Device and method for providing predictive information regarding change to a credit score based on credit scoring item

Abstract

The electronic device may include a processor that acquires the credit score of a target user and the item values ​​of a plurality of credit evaluation items, and for each of the plurality of credit evaluation items, selects a comparison group from among candidate groups in which a plurality of comparison users are grouped, based on the credit score of the target user, and for each of the plurality of credit evaluation items, determines the relevance level of the corresponding credit evaluation item to the credit score of the target user based on the item values ​​of the comparison users belonging to the comparison group and the item value of the target user, selects a target item from among the plurality of credit evaluation items using the relevance levels of the plurality of credit evaluation items, and provides predictive information regarding a change in the credit score based on the selected target item to the target user.

Description

Technology Field

[0001] The following describes a technology that provides predictive information regarding changes in credit scores based on credit evaluation items. Background Technology

[0002] Artificial intelligence (AI) currently plays a significant role in various fields. However, due to the complexity of AI models and their black-box nature, it is often difficult to understand or explain their decision-making processes. This includes credit rating in the financial services sector. means of solving the problem

[0003] A method for providing a prediction regarding a change in a credit score may include: obtaining the credit score of a target user and the item values ​​of a plurality of credit evaluation items; for each of the plurality of credit evaluation items, selecting a comparison group based on the credit score of the target user from among candidate groups in which a plurality of comparison users are grouped; for each of the plurality of credit evaluation items, determining the relevance level of the corresponding credit evaluation item to the credit score of the target user based on the item values ​​of the comparison users belonging to the comparison group and the item value of the target user; selecting a target item from among the plurality of credit evaluation items using the relevance levels of the plurality of credit evaluation items; and providing the target user with prediction information regarding a change in the credit score based on the selected target item.

[0004] The step of selecting the comparison group may include selecting, among the candidate groups, the candidate group with the largest number of comparison users having a credit score in a credit score range including the credit score of the target user as the comparison group.

[0005] The step of selecting the comparison group may include: for a first credit evaluation item among the plurality of credit evaluation items, selecting a first comparison group from among first candidate groups in which the plurality of comparison users are grouped according to the item value of the first credit evaluation item; and for a second credit evaluation item among the plurality of credit evaluation items, selecting a second comparison group from among second candidate groups in which the plurality of comparison users are grouped according to the item value of the second credit evaluation item.

[0006] The step of determining the degree of relevance may include: for each of the plurality of credit evaluation items, determining a comparative credit score based on the credit scores of comparison users belonging to a comparison group selected for the credit evaluation item; selecting a target group to which the target user belongs from among the candidate groups based on the item value of the target user's credit evaluation item; determining a target credit score based on the credit scores of comparison users belonging to the selected target group; and determining the degree of relevance of the credit evaluation item based on the difference between the determined comparative credit score and the target credit score.

[0007] The step of determining the degree of relevance may include: determining a comparative credit score for a first credit evaluation item among the plurality of credit evaluation items based on the credit scores of comparison users belonging to a comparison group selected for the first credit evaluation item; and determining the degree of relevance of a second credit evaluation item among the plurality of credit evaluation items based on the difference between the credit score of the target user and the comparative credit score determined for the first credit evaluation item.

[0008] The step of selecting the comparison group may include: obtaining the credit score of each of the plurality of comparison users and the item values ​​of the plurality of credit evaluation items of the comparison users; for each credit evaluation item, sorting the plurality of comparison users based on the item values ​​of the corresponding credit evaluation items of the plurality of comparison users; and grouping the plurality of comparison users into a plurality of candidate groups in which the number of comparison users included in the candidate groups differs from a threshold number or less.

[0009] The step of obtaining the credit score of the target user and the item values ​​of a plurality of credit evaluation items includes the step of obtaining input from the target user selecting the plurality of credit evaluation items among a plurality of candidate credit evaluation items that affect the user's credit score, and the step of providing the prediction information to the target user may further include the step of providing the target user with a guide explaining the predicted credit score according to a change in at least one of the plurality of credit evaluation items based on the relevance of the plurality of credit evaluation items.

[0010] The step of selecting the target item includes: a step of verifying, using a machine learning model, whether the target user's credit score increases according to a change in the item value of the selected target item; and a step of selecting the target item in response to the successful verification, wherein the machine learning model can be trained to output the user's credit score from input data including the item values ​​of the user's plurality of credit evaluation items.

[0011] The verification step may include: generating changed input data by changing the item value of the target item among the original input data containing the item values ​​of the plurality of credit evaluation items of the target user to a different value; outputting a changed credit score by applying the changed input data to the machine learning model; and comparing the credit score of the target user with the changed credit score.

[0012] The input data of the machine learning model may include the item values ​​of the plurality of credit evaluation items and the item values ​​of additional items.

[0013] The step of selecting the above target item may include: selecting at least one of the credit evaluation items having a correlation of a first sign as a target item; and restricting the selection of a credit evaluation item having a correlation of a second sign opposite to the first sign as a target item.

[0014] The step of providing the above prediction information to the target user may include the step of providing guide information for an increase in the credit score following a change in the item value of the target item.

[0016] The electronic device may include a processor that acquires the credit score of a target user and the item values ​​of a plurality of credit evaluation items, and for each of the plurality of credit evaluation items, selects a comparison group from among candidate groups in which a plurality of comparison users are grouped, based on the credit score of the target user, and for each of the plurality of credit evaluation items, determines the relevance level of the corresponding credit evaluation item to the credit score of the target user based on the item values ​​of the comparison users belonging to the comparison group and the item value of the target user, selects a target item from among the plurality of credit evaluation items using the relevance levels of the plurality of credit evaluation items, and provides predictive information regarding a change in the credit score based on the selected target item to the target user.

[0017] The processor may select, among the candidate groups, the candidate group with the largest number of comparison users having a credit score in a credit score range including the credit score of the target user as the comparison group.

[0018] The processor may, for a first credit evaluation item among the plurality of credit evaluation items, select a first comparison group from among first candidate groups in which the plurality of comparison users are grouped according to the item value of the first credit evaluation item, and for a second credit evaluation item among the plurality of credit evaluation items, select a second comparison group from among second candidate groups in which the plurality of comparison users are grouped according to the item value of the second credit evaluation item.

[0019] The processor can determine a comparative credit score for each of the plurality of credit evaluation items based on the credit scores of comparison users belonging to a comparison group selected for the credit evaluation item, select a target group to which the target user belongs from among the candidate groups based on the item value of the target user's credit evaluation item, determine a target credit score based on the credit scores of comparison users belonging to the selected target group, and determine the relevance of the credit evaluation item based on the difference between the determined comparative credit score and the target credit score.

[0020] The processor can determine a comparative credit score for a first credit evaluation item among the plurality of credit evaluation items based on the credit scores of comparison users belonging to a comparison group selected for the first credit evaluation item, and determine the relevance of a second credit evaluation item among the plurality of credit evaluation items based on the difference between the credit score of the target user and the comparative credit score determined for the first credit evaluation item.

[0021] The processor can obtain the credit scores of each of the plurality of comparison users and the item values ​​of the plurality of credit evaluation items of the comparison users, and for each credit evaluation item, sort the plurality of comparison users based on the item values ​​of the corresponding credit evaluation items of the plurality of comparison users, and group the plurality of comparison users into a plurality of candidate groups in which the number of comparison users included in the candidate groups has a difference of less than or equal to a threshold number.

[0022] The processor can obtain input from the target user selecting a plurality of candidate credit evaluation items among a plurality of candidate credit evaluation items that affect the user's credit score, and provide the target user with a guide explaining the predicted credit score based on a change in at least one of the plurality of credit evaluation items, based on the relevance of the plurality of credit evaluation items.

[0023] The processor verifies using a machine learning model whether the credit score of the target user increases according to a change in the item value of the selected target item, and in response to the successful verification, selects the target item, and the machine learning model can be trained to output the user's credit score from input data including the item values ​​of the user's plurality of credit evaluation items.

[0024] The processor can generate changed input data by changing the item value of the target item among the original input data containing the item values ​​of the plurality of credit evaluation items of the target user to a different value, output a changed credit score by applying the changed input data to the machine learning model, and compare the credit score of the target user with the changed credit score.

[0025] The input data of the machine learning model may include the item values ​​of the plurality of credit evaluation items and the item values ​​of additional items.

[0026] The above processor may select at least one of the credit evaluation items having a correlation of a first sign as a target item, and restrict the selection of a credit evaluation item having a correlation of a second sign opposite to the first sign as a target item.

[0027] The above processor can provide guide information for increasing the credit score based on a change in the item value of the above target item. Brief explanation of the drawing

[0028] FIG. 1 is a flowchart illustrating an example of a method in which an electronic device according to various embodiments provides a prediction regarding a change in a target user's credit score. FIG. 2 is a diagram illustrating an example of an operation in which an electronic device according to various embodiments groups multiple comparison users into multiple candidate groups for credit evaluation items. FIG. 3 is a flowchart illustrating an example of operation in which an electronic device according to various embodiments uses a machine learning model to determine a target item. FIG. 4 is a block diagram showing an example configuration of an electronic device according to various embodiments. FIG. 5 is a block diagram showing an example configuration of a prediction provision system regarding changes in credit scores according to various embodiments. Specific details for implementing the invention

[0029] Specific structural or functional descriptions of the embodiments are disclosed for illustrative purposes only and may be modified and implemented in various forms. Accordingly, actual implementations are not limited to the specific embodiments disclosed, and the scope of this specification includes modifications, equivalents, or substitutions included in the technical concept described by the embodiments.

[0030] Terms such as "first" or "second" may be used to describe various components, but these terms should be interpreted solely for the purpose of distinguishing one component from another. For example, the first component may be named the second component, and similarly, the second component may be named the first component.

[0031] When it is stated that a component is "connected" to another component, it should be understood that it may be directly connected to or joined to that other component, or that there may be other components in between.

[0032] The singular expression includes the plural expression unless the context clearly indicates otherwise. In this specification, terms such as "comprising" or "having" are intended to specify the existence of the described features, numbers, steps, actions, components, parts, or combinations thereof, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0033] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this specification.

[0034] Hereinafter, embodiments will be described in detail with reference to the attached drawings. In the description with reference to the attached drawings, identical components are given the same reference numeral regardless of the drawing number, and redundant descriptions thereof will be omitted.

[0036] FIG. 1 is a flowchart illustrating an example of a method in which an electronic device according to various embodiments provides a prediction regarding a change in a target user's credit score.

[0037] According to one embodiment, an electronic device may provide a guide for raising a target user's credit score. By comparing the target user with at least some of a plurality of comparison users, the electronic device may determine a credit evaluation item among the target user's credit evaluation items for which the target user's credit score may be raised upon a change in the item value. The electronic device may provide a guide to the user regarding the determined credit evaluation item (e.g., target item). In various embodiments of the present disclosure, the target item may refer to a credit evaluation item for which the target user's credit score is predicted to be raised upon a change in the item value of the target item.

[0038] In step (110), the electronic device can obtain the target user's credit score and the item values ​​of a plurality of credit evaluation items. The target user may refer to a user who is the subject of analysis for the method of determining credit evaluation items and providing guidance for raising the credit score.

[0039] Credit evaluation items may represent individual items and / or categories classifying factors that may influence the determination of a target user's credit score. A credit score may mean a score indicating the user's ability to pay for goods traded. For example, a credit score may be determined as one of an integer between 1 and 1000. However, the credit score is not limited to being determined as one of an integer between 1 and 1000, and the credit score may be determined or expressed as a selected grade (e.g., credit rating) among multiple grades based on a predetermined number of score ranges.

[0040] For example, credit evaluation items may include age, gender, income, the number of household members, the number of credit cards, the number of debit cards, the number of loans (also expressed as loan exposure in various embodiments of the present disclosure), the total amount of loan, and the number of delinquent payments. However, credit evaluation items are not limited to the examples described above and may include other items related to the user's credit.

[0041] The item values ​​and credit scores of credit evaluation items may differ depending on the user, and the target items and the amount of change (e.g., increase, decrease) of the item values ​​of the target items may differ depending on the user. For example, even if the first user and the second user have the same or similar credit scores, the first credit evaluation item (e.g., number of loans) may be determined as the target item for the first user, and the second credit evaluation item (e.g., the number of credit cards) may be determined as the target item for the second user.

[0042] In step (120), for each of the plurality of credit evaluation items, the electronic device can select a comparison group based on the target user's credit score from among candidate groups in which a plurality of comparison users are grouped.

[0043] Multiple comparison users may be grouped into multiple candidate groups. A comparison user may refer to another user being compared with a target user. Each comparison user may include item values ​​of multiple credit evaluation items of the comparison user. Multiple comparison users may be grouped into multiple candidate groups based on the item values ​​of each credit evaluation item. The grouping of multiple comparison users into multiple candidate groups is described in more detail later in FIG. 2.

[0044] The electronic device may select a comparison group from among candidate groups. A comparison group may refer to a group representing comparison user(s) having the same or similar credit score as the target user among multiple candidate groups. The comparison group may be used to identify the item values ​​of specific credit evaluation items of comparison users having the same or similar credit score as the target user. For example, the comparison group may include comparison users having item values ​​of specific credit evaluation items that are frequently possessed by comparison users with the same or similar credit score as the target user.

[0045] According to one embodiment, the electronic device may select, among candidate groups, the candidate group having the largest number of comparison users having a credit score within a credit score range that includes the target user's credit score as the comparison group. For example, the electronic device may determine (e.g., count) the number of comparison users among the comparison users included in each candidate group who have a credit score within a credit score range that includes the target user's credit score. The electronic device may select the group having the largest number of comparison users among the candidate groups as the comparison group.

[0046] The credit score range may be pre-set independently of the target user's credit score. For example, if the credit score of a user (e.g., target user, comparison user) is determined to be an integer between 0 and 1000, the credit score range may be determined to be one of 0 to less than 100, 100 to less than 200, 200 to less than 300, ..., 900 to less than 1000 (or 900 to 1000).

[0047] However, the credit score range according to the various embodiments of the present disclosure is not limited to being set independently of the target user's credit score. The credit score range may be set based on the target user's credit score. For example, an electronic device may determine a credit score range centered on the target user's credit score and having a predetermined score length (e.g., 100 points). For example, if the target user's credit score is 570 points, the credit score range including the target user's credit score may be determined to be 520 or greater and less than 570.

[0048] In step (130), the electronic device can determine the relevance level of the corresponding credit evaluation item to the target user's credit score for each of the plurality of credit evaluation items, based on the item values ​​of comparison users belonging to the comparison group and the item values ​​of the target user.

[0049] The relevance of each credit evaluation item may indicate the degree to which the item value of the corresponding credit evaluation item is related to the target user's credit score. The relevance may, for example, indicate the difference between the target user's credit score (or the credit score of the target group) and the credit score of the comparison group. The relevance of each credit evaluation item may be determined based on the comparison group determined for the corresponding credit evaluation item. The target group, as will be described later, may refer to a group selected from among a plurality of candidate groups and represent the target user. The relevance of the credit evaluation items will be described in more detail later in Tables 1 through 3.

[0050] In step (140), the electronic device may select a target item from among a plurality of credit evaluation items by using the correlations of the plurality of credit evaluation items. The target item may refer to a credit evaluation item in which the target user's credit score can be increased when the item value of the target item is changed.

[0051] For example, among the correlations of multiple credit evaluation items, the target item may select a credit evaluation item having the maximum or minimum correlation as the target item. According to one embodiment, if the correlation has a first sign (e.g., negative), it may be defined that the credit score can be increased through a change in the item value of the credit evaluation item. The electronic device may select a credit evaluation item having the minimum correlation as the target item. According to one embodiment, the electronic device may determine one or more credit evaluation items as target items. The electronic device may determine two or more credit evaluation items as target items in order that the correlations of the credit evaluation items have small values. According to one embodiment, the electronic device may determine the target item based on the sign of the correlation. The electronic device may determine at least one of the credit evaluation items having a correlation of the first sign as the target item. The electronic device may limit the determination of a credit evaluation item having a correlation of a second candidate opposite to the first sign as the target item.

[0052] In step (150), the electronic device can provide the target user with predictive information regarding changes in the credit score based on the selected target item.

[0053] Predictive information may include information regarding changes in the credit score resulting from changes in the item values ​​of credit evaluation items. For example, predictive information may include guide information indicating that an increase in the credit score is expected when the item values ​​of the target items are changed.

[0054] In various embodiments of the present disclosure, a plurality of credit evaluation items may be predetermined, for example, by a service provider. However, the plurality of credit evaluation items may be selected based on user input from among a plurality of candidate credit evaluation items set by the service provider. For example, a plurality of candidate credit evaluation items that affect the user's credit score may be set by the service provider. The user may select a credit evaluation item from among the plurality of candidate credit evaluation items for which the user can change the item value. Even if a change to the item value of a credit evaluation item is proposed to the user to increase the credit score, since the feasibility may vary depending on the user, the user may be asked to select a credit evaluation item for which the item value can be changed, and an analysis (e.g., determination of relevance, provision of predictive information) may be performed.

[0055] For example, an electronic device may obtain input from a target user selecting multiple credit evaluation items among multiple candidate credit evaluation items that affect the user's credit score. Based on the relevance of the multiple credit evaluation items, the electronic device may provide the target user with a guide explaining the predicted credit score according to a change in at least one of the multiple credit evaluation items.

[0056] However, in the various embodiments of the present disclosure, the predictive information is not limited to being determined based on an increase in credit rating. According to one embodiment, the predictive information may include information regarding the fact that the credit score does not change by more than a threshold score as a result of a change in the item value of the target item.

[0058] FIG. 2 is a diagram illustrating an example of an operation in which an electronic device according to various embodiments groups multiple comparison users into multiple candidate groups for credit evaluation items.

[0059] An electronic device according to one embodiment can group a plurality of comparison users into a plurality of candidate groups based on each credit evaluation item. As described above in FIG. 1, a comparison group and / or a target group may be selected from among the plurality of candidate groups.

[0060] The electronic device can divide the range of item values ​​of credit evaluation items of comparison users into multiple intervals. In this case, the number of comparison users having item values ​​included in each interval may be the same or similar (e.g., having a difference of less than or equal to a threshold number) across the multiple intervals. Each interval may correspond to each candidate group. Each candidate group may be represented by the minimum and maximum values ​​of the credit evaluation items of the comparison users included within that candidate group, and the item values ​​of the credit evaluation items of the comparison users included within that candidate group may be greater than or equal to the minimum value and less than or equal to the maximum value.

[0061] In step (210), the electronic device can obtain the credit score of each of the multiple comparison users and the item values ​​of the multiple credit evaluation items of the corresponding comparison user.

[0062] For example, the electronic device can obtain the credit scores of each of 100 comparison users (e.g., first comparison user to 100 comparison users) and the item values ​​of 10 credit evaluation items (e.g., first credit evaluation item to tenth credit evaluation item). Specifically, the electronic device can obtain the credit score of each comparison user and the item values ​​of each of the first credit evaluation item to tenth credit evaluation item of the comparison user.

[0063] In step (220), the electronic device can sort multiple comparison users for each credit evaluation item based on the item values ​​of the corresponding credit evaluation item of the multiple comparison users. If the item values ​​of the credit evaluation items of two or more comparison users are the same, the electronic device can sort two or more comparison users based on their credit scores.

[0064] In step (230), the electronic device can group multiple comparison users into multiple candidate groups in which the number of comparison users included in the multiple candidate groups differs from a threshold number.

[0065] According to one embodiment, the electronic device may set comparison users having a predetermined number of upper item values ​​and comparison users having lower item values ​​as separate candidate groups based on the item values ​​of credit evaluation items. A candidate group formed by grouping comparison users having upper or lower item values ​​based on a predetermined number may also be referred to as an outlier group. The electronic device may group the remaining comparison users into multiple candidate groups such that they have a difference of less than or equal to a threshold number. The number of comparison users included in the outlier group and the number of comparison users included in the candidate group may have a difference exceeding the threshold number. However, various embodiments of the present disclosure are not limited to grouping the outlier group into candidate groups, and the electronic device may omit the grouping of the outlier group and group all of the multiple comparison users into multiple candidate groups having a difference of less than or equal to a threshold number.

[0066] For example, the first comparison user to the 100th comparison user can be sorted based on the item value of the first credit evaluation item. Among the first comparison user to the 100th comparison user, five comparison users can be grouped into the first outlier group in order of the highest item value of the first credit evaluation item, and five comparison users can be grouped into the second outlier group in order of the lowest item value. The electronic device can divide the remaining 90 comparison users into 10 candidate groups. The electronic device can divide the 10 candidate groups such that each of them includes nine comparison users.

[0067] The alignment and grouping of comparison users can be performed independently for each credit rating item. For example, the electronic device may group multiple comparison users into multiple candidate groups for other credit rating items (e.g., a second credit rating item, a third credit rating item, ... a tenth credit rating item) independently of grouping multiple comparison users for a first credit rating item into multiple candidate groups. Even if the same multiple comparison users are grouped for each of the first through tenth credit rating items, the grouping results may differ depending on the credit rating item used in each grouping operation.

[0068] For example, regarding a first credit evaluation item among a plurality of credit evaluation items, the electronic device may group a plurality of comparison users into first candidate groups according to the item value of the first credit evaluation item. The electronic device may select a first comparison group from among the first candidate groups regarding the first credit evaluation item. Regarding a second credit evaluation item among a plurality of credit evaluation items, the electronic device may group a plurality of comparison users into second candidate groups according to the item value of the second credit evaluation item. The electronic device may select a second comparison group from among the second candidate groups. As described above, the first candidate groups may be different from the second candidate groups, and the first comparison group may be different from the second comparison group.

[0070] Tables 1 to 3 describe examples of operations in which an electronic device according to various embodiments determines the relevance of credit evaluation items.

[0071] According to one embodiment, the electronic device can determine the relevance of each credit rating item independently of candidate groups of other credit rating items.

[0072] For each of the multiple credit evaluation items, the electronic device can determine a comparative credit score based on the credit scores of comparison users belonging to a selected comparison group for the corresponding credit evaluation item.

[0073] As described above, the comparison group may be determined (e.g., selected) from among a plurality of candidate groups grouped for each credit evaluation item as the group with the largest number of comparison users having credit scores within a credit range including the target user's credit score. The comparison credit score may mean a score representing the credit scores of the comparison users included in the comparison group. The comparison credit score may, for example, be determined based on at least one of the average, mode, or median of the credit scores of the comparison users included in the comparison group. In various embodiments of the present disclosure, the comparison credit score may also be expressed as a major score.

[0074] The electronic device may select a target group to which the target user belongs from among candidate groups based on the item value of the target user's corresponding credit evaluation item. The target group for each credit evaluation item may be selected from among multiple candidate groups grouped for the corresponding credit evaluation item as the candidate group to which the item value of the target user's corresponding credit evaluation item belongs. For example, the electronic device may determine the range to which the item value of the target user's credit evaluation item belongs among the ranges of item values ​​corresponding to each candidate group. The electronic device may determine the candidate group corresponding to the range to which the item value of the target user's credit evaluation item belongs as the target group. The target group may represent comparison users (or a group of comparison users) having item values ​​that are the same as or similar to the item value of the target user's specific credit evaluation item.

[0075] An electronic device may determine a target credit score based on the credit scores of comparison users belonging to a selected target group. The target credit score may mean a score representing the credit scores of comparison users included in the target group. The target credit score may, for example, be determined based on at least one of the average, mode, or median of the credit scores of comparison users included in the target group. In various embodiments of the present disclosure, the target credit score may also be expressed as the marginal score of the target user (or target group).

[0076] An electronic device may determine the relevance of a credit evaluation item based on the difference between a determined comparison credit score and a target credit score. For example, the electronic device may determine the relevance as the value obtained by subtracting the target credit score from the comparison credit score. If the relevance of a specific credit evaluation item is a first sign (e.g., negative), it can be understood that the target group has a higher credit score than the comparison group, and that the specific credit evaluation item of the target user has a negative effect on the target user's credit score. Conversely, if the relevance of a specific credit evaluation item is a second sign (e.g., positive), it can be understood that the target group has a lower credit score than the comparison group, and that the specific credit evaluation item of the target user has a positive effect on the target user's credit score.

[0077] Table 1 describes an example of determining the relevance of each credit rating item independently of candidate groups of other credit rating items.

[0078] Number of loans Number of overdue cases Total loan amount age Compare credit scores 400 600 300 500 Target credit score 300 550 350 600 Relevance -100 -50 50 100

[0079] In Table 1, the correlation is calculated for four credit evaluation items (e.g., number of loans, number of delinquencies, total loan amount, and age). The correlation can be determined by subtracting the target credit score from the comparison credit score. The electronic device may determine the credit evaluation item with the minimum correlation among the credit evaluation items (e.g., number of loans) as the target item. Alternatively, the electronic device may determine the credit evaluation item with negative correlation (e.g., number of loans, number of delinquencies) as the target item.

[0080] The electronic device may provide predictive information based on target items. For example, regarding the number of loans, if the number of loans in the comparison group is 2 and the number of loans in the target group is 3, the electronic device may provide the target user with a prediction that an increase in credit score will occur if the number of loans is changed from 3 to 2. For example, regarding the number of delinquencies, if the number of delinquencies in the comparison group is 1 and the number of delinquencies in the target group is 2, the electronic device may provide the target user with a prediction that an increase in credit score will occur if the number of delinquencies is changed from 2 to 1.

[0081] In various embodiments of the present disclosure, the relevance of credit evaluation items is primarily described as being determined based on the difference between a target credit score and a comparison credit score, but is not limited thereto. According to one embodiment, an electronic device may determine the relevance of each credit evaluation item based on the difference between the target credit score for the corresponding credit evaluation item and the credit score of the target user. For example, the relevance of each credit evaluation item may be determined by subtracting the target user's credit score from the target credit score for the corresponding credit evaluation item. In Table 1, for example, the target user's credit score may be 500 points, the relevance of the number of loans may be determined as -200 points (e.g., (300-500) points), the relevance of the number of delinquencies as 50 points (e.g., (550-500) points), the relevance of the total loan amount as -150 points (e.g., (350-500) points), and the relevance of age as 100 points (e.g., (600-500) points).

[0083] According to one embodiment, the electronic device can determine the relevance of each credit rating item based on candidate groups of other credit rating items.

[0084] The electronic device can determine a comparative credit score for a first credit evaluation item among a plurality of credit evaluation items based on the credit scores of comparison users belonging to a comparison group selected for the first credit evaluation item.

[0085] The electronic device can determine the relevance of a second credit evaluation item among a plurality of credit evaluation items based on the difference between the target user's credit score and the comparison credit score determined for the first credit evaluation item. The electronic device can determine the relevance of the remaining credit evaluation items based on a value derived from the difference between the target user's credit score and the comparison credit score determined for the first credit evaluation item.

[0086] For example, the electronic device can determine the relevance of each credit evaluation item as the average of the differences between the comparative credit score for other credit evaluation items and the target user's credit score. For example, the electronic device can determine the difference between the comparative credit score for each credit evaluation item and the target user's credit score. The electronic device can determine the relevance of one credit evaluation item among multiple credit evaluation items as the average of the differences for the remaining credit evaluation items.

[0087] According to one embodiment, the difference between the target user's credit score and the comparative credit score determined for the first credit evaluation item may be interpreted as being independent (e.g., irrelevant) of the item value of the target user's first credit evaluation item. The difference between the target user's credit score and the comparative credit score determined for the first credit evaluation item may be interpreted as being attributable to a credit evaluation item other than the first credit evaluation item. For example, if a plurality of credit evaluation items includes 10 credit evaluation items, the difference between the target user's credit score and the comparative credit score determined for the first credit evaluation item may be determined to have been caused by the item value(s) of the target user's second credit evaluation item and / or third credit evaluation item and / or tenth credit evaluation item.

[0088] Table 2 shows an example of the difference between the comparison credit score and the credit score for each credit evaluation item, and Table 3 explains an example of the action of determining the relevance of each credit evaluation item based on other credit evaluation items.

[0089] Number of loans Number of overdue cases Total loan amount age Compare credit scores 400 600 900 500 difference 100 -200 -400 0

[0091] Number of loans Number of overdue cases Total loan amount age Number of loans - (500-400) / 3 (500-400) / 3 (500-400) / 3 Number of overdue cases (500-600) / 3 - (500-600) / 3 (500-600) / 3 Total loan amount (500-900) / 3 (500-900) / 3 - (500-900) / 3 age (500-500) / 3 (500-500) / 3 (500-500) / 3 - Relevance -166.67 -100 0 -133.33

[0092] In Tables 2 and 3, the correlation is calculated for four credit evaluation items (e.g., number of loans, number of delinquencies, total loan amount, and age). In Table 2, for each credit evaluation item, the selection of a comparison group and the determination of a comparison credit score may be performed. The difference between the comparison credit score and the target user's credit score for each credit evaluation item may be calculated. According to one embodiment, the difference between the comparison credit score and the target user's credit score may be determined by subtracting the comparison credit score from the target user's credit score. However, it is not limited thereto, and the difference may be determined by subtracting the target user's credit score from the comparison credit score. In the examples of Tables 2 and 3, the comparison credit score for the number of loans may be 400 points, the comparison credit score for the number of delinquencies may be 600 points, the comparison credit score for the total loan amount may be 900 points, and the comparison credit score for age may be 500 points, and the target user's credit score may be 500 points.

[0093] The relevance of each credit evaluation item can be calculated as the average of the differences of other credit evaluation items. In Table 3, for example, the relevance of the number of loans can be calculated as the average (e.g., -166.67) of the difference in the number of delinquencies (e.g., the difference in the comparison credit score for the number of delinquencies of 600 points from the target user's credit score of 500 points), the difference in the total loan amount (e.g., the difference in the comparison credit score for the total loan amount of 900 points from the target user's credit score of 500 points), and the difference in age (e.g., the difference in the comparison credit score for the age of 500 points from the target user's credit score of 500 points).

[0094] The electronic device may determine a credit evaluation item having the minimum correlation among the credit evaluation items (e.g., number of loans) as the target item. Alternatively, the electronic device may determine a credit evaluation item having a negative correlation (e.g., number of loans, number of delinquencies, age) as the target item.

[0095] The electronic device can provide predictive information based on target items. For example, regarding the number of loans, the electronic device can suggest changing the number of loans for a target user. For example, regarding the number of delinquencies, the electronic device can suggest changing the number of delinquencies.

[0096] For example, the electronic device may suggest changing the item value of the target item from the item value of the target group to the item value of the comparison group.

[0097] For example, the electronic device may determine a comparison group among a plurality of candidate groups grouped with respect to a target item, from among candidate groups in which the representative values ​​(e.g., mean, mode, median) of the credit scores of comparison users are higher than the credit score of the target user (hereinafter referred to as 'candidate groups satisfying conditions'). If candidate groups satisfying conditions exist, the electronic device may determine as the comparison group a candidate group having a comparison user that has a value most similar to the item value of the target item of the target user among the candidate groups. The electronic device may propose changing the item value of the target item to the item value of the comparison group.

[0099] FIG. 3 is a flowchart illustrating an example of operation in which an electronic device according to various embodiments uses a machine learning model to determine a target item.

[0100] An electronic device according to one embodiment can verify target items using a machine learning model. The machine learning model may refer to a model trained to output a user's credit score from input data including item values ​​of a plurality of credit evaluation items of the user. In various embodiments of the present disclosure, the machine learning model may also be expressed as a 'credit score estimation model'.

[0101] In step (310), the electronic device can use a machine learning model to verify whether the target user's credit score increases as a result of a change in the item value of the selected target item.

[0102] An electronic device can validate a target item by applying the modified input data to a machine learning model. For example, the electronic device can generate changed input data by changing the item value of the target item within the original input data to a different value. The original input data may include item values ​​of multiple credit evaluation items of the target user. The changed input data may refer to input data in which the item value of the target item within the original input data has been changed to a different value. The item value of the target item in the original input data may be changed to an item value of the comparison group. The item value of the comparison group may refer to an item value that falls within the range of item values ​​of the target item of the comparison group. For example, it may include representative values ​​(e.g., mean, median, mode) of the item values ​​of the target item of the comparison users included in the comparison group.

[0103] The electronic device can output a modified credit score by applying a machine learning model to the modified input data. The electronic device can compare the target user's credit score (e.g., original credit score) with the modified credit score. For example, the electronic device can determine that the target item is validated if the target user's credit score is greater than the modified credit score. The electronic device can determine that the validation of the target item has failed if the target user's credit score is equal to or less than the modified credit score. For example, the electronic device can determine that the target item is validated if the target user's credit score is greater than the modified credit score by more than a threshold score.

[0104] In step (320), the electronic device may select a target item in response if verification is successful. The electronic device may cancel the selection of the target item in response if verification of the target item fails. If the selection of the target item is canceled, the electronic device may select another credit evaluation item as the target item.

[0105] According to one embodiment, the input data of a machine learning model may include item values ​​of a plurality of credit evaluation items and item values ​​of additional items. Additional items may refer to credit evaluation items different from the plurality of credit evaluation items used for relevance calculation. The input data of the machine learning model may include item values ​​of more credit evaluation items than the credit evaluation items used for relevance calculation. A user (e.g., a service provider, a service user) may select a credit evaluation item to be used to provide predictive information to the user from among the credit evaluation items (e.g., a plurality of credit evaluation items and additional items) included in the input data of the machine learning model. Instead of calculating the relevance of all of the large number of credit evaluation items used as input data for the machine learning model, the electronic device may select and analyze credit evaluation items of interest to the user and provide the analysis results to the user.

[0107] FIG. 4 is a block diagram showing an example configuration of an electronic device according to various embodiments.

[0108] According to one embodiment, the electronic device (410) may include a processor (411), a memory (412), a communication unit (413), and an output unit (414).

[0109] The processor (411) can obtain the target user's credit score and multiple credit evaluation item values. The processor (411) can select a comparison group for each credit evaluation item. The processor (411) can determine the relevance of the multiple credit evaluation items. The processor (411) can select a target item among the multiple credit evaluation items and provide prediction information based on the target item to the target user.

[0110] Memory (412) may temporarily and / or permanently store at least one target item among a target user, a plurality of credit evaluation items, the target user's credit score, the item values ​​of the target user's plurality of credit evaluation items, comparison users, a plurality of candidate groups, and relevance. Memory (412) may store instructions for obtaining the target user's credit score and item values, selecting comparison groups, relevances of the plurality of credit evaluation items, selecting target items, and / or providing prediction information. However, this is purely an example, and the information stored in memory (412) is not limited to this.

[0111] The communication unit (413) can transmit and receive at least one of the prediction information, including the target user's credit score and item values, multiple candidate groups, comparison groups, relevance, etc., to an external device (e.g., another electronic device, server). The communication unit (413) can establish a wired communication channel and / or a wireless communication channel with the external device (e.g., another electronic device, server), and, for example, can establish communication with the external device through a long-distance communication network such as cellular communication, short-range wireless communication, LAN (local area network) communication, Bluetooth, WiFi (wireless fidelity) direct or IrDA (infrared data association), legacy cellular network, 4G and / or 5G network, next-generation communication, the Internet, or a computer network (e.g., LAN or WAN).

[0112] The output unit (414) may provide feedback to the user regarding the prediction information. For example, the output unit (414) may include a display and may display a visual representation indicating the prediction information. For example, the output unit (414) may include a speaker and may play a sound indicating the prediction information.

[0114] FIG. 5 is a block diagram showing an example configuration of a prediction provision system regarding changes in credit scores according to various embodiments.

[0115] According to one embodiment, a method for providing a prediction regarding a change in a credit score may be performed by a plurality of electronic devices (also referred to as a system for providing a prediction regarding a credit score in various embodiments of the present disclosure). A system for providing a prediction according to one embodiment may include a server (510) and a user terminal (520).

[0116] The server (510) can provide services to the user terminal (520). For example, the server (510) can provide a service (e.g., credit score management service) to the user terminal (520) that provides information on the determination (e.g., prediction, calculation) of the credit score and / or the change of the credit score.

[0117] The user terminal (520) can be any electronic device capable of installing and executing a service application related to the server (510), such as a computer, portable computer, wireless phone, mobile phone, smartphone, PDA (Personal Digital Assistants), or web tablet. At this time, the user terminal (520) can perform overall service operations, such as configuring the service screen, inputting data, transmitting and receiving data, and storing data, under the control of the application. For example, the user terminal (520) can access the service provided by the server (510) through the application.

[0118] A server (510) according to one embodiment may include a processor (511), memory (512), and a communication unit (513).

[0119] The processor (511) can obtain (e.g., receive) the credit score of the target user and the item values ​​of a plurality of credit evaluation items from the user terminal (520). The processor (511) can select a comparison group for each credit evaluation item. The processor (511) can determine the relevance of the plurality of credit evaluation items. The processor (511) can provide the target user by selecting a target item among the plurality of credit evaluation items and transmitting prediction information based on the target item to the user terminal (520).

[0120] Memory (512) may temporarily and / or permanently store at least one target item among a target user, a plurality of credit evaluation items, the target user's credit score, the item values ​​of the target user's plurality of credit evaluation items, comparison users, a plurality of candidate groups, and relevance. Memory (512) may store instructions for obtaining the target user's credit score and item values, selecting comparison groups, relevances of the plurality of credit evaluation items, selecting target items, and / or providing prediction information. However, this is purely an example, and the information stored in memory (512) is not limited to this.

[0121] The communication unit (513) can transmit and receive the target user's credit score and item values ​​and / or prediction information to and from an external device (e.g., user terminal (520)). The communication unit (513) can establish a wired communication channel and / or a wireless communication channel with the external device (e.g., user terminal (520)), and, for example, can establish communication with the external device through a long-distance communication network such as cellular communication, short-range wireless communication, LAN (local area network) communication, Bluetooth, WiFi (wireless fidelity) direct or IrDA (infrared data association), legacy cellular network, 4G and / or 5G network, next-generation communication, the Internet, or a computer network (e.g., LAN or WAN).

[0122] A user terminal (520) according to one embodiment may include a user input acquisition unit (521), a processor (522), a memory (523), a communication unit (524), and an output unit (525).

[0123] The user input acquisition unit (521) can acquire user input from a target user. The user input may include, for example, the target user's credit score and item values ​​of a plurality of credit evaluation items, a trigger input for a decision action for prediction information, and / or an input regarding the designation of a target item.

[0124] The processor (522) can transmit the target user's credit score and / or the item values ​​of the target user's credit evaluation items from the target user to the server (510). The processor (522) can receive prediction information from the server (510).

[0125] The memory (523) may temporarily and / or permanently store at least one of the following: a target user, a plurality of credit evaluation items, the target user's credit score, and prediction information. The memory (523) may store instructions for obtaining user input, obtaining information regarding the target user, and / or outputting prediction information. However, this is purely an example and the information stored in the memory (523) is not limited to this.

[0126] The communication unit (524) can transmit and receive information regarding user input, information regarding target users, and / or prediction information with an external device (e.g., server (510)). The communication unit (524) can establish a wired communication channel and / or a wireless communication channel with an external device (e.g., server (510)), and, for example, can establish communication with the external device through a long-distance communication network such as cellular communication, short-range wireless communication, LAN (local area network) communication, Bluetooth, WiFi (wireless fidelity) direct or IrDA (infrared data association), legacy cellular network, 4G and / or 5G network, next-generation communication, the Internet, or a computer network (e.g., LAN or WAN).

[0127] The output unit (525) may provide feedback to the user regarding the prediction information. For example, the output unit (525) may include a display and may display a visual representation indicating the prediction information. For example, the output unit (525) may include a speaker and may play a sound indicating the prediction information.

[0129] The embodiments described above may be implemented as hardware components, software components, and / or combinations of hardware and software components. For example, the devices, methods, and components described in the embodiments may be implemented using a general-purpose computer or a special-purpose computer, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing unit may execute an operating system (OS) and software applications executed on said operating system. Additionally, the processing unit may access, store, manipulate, process, and generate data in response to the execution of the software. For ease of understanding, the processing unit may be described as being used as a single unit, but those skilled in the art will understand that the processing unit may include multiple processing elements and / or multiple types of processing elements. For example, the processing unit may include multiple processors or one processor and one controller. In addition, other processing configurations, such as parallel processors, are also possible.

[0130] Software may include computer programs, code, instructions, or a combination of one or more of these, and may configure a processing unit to operate as desired or command the processing unit independently or collectively. Software and / or data may be permanently or temporarily embodied in any type of machine, component, physical device, virtual equipment, or computer storage medium or device so as to be interpreted by the processing unit or to provide instructions or data to the processing unit. Software may be distributed over networked computer systems and may be stored or executed in a distributed manner. Software and data may be stored on computer-readable recording media.

[0131] The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc., either alone or in combination, and the program instructions recorded on the medium may be those specifically designed and configured for the embodiment or may be those known and available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and hardware devices specifically configured to store and execute program instructions, such as ROMs and flash memory. Examples of program instructions include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc.

[0132] The hardware device described above may be configured to operate as one or more software modules to perform the operation of the embodiment, and vice versa.

[0133] In this document, each of the phrases such as "A or B", "at least one of A and B", "at least one of A or B", "A, B or C", "at least one of A, B and C", and "at least one of A, B, or C" may include any one of the items listed together in the corresponding phrase, or all possible combinations thereof.

[0134] Although the embodiments have been described above with reference to the limited drawings, those skilled in the art can apply various technical modifications and variations based thereon. For example, suitable results may be achieved even if the described techniques are performed in a different order than described, and / or if the components of the described system, structure, device, circuit, etc. are combined or assembled in a form different from described, or replaced or substituted by other components or equivalents.

[0135] Therefore, other implementations, other embodiments, and equivalents to the claims also fall within the scope of the claims set forth below.

Claims

Claim 1 A method for providing a prediction regarding a change in credit score performed by an electronic device comprises: obtaining credit scores and item values ​​of credit evaluation items for a plurality of comparison users and a target user; grouping the plurality of comparison users into a plurality of candidate groups based on the item values ​​of the credit evaluation items of the plurality of comparison users; - each candidate group corresponds to a range of item values ​​of the credit evaluation items; - among the candidate groups, selecting the candidate group with the largest number of comparison users having credit scores within a credit score range including the target user's credit score as a comparison group; determining a comparison credit score representing the credit scores of the comparison users belonging to the comparison group; among the candidate groups, selecting a candidate group corresponding to a range to which the item value of the credit evaluation item of the target user belongs as a target group; determining a target credit score representing the credit scores of the comparison users belonging to the target group; and determining a relevance level of the credit evaluation item representing the degree to which the item value of the credit evaluation item of the target user is related to the target user's credit score based on the difference between the comparison credit score and the target credit score. Claim 2 A method for providing a prediction regarding a change in credit score according to claim 1, wherein the step of selecting a comparison group comprises: determining a credit score range centered on the credit score of the target user and having a predetermined score length; counting the number of comparison users among the comparison users included in each candidate group who have a credit score belonging to the credit score range; and selecting the candidate group with the largest number of counted comparison users among the plurality of candidate groups as the comparison group. Claim 3 A method for providing a prediction regarding a change in a credit score according to claim 1, wherein the credit evaluation item comprises a plurality of credit evaluation items, and the method further comprises the steps of: for each of the plurality of credit evaluation items, performing a selection of a comparison group, a determination of a comparison credit score, a selection of a target group, a determination of a target credit score, and a determination of a correlation; among the plurality of credit evaluation items, selecting a target item using the correlations of the plurality of credit evaluation items; and providing a prediction information regarding a change in a credit score based on the selected target item to the target user. Claim 4 In claim 3, the step of performing, for each of the plurality of credit evaluation items, the selection of a comparison group, the determination of a comparison credit score, the selection of a target group, the determination of a target credit score, and the determination of a correlation comprises: for a first credit evaluation item among the plurality of credit evaluation items, the step of selecting a first comparison group from among first candidate groups in which the plurality of comparison users are grouped according to the item value of the first credit evaluation item; for a second credit evaluation item among the plurality of credit evaluation items, the step of selecting a second comparison group from among second candidate groups in which the plurality of comparison users are grouped according to the item value of the second credit evaluation item; for a first credit evaluation item among the plurality of credit evaluation items, the step of determining a comparison credit score based on the credit scores of comparison users belonging to the comparison group selected for the first credit evaluation item; and the step of determining the correlation of the second credit evaluation item among the plurality of credit evaluation items based on the difference between the credit score of the target user and the comparison credit score determined for the first credit evaluation item. Claim 5 delete Claim 6 A method for providing a prediction regarding a change in credit score according to claim 1, wherein the step of grouping into a plurality of candidate groups comprises: a step of sorting the plurality of comparison users based on the item values ​​of the credit evaluation items of the plurality of comparison users; and a step of grouping the plurality of comparison users into the plurality of candidate groups such that the number of comparison users included in the plurality of candidate groups has a difference of less than or equal to a threshold number. Claim 7 A method for providing a prediction regarding a change in credit score, wherein, in claim 1, the step of obtaining credit scores of a plurality of comparison users and a target user and item values ​​of a credit evaluation item includes the step of obtaining input from the target user selecting said credit evaluation item among a plurality of candidate credit evaluation items that affect the user's credit score, and the method further includes the step of providing a guide to the target user explaining the credit score predicted according to a change in said credit evaluation item based on the relevance of said credit evaluation item. Claim 8 In paragraph 3, the step of selecting the target item comprises: a step of verifying using a machine learning model whether the credit score of the target user increases according to a change in the item value of the selected target item; and a step of selecting the target item in response to the successful verification, wherein the machine learning model is trained to output the credit score of the user from input data including the item values ​​of the user's plurality of credit evaluation items, a method for providing a prediction regarding a change in credit score. Claim 9 In claim 8, the verification step comprises: generating changed input data by changing the item value of the target item among original input data including item values ​​of the plurality of credit evaluation items of the target user to a different value; outputting a changed credit score by applying the changed input data to the machine learning model; and comparing the credit score of the target user with the changed credit score, a method for providing a prediction regarding a change in a credit score. Claim 10 A method for providing a prediction regarding a change in credit score according to claim 8, wherein the input data of the machine learning model includes item values ​​of the plurality of credit evaluation items and item values ​​of additional items. Claim 11 A method for providing a prediction regarding a change in credit score, wherein, in paragraph 3, the step of selecting the target item comprises: a step of selecting at least one of the credit evaluation items having a correlation of a first sign as a target item; and a step of restricting the selection of a credit evaluation item having a correlation of a second sign opposite to the first sign as a target item. Claim 12 A method for providing a prediction regarding a change in credit score, wherein, in paragraph 3, the step of providing the prediction information to the target user includes the step of providing guide information for an increase in the credit score following a change in the item value of the target item. Claim 13 A computer-readable recording medium storing one or more computer programs comprising instructions for performing the method of any one of paragraphs 1 through 4 and paragraphs 6 through 12. Claim 14 An electronic device comprising a processor that obtains credit scores and item values ​​of credit evaluation items for a plurality of comparison users and a target user, and groups the plurality of comparison users into a plurality of candidate groups based on the item values ​​of the credit evaluation items of the plurality of comparison users; - each candidate group corresponds to a range of item values ​​of the credit evaluation items; - among the candidate groups, select the candidate group with the largest number of comparison users having credit scores in a credit score range including the credit score of the target user as the comparison group, and determine a comparison credit score representing the credit scores of the comparison users belonging to the comparison group; among the candidate groups, select a candidate group corresponding to a range to which the item value of the credit evaluation item of the target user belongs as the target group, and determine a target credit score representing the credit scores of the comparison users belonging to the target group; and determine a relevance level of the credit evaluation item representing the degree to which the item value of the credit evaluation item of the target user is related to the credit score of the target user based on the difference between the comparison credit score and the target credit score. Claim 15 An electronic device according to claim 14, wherein the processor determines a credit score range having a predetermined score length centered on the credit score of the target user, counts the number of comparison users among the comparison users included in each candidate group who have a credit score belonging to the credit score range, and selects the candidate group with the largest number of counted comparison users among the plurality of candidate groups as the comparison group. Claim 16 An electronic device according to claim 14, wherein the credit evaluation item comprises a plurality of credit evaluation items, and the processor performs, for each of the plurality of credit evaluation items, the selection of a comparison group, the determination of a comparison credit score, the selection of a target group, the determination of a target credit score, and the determination of a correlation, and selects a target item from among the plurality of credit evaluation items using the correlations of the plurality of credit evaluation items, and provides predictive information regarding a change in the credit score based on the selected target item to the target user. Claim 17 An electronic device according to claim 16, wherein the processor selects a first comparison group from among first candidate groups in which the plurality of comparison users are grouped according to the item value of the first credit evaluation item for a first credit evaluation item among the plurality of credit evaluation items, selects a second comparison group from among second candidate groups in which the plurality of comparison users are grouped according to the item value of the second credit evaluation item for a second credit evaluation item among the plurality of credit evaluation items, determines a comparison credit score based on the credit scores of comparison users belonging to the comparison group selected for the first credit evaluation item for a first credit evaluation item among the plurality of credit evaluation items, and determines the degree of relevance of the second credit evaluation item among the plurality of credit evaluation items based on the difference between the credit score of the target user and the comparison credit score determined for the first credit evaluation item. Claim 18 delete Claim 19 In paragraph 14, the processor is an electronic device that sorts the plurality of comparison users based on the item values ​​of the credit evaluation items of the plurality of comparison users and groups the plurality of comparison users into a plurality of candidate groups such that the number of comparison users included in the plurality of candidate groups has a difference of less than or equal to a threshold number. Claim 20 An electronic device according to claim 14, wherein the processor obtains input from the target user selecting a credit evaluation item among a plurality of candidate credit evaluation items that affect the user's credit score, and provides the target user with a guide explaining the credit score predicted according to the change in the credit evaluation item based on the relevance of the credit evaluation item. Claim 21 An electronic device according to claim 16, wherein the processor verifies using a machine learning model whether the credit score of the target user increases according to a change in the item value of the selected target item, and in response to the successful verification, selects the target item, and the machine learning model is trained to output the user's credit score from input data including the item values ​​of the user's plurality of credit evaluation items. Claim 22 An electronic device according to claim 21, wherein the processor generates changed input data by changing the item value of the target item among the original input data containing the item values ​​of the plurality of credit evaluation items of the target user to a different value, outputs a changed credit score by applying the changed input data to the machine learning model, and compares the credit score of the target user with the changed credit score. Claim 23 In paragraph 21, the input data of the machine learning model comprises item values ​​of the plurality of credit evaluation items and item values ​​of additional items, an electronic device. Claim 24 An electronic device according to claim 16, wherein the processor selects at least one of the credit evaluation items having a correlation of a first sign as a target item, and restricts the selection of a credit evaluation item having a correlation of a second sign opposite to the first sign as a target item. Claim 25 In paragraph 16, the processor is an electronic device that provides guidance information for raising a credit score in accordance with a change in the item value of the target item.

Images