JP2009163628A - Loan examination device and computer program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a loan in a further adequate loan condition matched to each individual by performing examination including the personal quality for money and consumption of a customer. <P>SOLUTION: A loan examination device includes, in addition to a credit examination means 12 for determining a loan condition based on attribute and credit information of a customer, a load aptitude information receiving means 14 which receives loan aptitude information of the customer transmitted from an automated loan dispensing machine and a loan condition correction means 15 which corrects, based on the loan aptitude information of the customer, the loan condition determined by the credit examination means 12. Since determination of the loan condition is performed based on not only an objective fact such as sex, age, annual income, or delinquent payment information or the like, but also the loan aptitude information of the customer, a loan matched to the trend or quality for money, consumption or the like of each customer can be provided. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a loan screening apparatus and a computer program that perform screening at the time of a loan contract.

  For example, in a loan contract using an automatic contract machine, a customer responds to examination items such as name, date of birth, address, etc., and sends the information to the computer of the management center of the financial institution. After reviewing the conditions (credit judgment), returning the result of the review to the automatic contract machine and accepting the result of the review under the relevant loan terms, a contract document will be created and a cash withdrawal card will be issued. . In other words, current loan contracts are based only on facts that can be grasped objectively, such as individual customer attributes (age, occupation, etc.) and credit information (other borrowing information, borrowing history such as delinquency information). We are examining. Accordingly, if there are a plurality of persons whose facts such as age, occupation, annual income, credit information, etc. are the same, the loan conditions of those persons are the same.

On the other hand, Patent Document 1 discloses a technology that adds physical and mental aptitude for driving to insurance premium calculation conditions in addition to objective conditions such as vehicle type, age, accident history, etc., for insurance premium payment. Yes.
JP 2004-102801 A

  As described above, loan contracts at financial institutions are currently examined only for examination items that can be objectively grasped. There is no review that takes into account the personal qualities of the customer, such as the sense of money or consumption.

  The present invention has been made in view of the above, and when making a loan contract using a contracting terminal device such as an automatic contract machine, an examination including personal qualities related to customer's personal money, consumption, etc. is performed. It is an object of the present invention to provide a loan examination apparatus and a computer program that can be executed and can be financed under more appropriate loan conditions suitable for each individual.

In order to solve the above-mentioned problem, the present invention according to claim 1 is a loan examination device for examining a loan to a customer,
Credit review means for determining loan conditions based on the customer attributes and credit information;
Loan aptitude information receiving means for receiving the loan aptitude information of the customer transmitted from the terminal device of the customer;
There is provided a loan examination device comprising: loan condition correction means for correcting the loan condition obtained by the credit examination means based on the loan suitability information of the customer received by the loan suitability information receiving means. .
In this invention of Claim 2, it comprises the loan suitability inquiry means which transmits the inquiry information for pulling out the loan suitability information of the customer to the terminal device of the customer,
2. The loan examination apparatus according to claim 1, wherein the loan suitability information receiving means is configured to receive a customer response to the inquiry information transmitted from the terminal device as the loan suitability information. .
In this invention of Claim 3, the said loan suitability inquiry means is set so that inquiry information including at least one of the inquiry information reflected in the loan limit and the inquiry information reflected in the loan interest rate can be transmitted. And
The loan suitability information receiving means is at least one of loan suitability information reflected in a loan limit and loan suitability information reflected in a loan interest rate as a response to each inquiry information transmitted from the loan suitability inquiry means. The loan examination apparatus according to claim 2, wherein the loan suitability information including one can be received.
In this invention of Claim 4, the said loan condition correction means is the loan suitability information reflected by the loan limit amount received by the said loan suitability information receiving means, and the loan suitability information reflected by the loan interest rate. The loan examination apparatus according to any one of claims 1 to 3, wherein the loan condition obtained by the credit examination unit is corrected based on the loan suitability information including at least one of them.
In this invention of Claim 5, the said loan aptitude information is shown by a score for every question of the said inquiry information,
The loan condition correcting means includes
A credential for obtaining a credibility as an answer to each inquiry information for at least one of the loan aptitude information reflected in the loan limit received by the loan aptitude information receiving means and the loan aptitude information reflected in the loan interest rate While having a degree calculation means,
A limit correction point is obtained by using the reliability of the loan suitability information reflected in the loan limit amount obtained from the credit calculation means, and the limit correction point is set to the loan limit amount obtained by the credit examination means. The interest rate correction point is calculated using the adjusted loan limit calculation means for calculating the adjusted loan limit in consideration and the reliability of the loan suitability information reflected in the loan interest rate obtained from the reliability calculation means. 5. The loan examination apparatus according to claim 4, further comprising at least one means of a modified loan interest rate calculation means for obtaining a modified loan interest rate by adding points to the loan interest rate obtained by the credit examination means. To do.
In this invention of Claim 6, the said reliability calculation means is
The reliability of at least one of the loan suitability information reflected in the loan limit received by the loan suitability information receiving means and the loan suitability information reflected in the loan interest rate from the score for each question of each loan suitability information A first step for determining a coefficient;
Using the confidence factor obtained in the first step, at least one of confidence for the loan suitability information reflected in the loan limit and confidence for the loan suitability information reflected in the loan interest rate. And providing a loan screening device according to claim 5.
In the present invention described in claim 7, both the inquiry information reflected in the loan limit amount and the inquiry information reflected in the loan interest rate, which are transmitted from the loan suitability inquiry means, directly relate to one question item. The loan examination apparatus according to any one of claims 2 to 6, wherein two types of questions, a question using a general expression and a question using an indirect expression, are set.
In the present invention according to claim 8, the question using the direct expression and the question using the indirect expression transmitted from the loan suitability inquiry means are transmitted at a predetermined time interval. The loan examination apparatus according to claim 7 is provided.
In the present invention as set forth in claim 9, in the second step of the reliability calculation means, the reliability of the loan aptitude information reflected in the loan limit and the loan aptitude information reflected in the loan interest rate When obtaining at least one confidence level, the confidence coefficient corresponding to the question using the indirect expression is higher than the confidence coefficient corresponding to the question using the direct expression. The loan examination apparatus according to any one of claims 6 to 8, wherein weighting is performed so that the degree of influence on the degree increases.
In this invention of Claim 10, It is a computer program introduced into the loan examination device which examines the loan to a customer,
Credit review means for determining loan conditions based on the customer attributes and credit information;
Loan aptitude information receiving means for receiving the loan aptitude information of the customer transmitted from the terminal device of the customer;
There is provided a computer program comprising: loan condition correcting means for correcting the loan condition obtained by the credit review means based on the loan suitability information of the customer received by the loan suitability information receiving means.
In this invention of Claim 11, the loan aptitude inquiry means which transmits the inquiry information for drawing out the loan aptitude information of the said customer to the said terminal device of a customer is provided,
11. The computer program according to claim 10, wherein the loan suitability information receiving unit is configured to receive a customer response to the inquiry information transmitted from the terminal device as the loan suitability information.
In the present invention according to claim 12, the loan suitability inquiry means is set to be capable of transmitting inquiry information including at least one of inquiry information reflected in a loan limit and inquiry information reflected in a loan interest rate. And
The loan suitability information receiving means is at least one of loan suitability information reflected in a loan limit and loan suitability information reflected in a loan interest rate as a response to each inquiry information transmitted from the loan suitability inquiry means. 12. The computer program according to claim 11, wherein the computer can receive loan suitability information including one.
In this invention of Claim 13, the said loan condition correction means is the loan suitability information reflected by the loan limit amount received by the said loan suitability information receiving means, and the loan suitability information reflected by the loan interest rate. The computer program according to any one of claims 10 to 12, wherein the loan condition obtained by the credit examination means is corrected based on loan suitability information including at least one of them.
In the present invention according to claim 14, the loan suitability information is indicated by a score for each question of the inquiry information,
The loan condition correcting means includes
A credential for obtaining a credibility as an answer to each inquiry information for at least one of the loan aptitude information reflected in the loan limit received by the loan aptitude information receiving means and the loan aptitude information reflected in the loan interest rate While having a degree calculation means,
A limit correction point is obtained by using the reliability of the loan suitability information reflected in the loan limit amount obtained from the credit calculation means, and the limit correction point is set to the loan limit amount obtained by the credit examination means. The interest rate correction point is calculated using the adjusted loan limit calculation means for calculating the adjusted loan limit in consideration and the reliability of the loan suitability information reflected in the loan interest rate obtained from the reliability calculation means. 14. The computer program according to claim 13, further comprising at least one means of a modified loan interest rate calculating means for obtaining an adjusted loan interest rate by adding points to the loan interest rate obtained by the credit examination means. .
In this invention of Claim 15, the said reliability calculation means is
The reliability of at least one of the loan suitability information reflected in the loan limit received by the loan suitability information receiving means and the loan suitability information reflected in the loan interest rate from the score for each question of each loan suitability information A first step for determining a coefficient;
Using the confidence factor obtained in the first step, at least one of confidence for the loan suitability information reflected in the loan limit and confidence for the loan suitability information reflected in the loan interest rate. 15. A computer program according to claim 14, further comprising a second step of obtaining one.
In the present invention described in claim 16, the inquiry information reflected in the loan limit amount and the inquiry information reflected in the loan interest rate, which are transmitted from the loan suitability inquiry means, are all directly related to one question item. The computer program according to any one of claims 11 to 15, wherein two types of questions, a question using a general expression and a question using an indirect expression, are set.
In the present invention according to claim 17, the question using the direct expression and the question using the indirect expression transmitted from the loan suitability inquiry means are transmitted at predetermined intervals. A computer program according to claim 16 is provided.
In the present invention as set forth in claim 18, in the second step of the reliability calculation means, the reliability of the loan aptitude information reflected in the lending limit and the loan aptitude information reflected in the loan interest rate. When obtaining at least one confidence level, the confidence coefficient corresponding to the question using the indirect expression is higher than the confidence coefficient corresponding to the question using the direct expression. The computer program according to any one of claims 15 to 17, wherein the computer program is obtained by weighting so that the degree of influence on the degree is increased.

  The present invention includes a loan aptitude information receiving means for receiving customer loan aptitude information transmitted from a customer terminal device, in addition to a credit screening means for determining loan conditions based on customer attributes and credit information, and a customer Based on the loan suitability information, the loan condition correcting means for correcting the loan condition obtained by the credit review means. The determination of financing conditions takes into account not only objective facts such as gender, age, annual income, and repayment information, but also information on the suitability of the customer's lending. You can make a loan. In other words, in the credit screening based on objective facts, even if the conditions of multiple customers are the same, for example, customers who are solid and have a high economic concept can be considered by adding the loan aptitude information of each customer. In this case, it is possible to obtain loans with more appropriate conditions suitable for each customer, such as obtaining more favorable loan conditions than those with wasteful consumers and low economic ideas.

  Hereinafter, embodiments of the present invention will be described in more detail with reference to the drawings. FIG. 1 is a diagram showing a schematic configuration of a loan examination apparatus 10 according to an embodiment of the present invention. The loan examination apparatus 10 is composed of a computer including a control unit 11 such as a CPU, is installed in a management center of a financial institution, and is connected to the automatic contract machine 20 via communication means. The automatic contracting machine 20 makes a loan contract without the customer going to the store window of a financial institution, an information input means for inputting customer attributes and the like, a scanner for reading identity verification documents and contracts, etc. (Not shown), a contract collection unit (not shown), and the like. The automatic contract machine 20 is installed and managed by the financial institution side, but is connected to the loan examination apparatus 10 via a communication means and used by a customer, and is used as a customer terminal apparatus. As the terminal device, not only the automatic contract machine 20 but also an information terminal such as a personal computer owned by a customer can be used as a terminal device by connecting to the loan examination device 10 via the Internet or a dedicated line.

  Computer programs such as credit review means 12, loan suitability inquiry means 13, loan suitability information receiving means 14, and loan condition correction means 15 are installed in the storage unit of the computer constituting the loan review apparatus 10. In addition, a customer database (customer DB) 16 in which customer information such as customer attributes is recorded is provided. The computer program can be provided by being stored in a recording medium such as a flexible disk, a hard disk, a CD-ROM, an MO (magneto-optical disk), a DVD-ROM, or can be transmitted through a communication line. .

  The credit examination means 12 determines the loan conditions based on the customer attributes and credit information. Specifically, the credit examination unit 12 includes an attribute information receiving unit 121, a credit information inquiry unit 122, and a determination unit 123. The customer attributes are name, gender, age, occupation, annual income, workplace, etc., and when these attributes are input by the customer in the automatic contract machine 20, the attribute information is received via the communication means. 121. Based on the attribute information, the attribute information receiving unit 121 accesses the customer database 16 and identifies the customer.

  The attribute information receiving unit 121 further passes the received attribute information to the credit information inquiry unit 122. The credit information inquiry means 122 makes an inquiry to the credit information institution 30 using the customer attribute information, and receives the credit information of the customer (other borrowing information, repayment arrears information, etc.). The judging unit 123 uses the customer attribute information received by the attribute information receiving unit 121 and the customer credit information received by the credit information inquiry unit 122 to provide a loan for the customer borrowing application via the automatic contractor 20. Determine the conditions. Loan conditions using customer attributes and credit information are generally determined for each financial institution that executes the loan. For example, in the case of annual income: 800 to 10 million yen, age: 40's, number of delinquent repayments: 0 times, etc., the lending limit is 300,000 yen, loan interest rate: 14%, or annual income: 300 In the case of -5 million yen, age: 20's, repayment delinquency: 3 times, etc., the loan limit that can be lent is 100,000 yen and the loan interest rate is 18%. For this reason, the determination means 123 applies a customer's attribute and credit information to these, and determines loan conditions automatically.

  The loan suitability inquiry means 13 transmits to the automatic contract 20 inquiry information for extracting customer loan suitability information. The loan suitability inquiry means 13 of the present embodiment is set to be capable of transmitting at least two types of inquiry information: inquiry information reflected in the loan limit and inquiry information reflected in the loan interest rate. The types of inquiry information are not limited to those related to the loan limit and those related to the loan interest rate, and if it is necessary to determine other loan conditions, it can be set to send the inquiry information about them. . However, since typical lending conditions are lending limit and lending interest rate, it is preferable to send at least one inquiry information among them, and further, lending limit as in this embodiment. It is more preferable that transmission of two types of inquiry information, that is, inquiry information reflected in the amount and inquiry information reflected in the loan interest rate, is an essential configuration.

  The contents of the inquiry information reflected in the loan limit and the inquiry information reflected in the loan interest rate are arbitrary, and can be freely set for each financial institution using the loan examination device 10. FIG. 2 shows an example. Here, the lending limit is related to the consumer's consumption tendency, and the lending interest rate is related to the customer's personal credibility and mental stability. A flashy person with a tendency to spend is considered to be more likely to be delinquent when lending a large amount of money. Repayment can be expected, so a low lending rate is acceptable. Therefore, as in the present embodiment, it can be said that the lending limit is preferably related to the customer's consumption tendency, and the lending interest rate is preferably related to the customer's personal reliability and mental stability. However, this is only an example, and depending on the financial institution, for example, it is possible to relate the consumption trend to the lending interest rate and to relate the personal confidence and stability to the lending limit. It is.

  The number of questions (number of inquiry information transmissions) is not limited. However, when the automatic contract machine 20 is used, it is necessary to consider that the occupation time of each customer of the automatic contract machine 20 does not become too long. In the present embodiment, as shown in FIG. 2, there are three types of inquiry information regarding consumption trends that affect the lending limit and inquiry information regarding personal trust and stability that affect the lending rate. It is set to inquire about the question item. Specifically, in the “consumption trend” category, three questions are set: “Waste or solid”, “I don't tend to escape from reality”, and “I can't stand”. In the category of “and stability”, three questions are set: “Break and keep promises”, “There is no economic idea”, and “Loose in time and good news”.

  In addition, for each question item, two types of questions, a question using direct expression and a question using indirect expression, are set, and a question using direct expression and an indirect expression are used. 2 questions each. When using the direct expression, it is relatively easy for the customer to intentionally make an answer that is different from the original answer in order to obtain favorable financing conditions, while when using the indirect expression In order to obtain favorable financing conditions, it is difficult to understand the desired answer, so it is difficult for the customer to enter the arbiter. For example, as shown in FIG. 3, if the question item “consumption trend” is “wasted house or solid house”, the question using direct expressions is “shopping is fun. "When you buy what you want, don't worry too much about the price, manufacturer, function, material, etc." In addition, questions using direct expressions and questions using indirect expressions may be sent simultaneously or consecutively and answered, but questions using direct expressions and questions using indirect expressions. It is preferable to transmit at intervals of a predetermined time. If sent at the same time or consecutively, the answer to the question using the direct expression that was previously answered remains in the memory, and the direct answer that was previously answered when answering the question using the indirect expression This is because there is a possibility that the consciousness to match the answers to questions using expressions will work strongly. By putting time between both questions, such consciousness will be reduced and the credibility of each answer will be high. Can be expected.

  The loan aptitude information receiving means 14 receives a customer's answer to each inquiry information (question) transmitted from the loan aptitude inquiry means 13. In the present embodiment, the inquiry information reflected in the loan limit and the inquiry information reflected in the loan interest rate are transmitted by the loan suitability inquiry means 13, so that each answer to this is reflected in the loan limit. Loan suitability information and loan suitability information reflected in the loan interest rate. The loan suitability information, which is the customer's answer, is indicated by a score for each question in the inquiry information. For example, as shown in FIG. 4A, “YES”: 1 point, “somewhat YES”: 0.5 point, “somewhat NO”: −0.5 point, “NO” : Set as −1 point. In the present embodiment, the loan condition correcting means 15 described later is set such that the higher the score is, the more advantageous the loan condition is. Therefore, in the question “I like to save money”, “YES” May be “1 point”, but for questions such as “will be wasted”, “YES” is set to “−1 point” and “NO” is set to “1 point”.

  The loan aptitude information reflected in the lending limit and the loan aptitude information reflected in the loan interest rate may include knowledge of money lending in addition to the above-mentioned “consumption trends” and “trust and stability”. Good. This is because the higher the knowledge of monetary lending, the more reasonable borrowing and reliable repayment can be expected. The higher the knowledge, the higher the lending limit and the lower the loan interest rate. . However, the loan aptitude information related to the above-mentioned “consumption trends” and “trust and stability” is more faithful to the personality qualities related to the money and consumption of each customer than the loan aptitude information related to money lending knowledge. In this embodiment, the loan condition correction unit 15 described later is configured to reduce the influence of the loan aptitude information on the knowledge of money lending in the present embodiment. For this reason, as shown in FIG. 4B, with respect to the knowledge of money lending, a score of “correct answer”: 0.5 points and “incorrect answer”: 0 points is given. Note that the financial institution using the loan examination device 10 can arbitrarily set whether or not to place importance on the knowledge of money lending and borrowing.

  The loan condition correcting unit 15 is a unit that corrects the loan condition obtained by the credit review unit 12 using the customer loan aptitude information received by the loan aptitude information receiving unit 14. The loan condition correcting unit 15 includes a confidence level calculating unit 151, a modified loan limit calculating unit 152, and a modified loan interest rate calculating unit 153. In the present embodiment, as described above, since the transmission of the two types of inquiry information, the inquiry information reflected in the loan limit and the inquiry information reflected in the loan interest rate, is essential, the modified loan limit Although both the calculation unit 152 and the modified loan interest rate calculation unit 153 are provided, if only one of the inquiry information is transmitted, the modified loan limit calculation unit 152 and the modified loan interest rate are correspondingly transmitted. Only one of the calculation means 153 is set.

  The credibility calculation means 151 is configured so that each of the loan suitability information reflected in the loan limit received by the loan suitability information receiving means 14 and the loan suitability information reflected in the loan interest rate is an answer to each inquiry information. Answers to questions that use indirect expressions rather than answers to questions that use direct expressions, because the greater the room for the customer's intention, the lower the credibility of the answers, and the smaller the answer, the higher the credibility of the answers. The value adjusted so that the impact on the limit adjustment point or interest rate adjustment point is greater is obtained as the credibility.

  The credibility calculation means 151 calculates the credibility using the first process 151a for obtaining the credibility coefficient from the score for each question of each loan aptitude information and the credibility coefficient for each question obtained in the first process 151a. And a second step 151b to be obtained. As described above, in this embodiment, for each question item, two types of questions, a question using a direct expression and a question using an indirect expression, are set, and a question using a direct expression is used. And 2 questions each using indirect expressions. Therefore, in the first step 151a, for each question item, a question reliability coefficient using direct expression and a question reliability coefficient using indirect expression are calculated. In this embodiment, since the question using the direct expression is asked first, and the question using the indirect expression is asked later, the former is sometimes referred to as the “first half problem” and the latter as the latter. This is called the “second half problem”.

In this embodiment, the reliability coefficient x1 of the question (first half problem) using the direct expression is
x1 = 1-ABS (a1-a2) / 2
(However, a1 and a2 are the scores of answers to questions (first half questions) using direct expressions.)
And
The credibility coefficient x2 of the question using the indirect expression (second half question) is
x2 = 1-ABS (b1-b2) / 2
(However, b1 and b2 are the scores of answers to questions (second half questions) using indirect expressions.)
It becomes.

  In other words, the credibility coefficient is a coefficient that focuses on the degree of variation in the scores for each question, and basically the same or almost the same score is obtained for the answers to two questions of the same type unless the customer's arbitrary intervention is involved. Should be. Therefore, the smaller the variation in the score between two questions of the same type, the higher the credibility coefficient, assuming that the customer's will is not intervening and the credibility as an answer is high.

  In this embodiment, three question items are set for the two categories of “consumption trend” and “trust and stability”, each using a direct expression (first half problem) and an indirect expression. Since there is a (second half problem), as shown in FIG. 5, twelve confidence coefficients are obtained in the range of “0 to 1”.

  In the second step 151b, credibility is obtained for each classification of loan suitability information (consumption trend) reflected in the loan limit and loan suitability information (trust and stability) reflected in the loan interest rate. Specifically, for each classification, the sum of the confidence coefficients of questions (first half questions) using direct expressions obtained in each of the three question items, and questions (second half problems) using indirect expressions The reliability is calculated by taking into account the total reliability coefficient. The purpose of the present invention is to adjust the loan conditions obtained by the credit screening means 12 according to the personal qualities related to the customer's individual financial sensation. However, if this score is reflected in the limit adjustment point or interest rate adjustment point as it is, the customer deliberately replied that an adjustment value favorable to him / her was obtained. In that case, it will be reflected as it is. Therefore, the credibility is obtained, and the credibility is reflected in the score that the customer answers, and the degree of influence of the score on each correction point is adjusted.

  On the other hand, when comparing a question using the direct expression (first half problem) and a question using the indirect expression (second half problem), it is more difficult for the customer to enter the indirect expression. This is a question (second half question). Therefore, it is preferable that the reliability coefficient of the question (second half problem) using indirect expression is more reflected. For this reason, in the present embodiment, the sum of the confidence coefficients of the questions using the direct expression (first half problem) and the sum of the confidence coefficients of the questions using the indirect expression (second half problem) are simply calculated. Rather than taking the average, the sum of the confidence factors of questions using the indirect expression (the second half of the question) is doubled and weighted and averaged so that the impact on the confidence level appears more strongly. ing. The specific calculation formula is as follows.

Credibility (%)
= [(Total reliability coefficient of the first half problem + Total reliability coefficient of the second half problem × 2) / 9] × 100

  In the above formula, “9” is the number of question items of the present embodiment, and therefore, the number of calculated reliability coefficients for the first half problem in one classification is 3, and the number of calculated reliability coefficients for the second half problem. This is the sum of 3 and 2 times. However, as described above, the total reliability coefficient of the question using the indirect expression (second half problem) is doubled, but how many times it can be arbitrarily set.

  First, the corrected loan limit calculating means 152 obtains a limit correction point using the confidence of the loan suitability information (consumption trend) reflected in the loan limit obtained by the confidence calculating means 151. The modified loan interest rate calculation means 153 first obtains the interest rate correction point using the credibility of the loan suitability information (trust and stability) reflected in the loan interest rate. A specific calculation formula is as follows (both limit correction points and interest rate correction points are expressed as “points” for convenience).

Points = (total score for the first half of the classification + total score for the second half of the classification x 2) x credibility of the classification + total score of the problem related to monetary loan

  It should be noted that the total score for the second half question was doubled here, because the influence of the answer that is difficult for the customer to give a priority is given priority, and it is of course possible to arbitrarily set how many times it is to be multiplied. In the present embodiment, as described above, the higher the knowledge of monetary borrowing, the more reasonable borrowing and the reliable repayment can be expected. Since it was used for adjustment, both the limit adjustment points and interest rate adjustment points are added with the “total points for monetary loan issues”. Whether or not to add the “total monetary loan score” is optional, and it is possible to avoid adding it at all, or by multiplying the total score by several times, the degree of influence on each revised value can be increased. It can also be adjusted arbitrarily.

  When the corrected loan limit calculating means 152 obtains the limit amendment point, the amended loan interest rate is obtained by adding the limit amendment point to the loan limit obtained by the credit examination means 12, and obtaining the amended loan interest rate. When calculating the interest rate correction point, the calculation unit 153 calculates the adjusted loan interest rate by adding the interest rate correction point to the loan interest rate obtained by the credit examination unit 12.

  It is optional how the limit correction points are reflected in the loan limit obtained by the credit review means 12, and the value of the limit correction points obtained is directly used as a monetary unit (for example, “10,000 yen”). The amount of change may be taken into account as an increase / decrease amount, or the amount of increase / decrease may be determined by appropriately dividing the obtained limit correction point value. The interest rate correction points can be arbitrarily reflected in the loan interest rate obtained by the credit review means 12, and the obtained interest rate correction points are reflected as they are, or a value that is a fraction is reflected as the rate of increase or decrease. You can make it.

  In this embodiment, as shown in FIG. 6 (a), the limit correction point is rounded off to increase or decrease in units of 50,000 yen to the loan limit obtained by the credit review means 12. The amount of increase or decrease is determined. Further, as shown in FIG. 6B, since the interest rate correction point may take a value of 10 or more, the rate of increase / decrease is set to be 1/10 of the interest rate correction point.

  In the present embodiment, the loan suitability information (consumption trend) reflected in the lending limit and the loan suitability information (trust and stability) reflected in the loan interest rate are divided into three categories. A question item is provided, and in each question item, a total of 4 questions, 2 questions in the first half and 2 questions in the second half, are set, and in the calculation of credibility and points, the weight of the latter half is doubled. As described above, the number of question items, the number of questions, the weighting value, and the like are arbitrary. Based on this point, the credibility coefficient, credibility, and points are expressed by general formulas as follows.

(1) Confidence factor of first half problem (x1)
x1 = 1-ABS (a1-a2) * (2 / k)
(However, k is the number of questions for each question item, and a1 and a2 are the scores of answers to questions (first half questions) using direct expressions.)
(2) Reliability factor of the latter half problem (x2)
x2 = 1-ABS (b1-b2) * (2 / k)
(However, b1 and b2 are the scores of answers to questions (second half questions) using indirect expressions.)
(3) credibility (y)
y = [(c + d * s) / {(j * k / 2) * (s + 1)}] * 100 [%]
(Where c is the total reliability coefficient of the first half problem for each classification, d is the total reliability coefficient of the second half problem for each classification, s is the weight of the second problem, and j is the number of question items for each classification.)
(4) Point (z)
z = (f + g * s) * y + h
(Where f is the total score of the first half problem for each classification, g is the total score of the second half problem for each classification, and h is the score of the knowledge problem of money lending.)

The reliability coefficient (x1) of the first half question and the confidence coefficient (x2) of the second half question may be obtained by the above formulas when the number of questions for each question item is two questions for each of the first half question and the second half question. When there are 3 or more questions in the first half and the second half, there is a concern that the above formula is inferior in accuracy. Therefore, as shown in the following formulas, a standard deviation suitable for calculating the degree of variation should be used. preferable. In the following formulas, it is assumed that the number of questions for the first and second half questions is the same. This is because, as in the present embodiment, when weighting the latter half problem in calculating the reliability, it is preferable that the upper limit of the score for the first half problem and the upper limit of the score for the second half problem be the same. .

  Next, the process for determining loan conditions using the loan review apparatus 10 of this embodiment will be described with reference to FIG. When a customer starts an application for a loan using the automatic contract machine 20 (S201), an agreement regarding acquisition of personal information is requested (S202). If the user does not agree, the application is canceled as it is (S203). If the user agrees, the product will be explained (S204). Next, after the identification document such as a driver's license is read by the scanner of the automatic contract machine 20 (S205), the customer inputs the name and date of birth (S206). The loan screening apparatus 10 receives the name and date of birth information input by the customer at the attribute information receiving means 121 (S101), and refers to the customer database 16 to check whether the customer is an existing customer. (S102). Next, the credit information inquiry means 122 makes a credit information inquiry to the credit information institution 30 and obtains the credit information of the customer (S103). Meanwhile, in the automatic contract machine 20, the customer has entered other attribute items such as address, telephone number, annual income, occupation, etc. (S207). Items are also received by the attribute information receiving means 121 (S104).

  Next, the loan examination device 10 uses the loan suitability inquiry means 13 to inquire information (classification: questions regarding “consumption trend”) reflected in the loan limit amount and inquiry information (classification: “trust”) reflected in the loan interest rate. First, a question using the direct expression (first half question) is transmitted to the automatic contract machine 20 (S105). In the present embodiment, as described above, the two classifications are classified into three question items, and a question using the direct expression (first half question) and a question using the indirect expression (second half) for each question item. 24 questions in total (2 questions) are prepared. In S105, 12 questions (first half question) using direct expressions are transmitted.

  When the automatic contract machine 20 receives a question using the direct expression (first half question) (S208), the customer inputs an answer (financing aptitude information) for each question (S209). This answer is transmitted from the automatic contract machine 20 to the loan examination apparatus 10, and the loan suitability information receiving means 14 of the loan examination apparatus 10 receives the answer (S107). The loan screening apparatus 10 includes the credit information acquired in S103 by the determination unit 123 of the credit screening unit 12 and the customer attribute information received by the attribute information reception unit 121 in S101 and S104 until the answer is received. And a credit determination for determining the loan conditions (S106). In the automatic contract machine 20, after the customer answers the question (first half problem) using the direct expression, the customer explains the contract contents and the basic knowledge about the money lending using a video or the like (S210). When the loan review apparatus 10 receives the completion of this explanation (S108), the loan suitability inquiry means 13 adds 12 questions using the indirect expression (second half problem) to 6 questions related to knowledge of money lending. (S109). Since the description of the contract content in S210 is an essential step in the contract using the automatic contract machine 20, this embodiment answers this question to the question (first half problem) using direct expression. After that, the time interval until the start of the answer to the question using the indirect expression (second half question) is assigned. As described above, it is considered that it is preferable to leave an interval between the two questions for a predetermined time because a highly credible answer can be obtained. However, in this embodiment, the contents of the contract are explained in this way. Since the time is used for this time interval, the occupation time of the automatic contract machine 20 is not increased more than necessary.

  When the automatic contract machine 20 receives a question using the indirect expression (second half question) and a question related to money lending (S211), the customer answers each question (S212) and transmits it to the loan review apparatus 10. To do. When the loan aptitude information receiving means 14 receives an answer to the question (second half problem) using the indirect expression transmitted from the automatic contracting machine 20 and a question regarding the loan loan knowledge (loan aptitude information) (S110), the loan The condition correction means 15 (credibility calculation means 151, corrected loan limit calculation means 152, corrected loan interest rate calculation means 153) operates to correct the loan conditions obtained in S106 (S111), The corrected loan interest rate is output as the examination result (S112). The examination result is transmitted to the automatic contract machine 20, and upon receiving the examination result (S213), the automatic contract machine 20 prompts the customer to confirm the contract contents (S214), and issues and receives a contract contents confirmation (S215). In addition, the steps of card issuance and receipt (S216) are executed to complete the contract (S217). When the loan review apparatus 10 confirms that the contract has been completed, the customer database 16 is updated based on the contract contents (S113).

(Concrete example)
FIG. 8 is a diagram showing a specific example when the loan condition correcting means 15 is operated for a customer whose credit limit is 300,000 yen and a loan interest rate (annual interest) is 18% in S106 of FIG. . The inquiry information by the loan suitability inquiry means 13 is classified into two categories of “consumption tendency” and “trust and stability”, and questions using the direct expression (problem in the first half) 2 for each of the three question items shown in the figure. Scores are shown for questions, 2 questions using indirect expressions (second half questions), and 6 questions that ask about knowledge about money lending.

From this score, for example, the credibility coefficient x1 of the first half problem and the credibility coefficient x2 of the second half problem of the classification “consumption trend” and the question item “whether it is a waster or a solid house” are obtained as follows.
x1 = 1-ABS (1-0.5) /2=0.75
x2 = 1-ABS (0.5-0.5) / 2 = 1

  Similarly, when the reliability coefficients x1 and x2 of each question item are obtained, each value shown in the “credibility coefficient” column of the figure is obtained.

As a result, the credibility is calculated as follows.
Confidence y1: Consumption trend
y1 = {0.75 + 1 + 1 + (1 + 1 + 0.5) * 2} /9*100=86.111 (%)
Confidence y2 regarding “trust and stability”:
y2 = {0.75 + 1 + 1 + (0.75 + 1 + 1) * 2} /9*100=91.666 (%)

The limit adjustment points and interest rate adjustment points are obtained as follows.
Limit correction point z1:
z1 = (3.5 + 2 * 2) * 0.86 + 1.5 = 7.95
Interest rate correction point z2:
z2 = (4.5 + 2.5 * 2) * 0.92 + 1.5 = 10.24

  The limit correction point “7.95” corresponds to “+3 or more and less than +8” in FIG. 6A, so in this embodiment, the increase / decrease amount is “+50,000 yen”, and the interest rate correction point “10.24” is This corresponds to “+8 or more and less than +13” in FIG. 6B, and the rate of increase / decrease is “−1.0%”. As a result, the revised loan limit, which is the examination result in S112 of FIG. 7, is “350,000 yen”, and the revised loan interest rate is “17.0%”. In other words, in the case of this customer, compared with those who have the same credit when judged only by attributes and credit information, the solidity in “consumption trend” and the quality such as meticulousness in “trust and stability” are reflected. As a result, more favorable financing conditions were obtained.

FIG. 1 is a diagram for explaining a schematic configuration of a loan examination apparatus according to an embodiment of the present invention. FIG. 2 is a diagram illustrating an example of inquiry information reflected on the loan limit and inquiry information regarding a loan interest rate. FIG. 3 is a diagram showing a specific example of each question of inquiry information reflected on the loan limit and inquiry information regarding the loan interest rate. FIGS. 4A and 4B are diagrams showing examples of scores given to customer responses (financing aptitude information). FIG. 5 is a diagram showing an example of the relationship between each question and the confidence coefficient. Fig. 6 (a) is a diagram showing an example of the relationship between the limit adjustment point and the increase / decrease amount of the loan limit. Fig. 6 (b) shows the relationship between the interest rate correction point and the increase / decrease rate of the loan interest rate. It is the figure which showed an example. FIG. 7 is a flowchart for explaining a loan condition determination process using the loan examination apparatus according to the embodiment. FIG. 8 is a diagram for explaining a specific example when the loan condition of the customer whose credit is determined is corrected by the loan condition correcting means.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Loan examination apparatus 11 Control part 12 Credit examination means 121 Attribute information reception means 122 Credit information inquiry means 123 Judgment means 13 Loan suitability inquiry means 14 Loan suitability information reception means 15 Loan condition correction means 151 Reliability calculation means 152 Amended loan limit Calculation means 153 Modified loan interest rate calculation means 20 Automatic contract machine

Claims (18)

A loan examination device for examining a loan to a customer,
Credit review means for determining loan conditions based on the customer attributes and credit information;
Loan aptitude information receiving means for receiving the loan aptitude information of the customer transmitted from the terminal device of the customer;
A loan examination device, comprising: loan condition correction means for correcting the loan condition obtained by the credit review means based on the loan suitability information of the customer received by the loan suitability information receiving means. A loan suitability inquiry means for transmitting inquiry information for extracting the customer loan suitability information to the customer terminal device;
2. The loan review apparatus according to claim 1, wherein the loan suitability information receiving unit is configured to receive a customer response to the inquiry information transmitted from the terminal device as the loan suitability information. The loan suitability inquiry means is set to be capable of transmitting inquiry information including at least one of inquiry information reflected in a loan limit and inquiry information reflected in a loan interest rate,
The loan suitability information receiving means is at least one of loan suitability information reflected in a loan limit and loan suitability information reflected in a loan interest rate as a response to each inquiry information transmitted from the loan suitability inquiry means. The loan examination apparatus according to claim 2, wherein the loan suitability information including one can be received.   The loan condition correction means includes at least one of the loan suitability information reflected by the loan limit amount and the loan suitability information reflected by the loan interest rate received by the loan suitability information receiving means. The loan examination apparatus according to claim 1, wherein the loan condition obtained by the credit examination unit is corrected based on the credit card. The loan suitability information is indicated by a score for each question of the inquiry information,
The loan condition correcting means includes
A credential for obtaining a credibility as an answer to each inquiry information for at least one of the loan aptitude information reflected in the loan limit received by the loan aptitude information receiving means and the loan aptitude information reflected in the loan interest rate While having a degree calculation means,
A limit correction point is obtained by using the reliability of the loan suitability information reflected in the loan limit amount obtained from the credit calculation means, and the limit correction point is set to the loan limit amount obtained by the credit examination means. The interest rate correction point is calculated using the adjusted loan limit calculation means for calculating the adjusted loan limit in consideration and the reliability of the loan suitability information reflected in the loan interest rate obtained from the reliability calculation means. 5. The loan examination apparatus according to claim 4, further comprising at least one means of a modified loan interest rate calculating means for obtaining a modified loan interest rate by adding points to the loan interest rate obtained by the credit examination means. The reliability calculation means includes:
The reliability of at least one of the loan suitability information reflected in the loan limit received by the loan suitability information receiving means and the loan suitability information reflected in the loan interest rate from the score for each question of each loan suitability information A first step for determining a coefficient;
Using the confidence factor obtained in the first step, at least one of confidence for the loan suitability information reflected in the loan limit and confidence for the loan suitability information reflected in the loan interest rate. The loan examination apparatus according to claim 5, further comprising a second step for obtaining one.   The inquiry information reflected in the loan limit and the inquiry information reflected in the loan interest rate sent from the loan suitability inquiry means are both direct and indirect questions for one question item. The loan examination apparatus according to any one of claims 2 to 6, wherein two types of questions, a question using an expression, are set.   The loan examination according to claim 7, wherein the question using the direct expression and the question using the indirect expression transmitted from the loan suitability inquiry means are transmitted at a predetermined time interval. apparatus.   In the second step of the credibility calculation means, at least one of the credibility of the loan aptitude information reflected in the loan limit and the credibility of the loan aptitude information reflected in the loan interest rate. When determining the credibility, the credibility coefficient corresponding to the question using the indirect expression has a greater influence on the credibility than the credibility coefficient corresponding to the question using the direct expression. The loan examination apparatus according to claim 6, wherein the loan examination device is obtained by weighting the balance. A computer program installed in a loan examination device for examining a loan to a customer,
Credit review means for determining loan conditions based on the customer attributes and credit information;
Loan aptitude information receiving means for receiving the loan aptitude information of the customer transmitted from the terminal device of the customer;
A computer program comprising: loan condition correcting means for correcting the loan condition obtained by the credit review means based on the loan suitability information of the customer received by the loan suitability information receiving means. A loan suitability inquiry means for transmitting inquiry information for extracting the customer loan suitability information to the customer terminal device;
The computer program according to claim 10, wherein the loan suitability information receiving unit is configured to receive a customer response to the inquiry information transmitted from the terminal device as the loan suitability information. The loan suitability inquiry means is set to be capable of transmitting inquiry information including at least one of inquiry information reflected in a loan limit and inquiry information reflected in a loan interest rate,
The loan suitability information receiving means is at least one of loan suitability information reflected in a loan limit and loan suitability information reflected in a loan interest rate as a response to each inquiry information transmitted from the loan suitability inquiry means. 12. The computer program according to claim 11, wherein the loan aptitude information including one can be received.   The loan condition correction means includes at least one of the loan suitability information reflected by the loan limit amount and the loan suitability information reflected by the loan interest rate received by the loan suitability information receiving means. The computer program according to any one of claims 10 to 12, wherein the loan condition obtained by the credit examination means is modified based on the credit card. The loan suitability information is indicated by a score for each question of the inquiry information,
The loan condition correcting means includes
A credential for obtaining a credibility as an answer to each inquiry information for at least one of the loan aptitude information reflected in the loan limit received by the loan aptitude information receiving means and the loan aptitude information reflected in the loan interest rate While having a degree calculation means,
A limit correction point is obtained by using the reliability of the loan suitability information reflected in the loan limit amount obtained from the credit calculation means, and the limit correction point is set to the loan limit amount obtained by the credit examination means. The interest rate correction point is calculated using the adjusted loan limit calculation means for calculating the adjusted loan limit in consideration and the reliability of the loan suitability information reflected in the loan interest rate obtained from the reliability calculation means. 14. The computer program according to claim 13, further comprising at least one means of a modified loan interest rate calculating means for obtaining a modified loan interest rate by adding points to the loan interest rate obtained by the credit examination means. The reliability calculation means includes:
The reliability of at least one of the loan suitability information reflected in the loan limit received by the loan suitability information receiving means and the loan suitability information reflected in the loan interest rate from the score for each question of each loan suitability information A first step for determining a coefficient;
Using the confidence factor obtained in the first step, at least one of confidence for the loan suitability information reflected in the loan limit and confidence for the loan suitability information reflected in the loan interest rate. 15. The computer program according to claim 14, further comprising a second step of obtaining one.   The inquiry information reflected in the loan limit and the inquiry information reflected in the loan interest rate sent from the loan suitability inquiry means are both direct and indirect questions for one question item. The computer program according to any one of claims 11 to 15, wherein two types of questions, a question using an expression, are set.   17. The computer program according to claim 16, wherein the question using the direct expression and the question using the indirect expression transmitted from the loan suitability inquiry means are transmitted at a predetermined time interval. .   In the second step of the credibility calculation means, at least one of the credibility of the loan aptitude information reflected in the loan limit and the credibility of the loan aptitude information reflected in the loan interest rate. When determining the credibility, the credibility coefficient corresponding to the question using the indirect expression has a greater influence on the credibility than the credibility coefficient corresponding to the question using the direct expression. The computer program according to claim 15, wherein the computer program is obtained by weighting.

Images