KR20210136729A - Method for providing automatic collection service for moneybox - Google Patents

Abstract

According to the present invention, disclosed is a method of providing a moneybox automatic-collection service. The method of providing a moneybox automatic-collection service includes the following steps of: calculating a current extra fund; activating saving rules for a saving account; comparing a past extra fund with the current extra fund; when the current extra fund is greater than the past extra fund, determining whether analysis data is sufficient; when the analysis data is sufficient, predicting a future extra fund; when the future extra fund is greater than or equal to the current extra fund, increasing a saving amount to be more than a previous saving amount, and, when the future extra fund is smaller than the current extra fund, reducing the saving amount to be less than the previous saving amount; confirming a final saving amount by postprocessing the saving amount; and transferring the saving amount from the parent amount to the saving amount. Therefore, the present invention is capable of improving a saving rate of a user.

Description

Method for providing automatic collection service for piggy bank

The present invention relates to a method of providing an automatic piggy bank collection service. Specifically, the present invention relates to a method of providing a service for automatically transferring an appropriate amount to a piggy bank by analyzing the user's parent account balance.

The content described in this section merely provides background information for the present embodiment and does not constitute the prior art.

The most common product among bank financial products is a savings product, and users can deposit and withdraw money at any time through the savings account provided by the savings product. As individual customers' interest in economic activities increases and economic knowledge becomes abundant, banks are providing a variety of financial products to satisfy customers of various tastes.

Recently, banks are providing a variety of small savings services that can increase users' savings rates by saving small amounts without burdening users.

However, in the case of the existing small savings service, the rules for saving such as category setting, saving frequency and ratio setting were complicated, so it was difficult for users to access it easily.

In addition, most of the savings methods of the small savings service are manually configured, so there was the inconvenience of always running the app or manually pressing the transfer button to save. Had a problem to do.

It is an object of the present invention to provide a method for providing an automatic piggy bank collection service capable of reducing the user's burden on savings by calculating an appropriate amount that the user will not be burdened with based on the user's master account balance change pattern.

Another object of the present invention is to provide a method for providing an automatic piggy bank collection service that selects an optimal prediction model from among various prediction models to accurately predict a user's balance change pattern to derive a final savings amount suitable for the user.

The objects of the present invention are not limited to the above-mentioned objects, and other objects and advantages of the present invention not mentioned may be understood by the following description, and will be more clearly understood by the examples of the present invention. Moreover, it will be readily apparent that the objects and advantages of the present invention may be realized by the means and combinations thereof indicated in the claims.

One aspect of the method for providing an automatic piggy bank collection service according to an embodiment of the present invention is the method for automatically transferring a savings amount from a mother account to a savings account. Activating a savings rule for the bank, comparing the past surplus funds with the current surplus funds, if the current surplus funds are larger than the past surplus funds, determining whether analysis data is sufficient, if the analysis data is sufficient, Predicting future spare funds, if the future spare funds are greater than or equal to the current spare funds, increase the savings than the previous savings, and if the future spare funds are smaller than the current spare funds, transfer the savings reducing the savings amount, post-processing the savings amount to determine a final savings amount, and transferring the savings amount from the mother account to the savings account.

In addition, in the post-processing of the savings amount, when the savings amount is less than or equal to the minimum savings amount, the minimum savings amount is determined as the final savings amount, and when the savings amount is greater than or equal to the maximum savings amount, and determining the maximum savings amount as the final savings amount.

In addition, when the analysis data is not sufficient, predicting a safe section further comprising the step of changing the maximum savings amount.

In addition, predicting the safe interval includes setting a plurality of intervals of the minimum balance, and determining which of the plurality of intervals the minimum balance after a preset period belongs to.

In addition, the post-processing of the savings may include rounding the savings amount to a final savings amount when the savings amount is between the minimum savings amount and the maximum savings amount.

In addition, according to claim 1, wherein the predicting of the future surplus funds comprises calculating an average error for each first past period of each of the n primary prediction models (M1), wherein the first past period is the first past period. The period from the time point to the current time point, and the ensemble secondary prediction model (M2) by selecting m primary prediction models (M1) having the smallest average error among the n primary prediction models (M1) calculating the average error for the first past period of , wherein m is a natural number greater than 2 and less than n; Selecting the smallest one as a final prediction model, and predicting the future surplus funds using the final prediction model.

In addition, the primary prediction model M1 includes a weighted moving average model, a simple exponential smoothing model, an Auto-regressive Integrated Moving Average (ARIMA) model, a Holt-Winters model, and a long short term memory (LSTM) model. , including at least one of the Prophet models.

In addition, in the step of calculating the average error of the primary prediction model M1 for the first past period, the primary prediction model M1 learns the second past period, and the second past period is a period from a second past time point that is earlier than the first past time point to the first past time point, and calculating an average error of the quadratic prediction model (M2) for the first past time period. , the secondary prediction model (M2) includes the step of learning for the second past period.

In addition, after calculating the basic savings amount and activating the savings rule, determining whether the saving time is the first saving time, and when the saving time is the first saving time, the saving amount is set to the It further includes the step of setting the basic savings amount.

In addition, after the step of determining the final savings amount, the steps of displaying the final savings amount, determining whether the savings rule for the savings account is deactivated, and determining whether the final balance exceeds a reference balance further includes

Another aspect of the method for providing an automatic savings account service according to an embodiment of the present invention is the method for automatically transferring a savings amount from a mother account to a savings account. Determining whether a user is a user or a low-income user, calculating the current spare funds, activating a savings rule for the savings account, comparing the past spare funds with the current spare funds, the current spare funds are the past If it is larger than the surplus funds, determining whether the analysis data is sufficient; if the analysis data is sufficient, predicting the future surplus funds; if the future surplus funds are greater than or equal to the current surplus funds, transfer the savings amount more, and when the future spare fund is smaller than the current spare fund, the savings amount is reduced from the previous savings amount, wherein the increase is calculated using a first parameter when the user of the parent account is the high-value user, Calculating using a second parameter different from the first parameter when the user of the mother account is the low-money user, post-processing the savings amount to determine a final savings amount, and transferring the savings amount from the mother account to the savings transfer to the account.

In addition, the magnitude of the increase is determined by the formula of Equation 1 below, and [Equation 1] Increase (P, M) = IL * P / (IV1/ (M / IV2) + P) where P is (The current spare fund / the future spare fund-1), wherein M is the current spare fund, the IL, the IV1 and IV2 are preset values, and the IL is the first or second parameter. do.

In addition, the size of the reduction is determined by the formula of Equation 2 below, and [Equation 2] reduction (P, M) = DL * P / ((M/DV2) /DV1) + P) where, the P is (the present spare fund/the future spare fund-1), wherein M is the future spare fund, and the DL, the DV1 and DV2 are preset values.

In addition, the calculating of the basic savings amount includes: calculating an exponential smoothing balance according to simple exponential smoothing of the mother account for a first period; multiplying the exponential smoothing balance by a preset basic savings ratio It includes the step of calculating

In addition, the step of calculating the current surplus funds includes: calculating the minimum balance maintained for the second period as a maintenance amount; calculating a reflected maintenance amount by multiplying the maintenance amount by a preset maintenance amount reflection ratio; and calculating the current spare funds by subtracting the reflected maintenance amount from the final balance at the current time.

According to the method of providing the automatic piggy bank collection service of the present invention, the user can reduce the user's burden on saving and improve the user's savings rate by automatically calculating and saving an appropriate amount that is not burdensome to the user.

In addition, the automatic piggy bank collection service providing method of the present invention selects an optimal prediction model from among various prediction models to accurately predict a user's balance change pattern, and through this, derive a final savings amount suitable for the user, and the user's burden The optimum savings rate can be derived from this limit.

The specific effects of the present invention in addition to the above will be described together while describing specific details for carrying out the invention below.

1 is a view for explaining a system in which a method for providing an automatic piggy bank collection service according to an embodiment of the present invention is performed.
2 is a flowchart illustrating a method of providing an automatic piggy bank collection service according to an embodiment of the present invention.
3 is a flowchart for describing in detail the steps of calculating the basic savings amount and selecting parameters of FIG. 2 .
FIG. 4 is a flowchart for explaining in detail the step of calculating the current spare funds of FIG. 2 .
5 is a graph for explaining a state in which the size of the reduction by the reduction function changes as the surplus funds decrease.
6 is a conceptual diagram for explaining a method of predicting a minimum balance section and determining a maximum savings amount according to the present invention.
7 is a graph for explaining a state in which the size of the increase by the increase function changes as the surplus funds increase.
8 is a flowchart for explaining in detail the step of predicting the future spare funds of FIG.
9 is a flowchart for explaining the primary prediction model (M1) and the secondary prediction model (M2).
10 is a flowchart for explaining in detail a step in which the primary prediction model M1 of FIG. 8 calculates an average error.
11 is a diagram for explaining the learning, evaluation and prediction period of the prediction models of the present invention.
12 is a flowchart for explaining in detail a step in which the secondary prediction model M2 of FIG. 8 calculates an average error.
13 is a flowchart for describing in detail the post-processing step of FIG. 2 .
14 is an exemplary diagram for explaining an activation interface of the present invention.
15 and 16 are time diagrams for explaining whether automatic collection is performed according to an activation time of the present invention.
17 is an exemplary diagram for explaining an interface in which a guide message is displayed according to an activation time of the present invention.

The terms or words used in this specification and claims should not be construed as being limited to a general or dictionary meaning. In accordance with the principle that the inventor can define a term or concept of a word in order to best describe his invention, it should be interpreted as meaning and concept consistent with the technical idea of the present invention. In addition, the embodiments described in the present specification and the configurations shown in the drawings are only one embodiment in which the present invention is realized, and do not represent all the technical spirit of the present invention, so they can be substituted at the time of the present application. It should be understood that there may be various equivalents and modifications and applicable examples.

Terms such as first, second, A, B, etc. used in this specification and claims may be used to describe various elements, but the elements should not be limited by the terms. The above terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, a first component may be referred to as a second component, and similarly, a second component may also be referred to as a first component. The term 'and/or' includes a combination of a plurality of related listed items or any of a plurality of related listed items.

The terms used in this specification and claims are used only to describe specific embodiments, and are not intended to limit the present invention. The singular expression includes the plural expression unless the context clearly dictates otherwise. It should be understood that terms such as “comprise” or “have” in the present application do not preclude the possibility of addition or existence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification in advance. .

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and should not be interpreted in an ideal or excessively formal meaning unless explicitly defined in the present application. does not

In addition, each configuration, process, process, or method included in each embodiment of the present invention may be shared within a range that does not technically contradict each other.

Hereinafter, a detailed description will be given of a method of providing a piggy bank automatic collection service that automatically transfers an appropriate amount to a piggy bank according to the user's parent account balance.

1 is a view for explaining a system in which a method for providing an automatic money collection service according to an embodiment of the present invention is performed.

Referring to FIG. 1 , a system for performing a method for providing an automatic piggy bank collection service according to an embodiment of the present invention may include a bank server 100 (hereinafter referred to as a server) and a user terminal 200 .

The server 100 may be the subject of the method for providing the automatic piggy bank collection service according to the embodiment of the present invention. The server 100 may manage a plurality of accounts for a specific user, and may perform a function of opening an account or transferring or remitting a specific amount of money.

Specifically, the server 100 may create a savings account for saving based on the user's mother account. In this case, the savings account may be provided to the user in the form of a piggy bank. However, the account number of the savings account may not be provided to the user. In addition, the user is prohibited from withdrawing only a partial amount from the savings account, and may only be able to withdraw the entire amount through a request to cancel the savings account (eg, request to open a piggy bank).

The opening of a savings account may proceed only with the user's parent account. In this case, the savings account opened on the basis of the mother account may be opened as a regular deposit account for demand deposit. Accordingly, a certain interest may be given to the savings account. However, this is only an example and the present invention is not limited thereto.

In this case, the parent account may include one or more linked accounts. Accordingly, the savings account may be connected to other accounts (eg, linked accounts) in addition to one specific account.

Hereinafter, for convenience of explanation, a savings account will be described as a piggy bank, and hereinafter, a savings account and a piggy bank will be used with the same meaning.

Subsequently, the server 100 may receive the setting of the savings rule for the piggy bank from the user. A user may set a savings rule through the user terminal 200 . The savings rule may be designed to automatically transfer an amount determined according to a predetermined algorithm from the balance of the mother account to the piggy bank at predetermined intervals.

For example, the savings rule may be to automatically transfer the final savings amount determined according to a predefined algorithm from the balance of the parent account to the piggy bank every predetermined period. As an embodiment, the savings rule sets an appropriate savings amount that will not be burdensome to the user based on the balance change pattern of the parent account for a predetermined number of times (eg, 1-2 times) in a predetermined period (eg, one week). This includes the method of automatic debit from the bank to the piggy bank. This savings rule can be called a piggy bank automatic collection service.

In this case, the user may activate the savings rule in the piggy bank opening step through the user terminal 200 .

Subsequently, when the savings amount exceeds a preset maximum or there is a request to cancel the piggy bank (eg, a request to open the piggy bank), the savings amount of the piggy bank may be transferred to the mother account. However, this is only an example, and the termination condition of the piggy bank may be variously modified and implemented.

An application (eg, a banking app) used to communicate with the server 100 may be installed in the user terminal 200 . The user may create a piggy bank, set or change a savings rule applied to the piggy bank, or request cancellation of the piggy bank through an application installed in the user terminal 200 .

In addition, the application does not expose the savings amount of the piggy bank, but may expose different item images that change according to a preset standard according to the size of the savings amount. The user can guess the amount of savings accumulated in the piggy bank by looking at the exposed item image.

In addition, the application may allow the user to check the amount of savings in the piggy bank through a peek function provided once at a certain period.

The method of providing the automatic piggy bank collection service of the present invention may be performed only on the server 100 or may be implemented by being divided into the server 100 and the user terminal 200 .

In an embodiment of the present invention, the user terminal 200 may refer to various electronic devices such as a personal computer (PC), a notebook computer, a tablet, a mobile phone, a smart phone, and a wearable device (eg, a watch-type terminal). .

In addition, the user terminal 200 includes an input unit for receiving a user's input, a display unit for displaying visual information, a communication unit for transmitting and receiving signals to and from the outside, and processing data and controlling each unit inside the user terminal 200, It may include a control unit for controlling data transmission/reception between the two. Hereinafter, what the control unit performs inside the user terminal 200 according to the user's command is collectively referred to as the user terminal 200 performs.

In addition, the server 100 is a computer system that provides information or services to the user terminal 200 through a communication network, and may refer to a computer program or a device.

In addition, the above-described server may be directly operated or managed by a specific company or individual, or may be outsourced, and may be operated by the same entity.

In addition, since the name of the server is not specified, the functions performed by each server can be performed in other servers as well, and the functions performed in one server can be performed separately in multiple servers. Of course.

Hereinafter, a method for providing an automatic piggy bank collection service according to an embodiment of the present invention will be described in detail.

2 is a flowchart illustrating a method of providing an automatic piggy bank collection service according to an embodiment of the present invention.

Referring to FIG. 2 , in the method of providing a piggy bank automatic collection service according to an embodiment of the present invention, first, a final balance, a basic savings amount, a maintenance amount, and a current surplus fund are calculated ( S110 ).

In this case, the automatic piggy bank collection service according to the present embodiment may be performed periodically. For example, the automatic piggy bank collection service according to the present embodiment may be performed on a weekly basis. However, the present embodiment is not limited thereto.

If the automatic piggy bank collection service according to the present embodiment is performed on a weekly basis, the piggy bank automatic collection service may be started at the same time every week. For example, an automatic piggy bank collection service may be started as of 00:00 on Friday. In this case, the automatic piggy bank collection service may be processed during Saturday. Saturdays have relatively fewer deposits and withdrawals in the mother account compared to weekdays, and the deposit and withdrawal patterns for one week can be clearly analyzed, so it may be an appropriate period for the automatic piggy bank collection service according to the present embodiment to be performed.

Hereinafter, for convenience, it will be described that the automatic piggy bank collection service according to the present embodiment is performed in units of one week and starts at 00:00 on Friday.

The automatic piggy bank collection service according to the present embodiment may create various contexts based on a service start time, such as 0 o'clock on Friday. For example, the generated context includes deposit/withdrawal amount and rate, amount and rate by deposit/withdrawal classification, deposit/withdrawal pattern (e.g., loan repayment, card payment, etc.), first transaction time, average number of transactions per month ( For example, it may include at least one of the last 6 months) and automatic collection sessions.

At this time, the final balance, the basic savings amount, the maintenance amount, and the current surplus funds may also be calculated. In addition, the minimum savings amount and the maximum savings amount may be preset. For example, the minimum savings amount may be 1,000 won, and the maximum savings amount may be 5000 won. However, the present embodiment is not limited thereto.

The final balance may be a balance at the time of processing. For example, if the automatic piggy bank collection service starts at 00:00 on Friday, 00:00 on Friday may be the reference point of the final balance.

3 is a flowchart for describing in detail the steps of calculating the basic savings amount and selecting parameters of FIG. 2 .

2 and 3, the step of calculating the basic savings amount (S110) first calculates an exponentially smoothed balance (S111).

Here, the exponentially smoothed balance may mean a balance predicted by a simple exponentially smoothing method. The exponential smoothing balance may be derived using, for example, the balance of the last two weeks. However, the present embodiment is not limited thereto.

Next, a basic savings amount is calculated (S112).

The basic savings amount may be calculated by multiplying the exponential smoothing balance by the basic savings ratio. In this case, the basic savings rate may be a preset parameter. For example, the basic savings rate may be 1%. However, the present embodiment is not limited thereto.

Next, it is determined whether the basic savings amount is greater than or equal to the reference value (S113).

In this case, the reference value may be a preset value. For example, the reference value may be 5000 won. However, the present embodiment is not limited thereto. If the basic savings amount is greater than or equal to the reference value, the user of the parent account may be classified as a high-value user. Conversely, when the basic savings amount is smaller than the reference value, the user of the mother account may be determined as the low-value user.

If the basic savings amount is greater than or equal to the reference value, the first parameter is selected (S114).

Here, the first parameter may be a parameter for a high-value user. The first parameter is a preset parameter, and may be a large amount increase limit. The first parameter may be used later in the increment function. For example, the first parameter may be 2000 won. However, the present embodiment is not limited thereto.

If the basic savings amount is less than the reference value, the second parameter is selected (S115).

Here, the second parameter may be a parameter for a low-cost user. The second parameter is a preset parameter, and may be a low-cost increase limit. The second parameter may be used later in the increment function. For example, the second parameter may be 1,000 won. However, the present embodiment is not limited thereto.

The method of providing the automatic piggy bank collection service according to the present embodiment may select a parameter for an increase limit by classifying a high-value user and a low-value user according to an account balance. Accordingly, when determining the size of the increase in the future savings amount, it is possible to derive a more appropriate final savings amount by varying the limit of increase between the high-value user and the low-income user.

FIG. 4 is a flowchart for explaining in detail the step of calculating the current spare funds of FIG. 2 .

2 and 4, the maintenance amount is calculated first (S115).

The maintenance amount may mean a minimum balance maintained for a period corresponding to 99% of the maintenance amount calculation period during the maintenance amount calculation period. For example, if the maintenance amount calculation period is 60 days, it may mean the minimum balance maintained for 1425 hours, which is 99% of the 60 days during the last 60 days.

Next, the reflected maintenance amount is calculated (S116).

The reflected maintenance amount may be a value obtained by multiplying the maintenance amount by a maintenance amount reflection ratio. In this case, the maintenance amount reflection ratio may be a preset parameter. For example, the maintenance amount reflection ratio may be 60%. However, the present embodiment is not limited thereto.

Next, the current surplus funds are calculated (S117).

The current spare funds may mean spare funds this week when the service is performed on a weekly basis. The current surplus funds may be calculated by deducting the reflected maintenance amount from the final balance.

If it is not the time when the automatic piggy bank collection service according to the present embodiment is performed for the first time, information on past surplus funds and previous savings amounts calculated in the past may exist. In the case of past surplus funds, in the case of services performed on a weekly basis, it may mean the surplus funds from the previous week, and the previous savings may mean the savings finally deposited in the previous week.

Again, referring to FIG. 2, it is determined whether the savings rule for the savings account is activated (S120).

The savings rule may be determined on/off by the user. The savings rule may be activated only when the on-state is on as of 00:00 on Friday. If the savings rule is not activated, the automatic piggy bank collection service may be terminated.

Conversely, when the savings rule is activated, it is then checked whether it is the time of the first savings (S130).

Whether it is the time of the initial savings can be confirmed through the context of the time of the initial transaction created previously. If it is the time of the first deposit, data on the past may not exist. That is, data on past spare funds or previous savings may not exist.

Therefore, in the case of the initial savings time, the savings amount is set as the basic savings amount (S140).

The basic savings amount may be calculated in step S110. That is, in the case of the initial point in time when there is no data on previous savings, the savings amount can be reasonably set through the basic savings amount. The basic savings amount is a product of the exponential smoothing balance and the basic savings ratio, and may be determined according to a preset basic savings ratio.

If it is not the time of the initial savings, the current surplus funds and the past surplus funds can be compared. In other words, in the weekly service, the surplus funds of the previous week and the surplus funds of this week can be compared.

Therefore, if it is not the time of the initial savings, it is determined whether the current excess funds are greater than the past spare funds (S150).

If the current spare funds are not greater than the past spare funds, it is determined whether the current spare funds are smaller than the past spare funds (S160).

If the current spare funds are smaller than the past spare funds, the savings are set to be deducted by the first reduction from the previous savings (S170).

In this case, the first reduction may be defined by Equation 1 below.

<Equation 1>

Decrease (P, M)=DL * P / ((M/DV2) /DV1) + P)

Here, P is a rate of change, and M may be a spare fund. The DL, the DV1, and the DV2 are preset parameters, and may be a reduction limit, a reduction variable 1, and a reduction variable 2, respectively. For example, the depreciation limit may be 3000 won, and the depreciation variable 1 and the depreciation variable 2 may be 50 and 5000, respectively. However, the present embodiment is not limited thereto.

The first reduction may be calculated as a reduction ((past surplus funds/present surplus funds-1), present surplus funds).

5 is a graph for explaining a state in which the size of the reduction by the reduction function changes as the surplus funds decrease.

The horizontal axis of the graph of FIG. 5 is the decrease ratio of M (surplus funds) of the decrease (P, M) function, and the vertical axis is the value of the decrease (P, M) function. At this time, the reduction limit DL was set at KRW 3,000.

Referring to FIG. 5 , depending on the size of the surplus funds, the change pattern of the reduced amount may vary. In other words, when there is a large amount of spare funds (3,000,000 KRW), the amount of the reduction increases relatively slowly when the spare funds gradually decrease, and when the spare funds are small (30,000 KRW), it is relatively when the spare funds decrease gradually. The size of the reduction can increase rapidly. When the surplus funds are the value (300,000) therebetween, when the surplus funds gradually decrease, the degree to which the amount of reduction increases may be between a case where the surplus funds are large and a case where the surplus funds are small.

In the method of providing a piggy bank automatic 4 collection service according to the present embodiment, the size of the reduction increases relatively quickly when the excess funds are small and focuses on preserving the balance of the mother account, and when the spare funds are large, the size of the reduction is relatively large. It can increase slowly so that the amount of savings in your savings account does not decrease too much. Through this, it is possible to efficiently manage the savings ratio according to the size of the user's spare funds.

Referring back to FIG. 2 , in step S160, if the current spare funds are not smaller than the past spare funds, it corresponds to the case where the current spare funds and the past spare funds are the same. Accordingly, in this case, the savings amount is set as the previous savings amount (S180).

The basic savings amount may be calculated in step S110. That is, if the current surplus funds and the past surplus funds are the same, the savings ratio can be maintained at the same level by setting the same amount as the previous savings amount.

If, in step S150, if the current surplus funds are greater than the past surplus funds, it is determined whether the analysis data is sufficient (S190).

Determining whether the analysis data is sufficient may include determining whether the deposit/withdrawal transaction history is more than N weeks. In this case, N may be a preset value. For example, N weeks may be 24 weeks. However, the present embodiment is not limited thereto.

In addition, determining whether the analysis data is sufficient may include determining whether the deposit and withdrawal transaction details are M or more on average per month. In this case, M may be a preset value. For example, M items may be 15 items. However, the present embodiment is not limited thereto.

If it is determined that the analysis data is insufficient, a safe interval prediction is performed (S200).

The safe interval prediction may be a step of setting the savings amount as safely as possible so that the actual balance does not become negative because the savings amount is large when there is insufficient analysis data.

Predicting the safe interval may include setting a plurality of intervals of the minimum balance, and determining which of the plurality of intervals the minimum balance after a preset period belongs to.

6 is a conceptual diagram for explaining a method of predicting a minimum balance section and determining a maximum savings amount according to the present invention.

Specifically, referring to FIG. 6 , a plurality of sections of the minimum balance may be set. In FIG. 6 , six sections can be divided based on the unit of 1,000 won. For example, the sections may be divided into less than 1,000 won, more than 1,000 won and less than 2,000 won, more than 2,000 won and less than 3,000 won, more than 3,000 won and less than 4,000 won, more than 4,000 won and less than 5,000 won, and more than 5,000 won. However, since this is only an example, the present embodiment is not limited thereto.

Then, it is determined which of the plurality of sections the balance of the user's mother account after a preset period belongs.

In this case, the preset period may be, for example, 8 days. That is, when the automatic piggy bank collection service is performed on a weekly basis, the section to which the balance of the next week belongs can be predicted. However, the present embodiment is not limited thereto.

Predicting the minimum balance interval can be optimized to select the highest possible amount interval within a safe range. This is because, when an amount greater than the minimum balance is calculated as a savings amount, the balance may become negative, so it should be avoided first.

Predicting the minimum balance interval can be done using the user's recent deposit and withdrawal data and common information of all customers. The input for predicting the minimum balance interval is, for example, the user's current balance, this week maintenance fund, number of withdrawals this week, total withdrawal amount this week, average withdrawal amount this week, account opening parking, deposit rate this week, number of deposits this week, check card this week It may include at least one of the withdrawal amount, the total deposit amount this week, the average deposit amount this week, and the number of withdrawals from the check card this week. However, the present embodiment is not limited thereto.

Predicting the minimum balance interval may use, for example, a Light Gradient Boosting Machine (LightGBM) algorithm. However, the present embodiment is not limited thereto.

Again, referring to FIG. 2, after the prediction of the minimum balance section is completed, the maximum savings amount is changed (S210).

If the maximum savings amount is set at 5,000 won in advance, the maximum savings amount may be lowered or maintained. Specifically, referring to FIG. 6 , when the minimum balance interval prediction is less than 1,000 won or more than 1,000 won and less than 2,000 won, the maximum savings amount may be changed to 1,000 won. If the minimum balance interval prediction is 2,000 won or more and less than 3000 won, the maximum savings amount may be changed to 2000 won, and if the minimum balance interval prediction is 3000 won or more but less than 4,000 won, the maximum savings amount may be changed to 3000 won. If the minimum balance interval forecast is 4000 won or more but less than 5000 won, the maximum savings amount is changed to 4000 won, and if the minimum balance interval forecast is 5000 won or more, the maximum savings amount may be maintained at 5000 won. However, the present embodiment is not limited thereto.

The maximum savings amount may be used when determining the final savings amount in a later post-processing step (S270).

Next, the savings amount is set to increase by the first increase from the previous savings amount (S220).

In this case, the first increase may be defined by Equation 2 below.

<Equation 2>

Increment(P, M)=IL * P / (IV1 / (M / IV2) + P)

Here, P is a rate of change, and M may be a spare fund. The IL, IV1, and IV2 are preset parameters, and may be an increase limit, an increase variable 1, and an increase variable 2, respectively.

In this case, the increase limit may have different parameters for high-income users and low-cost users. That is, a high amount increase limit (first parameter) may be used for a high value user, and a low price increase limit (second parameter) may be used for a low value user.

For example, the first parameter, that is, the high-value increase limit may be 2,000 won, and the second parameter, that is, the low-priced increase limit, may be 1,000 won. Increment variable 1 and Increment variable 2 may be 50 and 5000, respectively. However, the present embodiment is not limited thereto.

The first increase may be calculated as an increase ((current surplus funds/past surplus funds-1), current surplus funds).

7 is a graph for explaining a state in which the size of the increase by the increase function changes as the surplus funds increase.

The horizontal axis of the graph of FIG. 7 is the increase rate of M (extra funds) of the increase (P, M) function, and the vertical axis indicates the value of the increase (P, M) function. At this time, the increase limit IL was set at 1,000 won.

Referring to FIG. 7 , depending on the size of the surplus funds, the change pattern of the increased amount may vary. That is, in the case of a large amount of surplus funds (3,000,000 won), the amount of increase increases relatively quickly when the surplus funds gradually increase. The size of the increment may increase slowly. When the surplus funds are the value in between (300,000), when the surplus funds gradually increase, the degree of increase in the amount of the surplus funds may be between a case where the surplus funds are large and a case where the surplus funds are small.

The method of providing the automatic piggy bank collection service according to this embodiment concentrates on preserving the balance of the mother account by increasing the amount of increase relatively slowly when the excess funds are small, and increases the size of the increase relatively quickly when the spare funds are large. It can be increased so that the amount of savings saved into the savings account increases significantly. Through this, it is possible to efficiently manage the savings ratio according to the size of the user's spare funds.

Again, referring to FIG. 2, when it is determined that the analysis data is sufficient in step S190, the future spare funds are predicted (S230). In this case, in the case of a service performed on a weekly basis, the future surplus funds may mean the borrower surplus funds.

8 is a flowchart for explaining in detail the step of predicting the future surplus funds of FIG. 2 , and FIG. 9 is a flowchart for explaining the primary prediction model and the secondary prediction model.

Referring to FIG. 8 , first, the average error of the primary prediction model is calculated ( S231 ).

Specifically, referring to FIG. 9 , the primary prediction model M1 may be plural. The primary prediction model (M1) is, for example, a weighted moving average model, a simple exponential smoothing model, an ARIMA (Auto-regressive Integrated Moving Average) model, a Holt-Winters model, and a long short term memory (LSTM) model. It may include at least one of a model and a prophet model. However, the present embodiment is not limited thereto.

10 is a flowchart for explaining in detail the step in which the primary prediction model M1 of FIG. 8 calculates the average error, and FIG. 11 is a diagram for explaining the learning, evaluation and prediction period of the prediction models of the present invention. .

Referring to FIG. 10 , the primary prediction model M1 learns about the second past period ( S231a ).

Specifically, referring to FIG. 11 , a first past period may exist in the past from the present, that is, the prediction time. That is, the first past period may be the most recent past. The second past period may exist in a past direction from the first past period. That is, the second past period may be a past period connected to the first past period and located before the first past period.

For example, the first past period may be a period of one week, and the second past period may be a period of 23 weeks. However, the present embodiment is not limited thereto.

The first prediction model M1 may be trained using data of the second past period. Through this, the primary prediction model M1 may determine the internal parameters.

Referring again to FIG. 10 , the primary prediction model calculates an average error for the first past period ( S231b ).

The primary prediction model M1 may perform prediction for the first past period according to the determined internal parameters. However, since the first past period is in the past from the present time, it may be a period in which data already exists. Therefore, the average error can be calculated by comparing the data that actually exists with the data predicted by the primary prediction model M1.

The average error may be calculated for each of the plurality of primary prediction models M1.

Again, referring to FIG. 8 , the average error of the secondary prediction model M2 is calculated ( S232 ).

Specifically, referring to FIG. 9 , the secondary prediction model M2 may be an ensemble prediction model by selecting models having the smallest mean error among the primary prediction models M1. In this case, a plurality of secondary prediction models M2 may be derived according to the number of ensemble models.

For example, the TOP 2 ensemble model may be a prediction model in which two prediction models having the smallest average error among the primary prediction models M1 are ensembled. Also, the TOP 6 ensemble model may be a prediction model in which six prediction models having the smallest average error among the primary prediction models M1 are ensembled.

If there are n primary prediction models M1, a maximum of n-1 secondary prediction models M2 may exist. That is, the quadratic prediction model M2 includes Top m ensemble models, where m may be 2 to n.

12 is a flowchart for explaining in detail a step in which the secondary prediction model of FIG. 8 calculates an average error.

Referring to FIG. 12 , the secondary prediction model learns about the second past period ( S232a ).

Specifically, referring to FIG. 11 , the secondary prediction model M2 may be trained using data of the second past period. Through this, the secondary prediction model M2 may determine the internal parameters.

Again, referring to FIG. 12 , the secondary prediction model calculates an average error for the first past period ( S232b ).

The secondary prediction model M2 may perform prediction with respect to the first past period according to the determined internal parameters. However, since the first past period is in the past from the present time, it may be a period in which data already exists. Therefore, the average error may be calculated by comparing the data that actually exists with the data predicted by the secondary prediction model M2.

The average error may be calculated for each of the plurality of secondary prediction models M2.

Again, referring to FIG. 8 , a final prediction model is selected ( S233 ).

Specifically, referring to FIG. 9 , as the final prediction model, any one of a primary prediction model M1 and a secondary prediction model M2 may be selected. Since the average error of the primary prediction model (M1) and the secondary prediction model (M2) is calculated respectively, the prediction model with the lowest average error among the primary prediction model (M1) and the secondary prediction model (M2) is the final prediction can be selected as a model.

Again, referring to FIG. 8, the future spare funds are predicted using the final prediction model (S234).

Specifically, referring to FIG. 9 , since the final prediction model is selected as the prediction model having the lowest average error among the primary prediction model M1 and the secondary prediction model M2, the most accurate prediction may be possible. Therefore, the final prediction model can predict the future surplus funds relatively accurately.

Again, referring to FIG. 2, it is determined whether the future spare funds are greater than or equal to the current spare funds (S240).

If the future surplus funds are smaller than the current surplus funds, the savings are set to be deducted by the second reduction from the previous savings (S250).

In this case, the second reduction may be defined by Equation 1 described above.

That is, the second reduction may be defined as a reduction ((current spare funds/future spare funds-1), future spare funds).

If the future spare funds are greater than or equal to the current spare funds, the savings amount is set to increase by a second increase from the previous savings amount (S260).

In this case, the second increase may be defined by Equation 2 described above.

That is, the second increase may be defined as an increase (current surplus funds/past surplus funds-1, present surplus funds).

Then, the final savings amount is determined by post-processing the savings amount (S270).

13 is a flowchart for describing in detail the post-processing step of FIG. 2 .

Referring to FIG. 13 , it is first determined whether the savings amount is greater than or equal to the maximum savings amount ( S271 ).

If the savings amount is greater than or equal to the maximum savings amount, the final savings amount is set as the maximum savings amount (S272).

That is, the post-processing may be performed so that an amount greater than the maximum amount of savings that is finally set in advance is not transferred to the savings account.

If the savings amount is less than the maximum savings amount, it is determined whether the savings amount is less than or equal to the minimum savings amount (S273).

If the savings amount is less than or equal to the minimum savings amount, the final savings amount is set as the minimum savings amount (S272).

That is, post-processing may be performed so that an amount smaller than the final minimum amount of savings is not transferred to the savings account.

In FIG. 13 , steps S271 and S273 are shown to be sequentially performed, but the present embodiment is not limited thereto. That is, the order of steps S271 and S273 may be reversed.

If, in step S273, the savings amount is greater than the minimum savings amount, the savings amount may be located between the maximum savings amount and the minimum savings amount. In this case, the final savings amount is determined by rounding the savings amount to a specific unit (S275).

Here, the specific unit may be, for example, a 100 won unit. Through this, the final savings amount transferred from the parent account to the savings account may be a neatly rounded amount in units of 100 won or less. However, the present embodiment is not limited thereto.

Again, referring to FIG. 2, the final savings amount is displayed (S280).

The user can check how much the final savings amount is before the account transfer of the final savings amount is carried out. Accordingly, the user may have the opportunity to approve or object to the final savings being transferred to the savings account. Approval of the user's transfer may not require any special action. However, in order for the user to object, it may be necessary to change the savings rule from an on state to an off state. However, the present embodiment is not limited thereto.

Next, it is determined whether the final balance is sufficient (S290).

If the final balance is not sufficient, the final savings amount is not transferred (S320). In this case, whether the final balance is sufficient may be determined based on a preset final transferable balance. For example, the final transferable balance may be 10,000 won. That is, if the final balance is 10,000 won or less, the transfer of the final savings amount may not be performed. However, the present embodiment is not limited thereto.

If, if the final balance is sufficient, it is determined whether the savings rule is activated (S300).

The user may turn off the savings rule in the final savings amount display step ( S280 ). Accordingly, if the savings rule is turned off, the savings rule may be deactivated. Conversely, when the user keeps the savings rule on, the savings rule may be continuously activated.

If the savings rule is activated, the final savings amount is transferred to the savings account (S310). If the savings rule is deactivated, the final savings amount is not transferred (S320).

The method of providing the automatic piggy bank collection service according to the present embodiment can efficiently set the increase and decrease of the savings amount by using data of the current surplus funds, the past surplus funds, and the predicted data of the future surplus funds.

In addition, in the prediction of future spare funds, the accuracy of prediction can be increased by selecting an optimal model from among a plurality of prediction models. Furthermore, by designing the increase/decrease ratio of the increase function and the decrease function to be different according to the amount of spare funds, an optimized transfer according to the user's spare funds may be possible.

In addition, the method for providing the automatic piggy bank collection service according to the present embodiment may predict a safe section when the analysis data is not sufficient and appropriately adjust the savings amount so that the savings amount is not greater than the future balance.

In this way, the user can automatically perform maximum savings without burdensome limits.

14 is an exemplary diagram for explaining an activation interface of the present invention.

Referring to FIG. 14 , the first screen 201 may include an activation area A1 and an activation button B1. The activation area A1 may be an area capable of controlling whether to turn on or off the automatic piggy bank collection service according to the present embodiment.

The activation button B1 may be a part capable of turning on or off the automatic piggy bank collection service according to the present embodiment by a touch. The activation button B1 may display the current state of the automatic piggy bank collection service according to the present embodiment in the form of a slide. However, the present embodiment is not limited thereto.

The first screen 201 may be in a state in which the automatic piggy bank collection service according to the present embodiment is turned off. At this time, when the user touches the activation button B1 of the first screen 201, the automatic piggy bank collection service according to the present embodiment is turned on as shown in the second screen 202 through an electronic signature such as PIN authentication or biometric authentication. It can be (On).

The second screen 202 also includes an activation area A1 and an activation button B1 like the first screen 201, but the display of the activation button B1 is different from the first screen 201 in this embodiment. An On state of the automatic piggy bank collection service may be displayed.

15 and 16 are time diagrams for explaining whether automatic collection is performed according to an activation time of the present invention.

First, referring to Figure 15, if the service is performed on a weekly basis, if automatic collection is on and activated before 00:00 on Friday, which is the basis, after 00:00 on Friday, automatic collection is deactivated In any case, automatic collection can be performed. In this case, the time point at which the automatic collection is performed may be around Saturday. The final savings amount display step ( S280 ) of FIG. 2 is a step after the automatic collection is performed, and may be a separate step from the automatic collection deactivation of FIG. 15 .

Subsequently, referring to FIG. 16 , when automatic collection is turned off and deactivated before 00:00 on Friday, which is the basis when the service is performed on a weekly basis, automatic collection is activated again after 00:00 on Friday Automatic collection may not be performed even in this case. In this case, the time point at which the automatic collection is performed may be around Saturday.

17 is an exemplary diagram for explaining an interface in which a guide message is displayed according to an activation time of the present invention.

16 and 17 , the third screen 203 may include an activation area A1, an activation button B1, and a message Ma. In FIG. 16 , when automatic collection is activated after 00:00 on Friday (before 00:00 on Saturday), a message Ma may be displayed as shown in the third screen 203 .

The message Ma may notify the user that the automatic collection is performed in the next week instead of this week if the time point is late even if the user activates the automatic collection. When the user touches the message Ma, the side message S1 may be slid and displayed as shown in the fourth screen 204 .

The fourth screen 204 may display a side message S1. As for the side message S1, a push or message displayed to the user may be sequentially displayed in chronological order. Through this, the user can definitively determine whether to perform automatic collecting.

The above description is merely illustrative of the technical idea of this embodiment, and various modifications and variations will be possible by those skilled in the art to which this embodiment belongs without departing from the essential characteristics of the present embodiment. Accordingly, the present embodiments are intended to explain rather than limit the technical spirit of the present embodiment, and the scope of the technical spirit of the present embodiment is not limited by these embodiments. The protection scope of this embodiment should be interpreted by the following claims, and all technical ideas within the equivalent range should be interpreted as being included in the scope of the present embodiment.

Claims (15)

A method of providing an automatic piggy bank collection service that automatically transfers a savings amount from a mother account to a savings account,
calculating the current surplus funds;
activating a savings rule for the savings account;
comparing the past spare funds with the current spare funds;
determining whether the analysis data is sufficient when the current spare funds are greater than the past spare funds;
predicting future surplus funds when the analysis data is sufficient;
increasing a savings amount from a previous savings amount when the future spare funds are greater than or equal to the current spare funds, and decreasing the savings amount from the previous savings amounts when the future spare funds are smaller than the current spare funds;
post-processing the savings amount to determine a final savings amount; and
transferring the savings amount from the mother account to the savings account;
How to provide an automatic piggy bank collection service.
According to claim 1,
Post-processing the savings amount,
If the savings amount is less than or equal to the minimum savings amount, the minimum savings amount is determined as the final savings amount;
When the savings amount is greater than or equal to the maximum savings amount, determining the maximum savings amount as the final savings amount
How to provide an automatic piggy bank collection service.
3. The method of claim 2,
When the analysis data is not sufficient, predicting a safe section further comprising the step of changing the maximum savings
How to provide an automatic piggy bank collection service.
4. The method of claim 3,
The step of predicting the safety interval,
setting a plurality of sections of the minimum balance;
Comprising the step of determining which of the plurality of sections the minimum balance after a preset period belongs
How to provide an automatic piggy bank collection service.
3. The method of claim 2,
Post-processing the savings amount,
When the savings amount is between the minimum savings amount and the maximum savings amount, rounding the savings amount to a specific unit and determining the final savings amount
How to provide an automatic piggy bank collection service.
According to claim 1,
The step of predicting the future surplus funds,
calculating an average error for a first past period of each of the n primary prediction models, wherein the first past period is a period from a first past time to a present time;
Among the n primary prediction models, m primary prediction models (M1) having the smallest average error are selected to calculate the average error for the first past period of each ensemble secondary prediction model, wherein m is 2 greater than and a natural number smaller than n;
selecting one of the primary prediction model and the secondary prediction model having the smallest average error as a final prediction model;
Comprising the step of predicting the future surplus funds using the final prediction model
How to provide an automatic piggy bank collection service.
7. The method of claim 6,
The primary prediction model is a weighted moving average model, a simple exponential smoothing model, an Auto-regressive Integrated Moving Average (ARIMA) model, a Holt-Winters model and a long short term memory (LSTM) model, a Prophet model. ) containing at least one of the models
How to provide an automatic piggy bank collection service.
7. The method of claim 6,
Calculating the average error for the first past period of the primary prediction model comprises:
wherein the first prediction model learns about a second past time period, wherein the second past time period is a period from a second past time point that is earlier than the first past time point to the first past time point,
Calculating the average error for the first past period of the secondary prediction model comprises:
The secondary prediction model comprising the step of learning for the second past period
How to provide an automatic piggy bank collection service.
According to claim 1,
Calculating the basic savings amount;
After activating the savings rule, determining whether the saving time is the first saving time;
If the saving time is the first saving time, further comprising the step of setting the saving amount as the basic saving amount
How to provide an automatic piggy bank collection service.
According to claim 1,
After the step of determining the final savings amount,
displaying the final savings amount;
determining whether the savings rule for the savings account is deactivated;
Further comprising the step of determining whether the final balance exceeds the reference balance
How to provide an automatic piggy bank collection service.
A method of providing an automatic piggy bank collection service that automatically transfers a savings amount from a mother account to a savings account,
determining whether the user of the mother account is a high-income user or a low-income user by calculating a basic savings amount;
calculating the current surplus funds;
activating a savings rule for the savings account;
comparing the past spare funds with the current spare funds;
determining whether the analysis data is sufficient when the current spare funds are greater than the past spare funds;
predicting future surplus funds when the analysis data is sufficient;
When the future spare funds are greater than or equal to the current spare funds, the savings amount is increased than the previous savings amount, and when the future spare funds are smaller than the current spare funds, the savings amount is reduced from the previous savings amount, but the increase calculating using a first parameter when the user of the mother account is the high-value user, and calculating using a second parameter different from the first parameter when the user of the mother account is the low-income user;
post-processing the savings amount to determine a final savings amount; and
transferring the savings amount from the mother account to the savings account;
How to provide an automatic piggy bank collection service.
12. The method of claim 11,
The size of the increase is determined by the formula of Equation 1 below,
[Equation 1]
Increment(P, M)=IL * P / (IV1 / (M / IV2) + P)
Here, P is (the current spare fund / the future spare fund-1), M is the current spare fund, the IL, the IV1 and IV2 are preset values, and the IL is the first or first 2 parameters
How to provide an automatic piggy bank collection service.
12. The method of claim 11,
The size of the reduction is determined by the formula of Equation 2 below,
[Equation 2]
Decrease (P, M)=DL * P / ((M/DV2) /DV1) + P)
Here, the P is (the current spare fund / the future spare fund-1), the M is the future spare fund, and the DL, the DV1 and DV2 are preset values.
How to provide an automatic piggy bank collection service.
12. The method of claim 11,
The step of calculating the basic savings amount,
Calculating the exponential smoothing balance according to the simple exponential smoothing of the mother account during the first period;
Calculating the basic savings amount by multiplying the exponential smoothing balance by a preset basic savings ratio
How to provide an automatic piggy bank collection service.
12. The method of claim 11,
The step of calculating the current surplus funds,
Calculating the minimum balance maintained for the second period as a maintenance amount;
calculating a reflected maintenance amount by multiplying the maintenance amount by a preset maintenance amount reflection ratio;
Comprising the step of calculating the current spare funds by subtracting the reflected maintenance amount from the final balance at the current time
How to provide an automatic piggy bank collection service.

Images