r1_18) + (r1_19) + (r1_20)} / {(r1_-7) + (r1_-6) + (r1_-5) + (r1_-4) + (r1_-3) + (r1_-2) + (r1_-1)}.Additionally, over1_fw4 can be calculated as {(r1_21) + (r1_22) + (r1_23) + (r1_24) + (r1_25) + (r1_26) + (r1_27)} / {(r1_-7) + (r1_-6) + (r1_-5) + (r1_-4) + (r1_-3) + (r1_-2) + (r1_-1)}. Finally, over1_fw5 can be calculated as [{(r2_28)+ (r2_29)+ (r2_30)+ (r2_31)+ (r2_32)+ (r2_33)+ (r2_34)} / {(r2_0) + (r2_1) + (r2_2) + (r2_3) + (r2_4) + (r2_5) + (r2_6)}] * [{(r1_0) + (r1_1) + (r1_2) + (r1_3) + (r1_4) + (r1_5) + (r1_6)} / {(r1_-7)+ (r1_-6)+ (r1_-5)+ (r1_-4)+ (r1_-3)+ (r1_-2)+ (r1_-1)}.

[0060] FIG. 7 is a diagram showing an electronic device according to one embodiment of the present disclosure determining a correction factor. Referring to the table (700), the total sales volume (sold_patched_w1) of an item group during a specific past period (e.g., the period from day d-7 to day d-1, with the current date being day d) can be confirmed to be 4,948,264.80. The electronic device (110) can determine the total sales volume prediction value of an item group by a second sales volume prediction algorithm corresponding to a specific period by multiplying the total sales volume of the item group during a specific past period by the proportional value of the sum of the ratio values ​​of the days included in the specific period relative to the sum of the ratio values ​​of the days included in the specific past period. For example, the value obtained by multiplying sold_patched_w1(4,948,264.80) by over1_fw1(0.81286339), over1_fw2(1.047182324), over1_fw3(1.014613499), over1_fw4(1.001798799), and over1_fw5(0.946736608), respectively, can be determined as the total sales volume prediction value of the item group by the second sales volume prediction algorithm corresponding to a specific period (fw1, fw2, fw3, fw4, fw5).

[0061] Additionally, the electronic device (110) can determine the value obtained by dividing the total sales volume prediction value of an item group by the second sales volume prediction algorithm corresponding to a specific period by the total sales volume prediction value of an item group by the first sales volume prediction algorithm as a correction coefficient corresponding to a specific period.

[0062] FIG. 8 is a diagram showing that an electronic device (110) according to one embodiment of the present disclosure determines a daily sales volume prediction value of a target item. The electronic device (110) can determine a daily sales volume prediction value for each day of the period from 2023-09-25 to 2023-10-08 for a target item with item identification number 112753.

[0063] Specifically, the period from 2023-09-25 to 2023-10-01 may be a specific period of fw1 as week identification number 1. Additionally, the period from 2023-10-02 to 2023-10-08 may be a specific period of fw2 as week identification number 2. Here, the correction factor for the fw1 period may be 0.842294, and the correction factor for the fw2 period may be 1.035757. For specific periods of fw1 and fw2, the electronic device (110) can calculate a weekly sales volume prediction value after correction (a corrected total sales volume prediction value of a target item corresponding to a specific period) by multiplying the correction factor by the weekly sales volume prediction value before correction (a total sales volume prediction value of a target item corresponding to a specific period by the first sales volume prediction algorithm). The electronic device (110) can determine the daily sales volume prediction value after correction by multiplying the weekly sales volume prediction value after correction by a daily division factor. The daily sales volume prediction value before correction compared with this may be determined by the first sales volume prediction algorithm.

[0064] In one embodiment, the electronic device (110) may determine a daily division factor based on the ratio value of each day included in a specific period. Specifically, the daily division factor may be determined to be proportional to the ratio value of each day included in a specific period. For example, the electronic device (110) may determine the daily division factor of a specific date based on the ratio value of a specific date included in a specific period divided by the sum of the ratio values ​​of each of all days included in a specific period. For example, the electronic device (110) may determine the daily division factor of a specific date by the ratio value of a specific date included in a specific period divided by the sum of the ratio values ​​of each of all days included in a specific period.

[0065] In the flowcharts according to the present disclosure, each step of the method or algorithm is described in a sequential order, but the steps may be performed in any combination other than sequentially. Description of the flowcharts or flowcharts of the present disclosure does not exclude changes or modifications to the method or algorithm and does not imply that any step is essential or desirable. In one embodiment, at least some steps may be performed in parallel, iteratively, or heuristically. In another embodiment, at least some steps may be omitted or other steps may be added.

[0066] Various embodiments according to the present disclosure may be implemented as software on a machine-readable storage medium. The software may be software for implementing the various embodiments described in the present disclosure. The software may be inferred from the various embodiments described in the present disclosure by programmers skilled in the art to which the present disclosure pertains. For example, the software may be a program containing machine-readable instructions (e.g., instructions, code, or code segments). The machine may be a device capable of operating according to instructions called from the storage medium, for example, a computer. In one embodiment, the machine may be a computing device according to the various embodiments described in the present disclosure. In one embodiment, the processor of the machine may execute the called instruction to cause the components of the machine to perform functions corresponding to the instruction. The storage medium may mean any type of recording medium in which data is stored that can be read by the machine. The storage medium may include, for example, ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc. In one embodiment, the storage medium may be implemented in a distributed form on a networked computer system, etc. In this case, the software may be stored and executed in a distributed manner on the computer system, etc. In another embodiment, the storage medium may be a non-transitory storage medium. A non-transitory storage medium refers to a tangible medium that exists regardless of whether data is stored semi-permanently or temporarily, and does not include a signal that propagates transitorily.

[0067] Although the technical concept according to the present disclosure has been described by various embodiments above, the technical concept according to the present disclosure includes various substitutions, modifications, and alterations that can be made within the scope of understanding of a person skilled in the art to which the present disclosure pertains. Furthermore, it should be understood that such substitutions, modifications, and alterations may be included within the scope of the appended claims. Explanation of the symbols delete

Claims

Claim 1 A method for predicting sales volume on a sales platform, performed by an electronic device, comprises: a step of determining a total sales volume prediction value of said target item corresponding to a specific period based on a first sales volume prediction algorithm that predicts sales volume based on the past sales volume of said target item; a step of determining a total sales volume prediction value of said item group corresponding to the specific period based on a second sales volume prediction algorithm that predicts sales volume based on the past sales volume of said item group including said target item; a step of determining a correction coefficient based on the total sales volume prediction value of said item group corresponding to the specific period; a step of determining a corrected total sales volume prediction value of said target item by applying the correction coefficient to the total sales volume prediction value of said target item; and a step of determining a daily sales volume prediction value of said target item corresponding to a specific date included in said specific period by applying a date-based division coefficient to the corrected total sales volume prediction value of said target item, wherein the step of determining the correction coefficient comprises a step of determining a total sales volume prediction value of each of said items included in said item group corresponding to the specific period based on the first sales volume prediction algorithm. A method comprising the step of determining a correction coefficient based on the sum of the total sales volume prediction values ​​of each of the items and the total sales volume prediction value of the item group, wherein the correction coefficient is determined based on the value obtained by dividing the total sales volume prediction value of the item group by the sum of the total sales volume prediction values ​​of each of the items. Claim 2 A method according to claim 1, wherein the item group including the target item is a set of all items sold on the sales platform. Claim 3 delete Claim 4 delete Claim 5 A method according to claim 1, wherein the second sales volume prediction algorithm comprises logic for determining a ratio value corresponding to each date based on the attributes of each date from the current day to a specific future date. Claim 6 In paragraph 5, the above attribute is based on at least one of the day of the week, the date and the date of a specific event. Claim 7 In claim 5, the method wherein the ratio value is determined by an artificial neural network-based learning model that has been trained based on Light GBM (Light Gradient Boosting Machine). Claim 8 In paragraph 5, the method wherein the above-mentioned date-by-date division coefficient is determined based on the ratio value of each of the dates included in the above-mentioned specific period. Claim 9 In paragraph 8, the method wherein the above-mentioned date division coefficient is proportional to the ratio value of each of the dates included in the above-mentioned specific period. Claim 10 In claim 5, the second sales volume prediction algorithm comprises logic for determining a predicted total sales volume of the item group for a specific period based on the total sales volume of the item group from a specific past date to the present, the sum of the ratio values ​​of each day from the specific past date to the present, and the sum of the ratio values ​​of each day of the specific period. Claim 11 A method according to claim 10, wherein the total sales volume prediction value of the item group for a specific period determined by the second sales volume prediction algorithm is determined based on the value obtained by dividing the total sales volume of the item group from the past specific date to the present by the sum of the ratio values ​​for each day from the past specific date to the present and multiplying it by the sum of the ratio values ​​for each day of the specific period. Claim 12 A method according to claim 1, wherein the corrected total sales volume prediction value of the target item is determined based on the value obtained by multiplying the total sales volume prediction value of the target item by the correction coefficient. Claim 13 A method according to claim 1, wherein the daily sales volume forecast value of the target item corresponding to a date included in the specific period is determined based on the value obtained by dividing the corrected total sales volume forecast value of the target item by the sum of the daily division coefficients for each date included in the specific period and multiplying it by the daily division coefficient corresponding to the specific date. Claim 14 An electronic device comprising: one or more processors; and one or more memories storing instructions to be executed by said one or more processors, wherein, upon execution of said instructions, said one or more processors are configured to perform a method according to any one of claims 1, 2 and 5 through 13. Claim 15 A non-transient computer-readable recording medium having instructions to be executed by one or more processors, wherein the instructions are configured such that, upon execution of the instructions, the one or more processors perform a method according to any one of claims 1, 2 and 5 through 13.

Images