CN114912833A - Sponsor screening method, sponsor screening apparatus, and computer-readable storage medium - Google Patents

Abstract

The embodiment of the application discloses a sponsor screening method, a sponsor screening device and a computer readable storage medium, which are used for screening out a target sponsor for processing a current order. The method in the embodiment of the application comprises the following steps: the method comprises the steps of obtaining an order risk value and a preset risk threshold value of a current order, determining a plurality of alternative sponsors, wherein the plurality of alternative sponsors have corresponding sponsor risk values, carrying out fuzzy logic processing on the current order risk value, the preset risk threshold value and the sponsor risk value of the alternative sponsor by utilizing a fuzzy logic reasoning algorithm aiming at each alternative sponsor to obtain rejection probability corresponding to the alternative sponsor, wherein the rejection probability is used for representing the possibility that the alternative sponsor rejects the current order, and the alternative sponsor of which the rejection probability does not exceed the preset rejection probability threshold value is used as a target sponsor for processing the current order.

Description

Employer screening method, employer screening apparatus, and computer-readable storage medium

technical field

The embodiments of the present application relate to the field of capital screening, and more specifically, to a capital screening method, a capital screening device, and a computer-readable storage medium.

Background technique

The loan assistance platform has a capital pool, and the capital pool has multiple capitals. The loan assistance platform can receive orders and match the orders to the capital. The contract signed by the capital and the loan platform stipulates a preset risk threshold. If the capital of the capital exceeds the risk value of the capital With the preset risk threshold, the loan assistance platform needs to pay the penalty fee to the capital. In order to increase the profit of the general market, it is necessary to reduce the situation that the loan assistance platform pays the penalty fee to the capital. Therefore, it is necessary to filter out the processing orders from the alternative capital in the capital pool. The target capital, so that the capital risk value of the target capital after processing the order is less than or equal to the preset risk threshold as much as possible.

The existing capital screening method is to obtain the order risk value of the current order, the preset risk threshold, and the capital risk value of each capital. You can first determine whether the order risk value is less than or equal to the preset risk threshold. Whether the capital risk value of each candidate capital is less than or equal to the preset risk threshold, the judgment result is obtained, and according to the judgment result, the candidate capital whose capital risk value is less than or equal to the preset risk threshold is regarded as the target capital for processing the current order.

However, for this method of capital selection, under the condition that the order risk is less than the preset risk threshold, the alternative capital can be used as the target capital for processing the current order as long as the capital risk is less than or equal to the pre-set risk threshold. When the capital risk value of the capital selection party is about to reach the preset risk threshold, the current order can still be processed. After the loan is successful, the probability that the capital risk of the target capital is greater than the preset risk threshold is increased, and the loan assistance platform needs to be sent to the target capital. The probability of paying a penalty fee reduces the profit of the broader market.

SUMMARY OF THE INVENTION

The embodiments of the present application provide a method, device and a computer-readable storage medium for screening an employer, which are capable of screening out a target employer for processing a current order.

In the first aspect, the embodiments of the present application provide a method for screening capital, including:

Obtain the order risk value and preset risk threshold of the current order;

determining a plurality of alternative capital parties, the plurality of alternative capital parties have corresponding capital party risk values;

For each candidate capital, fuzzy logic is used to perform fuzzy logic processing on the risk value of the current order, the preset risk threshold and the capital risk value of the candidate capital, to obtain the candidate capital The corresponding rejection probability; the rejection probability is used to indicate the possibility that the alternative employer rejects the current order;

The candidate employer whose rejection probability does not exceed a preset rejection probability threshold is used as the target employer for processing the current order.

Optionally, before performing fuzzy logic processing on the capital risk value, the current order risk value and the preset risk threshold by using a fuzzy logic inference algorithm, the method further includes:

Obtain the historical order amount and historical order risk value of each said alternative capital;

For each candidate capital, the capital risk value of the candidate capital is obtained according to the historical order amount and historical order risk value of the candidate capital.

Optionally, determining the capital risk value of the candidate capital according to the historical order amount and historical order risk of the candidate capital, including:

A weighted average of the historical order amount and the historical order risk value of the candidate capital is performed to obtain the capital risk value of the candidate capital.

Optionally, the fuzzy logic inference algorithm is used to perform fuzzy logic processing on the capital risk value, the current order risk value, and the preset risk threshold to obtain the rejection of the current order by the alternative capital party. probabilities, including:

Taking the difference between the capital risk value and the preset risk threshold as the first input quantity, and taking the difference between the current order risk value and the capital risk value as the second input quantity;

Inputting the first input into the first membership function to obtain the membership of the first input, and inputting the second input into the second membership function to obtain the membership of the second input;

According to the membership degree of the first input quantity and the membership degree of the second input quantity, the target rejection willingness ambiguous mark corresponding to the first input quantity and the second input quantity and each target rejection willingness ambiguous mark are determined. affiliation;

The rejection probability is obtained by performing a weighted average of the membership degree of each target rejection willingness ambiguous mark and the preset weight corresponding to each target rejection willingness ambiguous mark.

Optionally, according to the degree of membership of the first input amount and the degree of membership of the second input amount, determine the target rejection willingness ambiguous mark corresponding to the first input amount and the second input amount and each Membership of vague markers of target rejection willingness, including:

In the preset fuzzy rule base, the rejection intention fuzzy marks corresponding to the membership degree of the first input quantity and the membership degree of the second input quantity are used as the target rejection intention fuzzy mark;

According to the membership degree of the first input quantity corresponding to the target rejection willingness ambiguous mark and the membership degree of the second input quantity, determine the intensity value of the target rejection willingness ambiguous mark;

According to the strength value of at least one of the target rejection intention ambiguous marks, the membership degree of each target rejection intention ambiguous mark is determined.

Optionally, determining the intensity value of the target rejection intention fuzzy mark according to the membership degree of the first input quantity corresponding to the target rejection intention fuzzy mark and the membership degree of the second input quantity, including:

The smaller of the membership degree of the first input quantity corresponding to the target rejection willingness ambiguous mark and the membership degree of the second input quantity is taken as the strength value of the target rejection willingness ambiguous mark.

Optionally, determining the degree of membership of each ambiguous mark of target rejection willingness according to the intensity value of at least one of the target rejection willingness ambiguous marks, including:

In at least one of the target rejection intention ambiguous marks, determine the target rejection intention ambiguous marks belonging to the same type;

The maximum value of the intensity values of the same kind of target rejection willing ambiguous marks is taken as the membership degree of this kind of target rejection willing ambiguous marks.

In a second aspect, the embodiments of the present application provide a device for screening employers, including:

Obtaining unit, used to obtain the order risk value and preset risk threshold of the current order;

a determining unit, configured to determine a plurality of candidate capitals, the plurality of candidate capitals have corresponding capital risk values;

The processing unit is configured to perform fuzzy logic processing on the current order risk value, the preset risk threshold and the capital risk value of the alternative capital by using a fuzzy logic inference algorithm for each candidate capital, to obtain The rejection probability corresponding to the alternative capital; the rejection probability is used to indicate the possibility of the alternative capital rejecting the current order;

As a unit, it is used for taking the candidate employer whose rejection probability does not exceed the preset rejection probability threshold as the target employer for processing the current order.

In a third aspect, an embodiment of the present application provides a device for screening employers, including:

Central processing unit, memory, input and output interface, wired or wireless network interface and power supply;

the memory is a short-term storage memory or a persistent storage memory;

The central processing unit is configured to communicate with the memory and execute the operations of the instructions in the memory to perform the aforementioned method of screening for employers.

In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, where the computer-readable storage medium includes instructions, when the instructions are executed on a computer, the computer causes the computer to execute the foregoing method for screening employers.

In a fifth aspect, an embodiment of the present application provides a computer program product containing instructions, which, when the computer program product is run on a computer, enables the computer to execute the aforementioned method for screening employers.

It can be seen from the above technical solutions that the embodiments of the present application have the following advantages: for each candidate capital, the fuzzy logic inference algorithm can be used to blur the current order risk value, the preset risk threshold and the capital risk value of the candidate capital. Logical processing to obtain the rejection probability corresponding to the alternative employer, the rejection probability is used to indicate the possibility of the alternative employer rejecting the current order, and the alternative employer whose rejection probability does not exceed the preset rejection probability threshold is used as the target employer for processing the current order. The requirement that the alternative funder can be used as the target funder for processing the current order has been improved. After the loan is successful, the probability that the funder risk of the target funder is greater than the preset risk threshold is reduced, and the need for the loan assistance platform to pay the penalty fee to the target funder is reduced. The probability increases the profit of the broader market profit.

Description of drawings

FIG. 1 is a schematic flowchart of a method for screening an employer disclosed in an embodiment of the application;

2 is a schematic flowchart of a fuzzy logic processing method disclosed in an embodiment of the present application;

3 is a schematic diagram of a first membership function disclosed in an embodiment of the present application;

4 is a schematic diagram of a second membership function disclosed in an embodiment of the present application;

5 is a schematic diagram of a membership function of rejection willingness disclosed in an embodiment of the present application;

6 is a schematic diagram of a fuzzy logic inference curved surface disclosed in an embodiment of the present application;

7 is a schematic structural diagram of a capital screening device disclosed in an embodiment of the present application;

8 is a schematic structural diagram of another capital screening device disclosed in an embodiment of the present application;

FIG. 9 is a schematic structural diagram of another capital screening device disclosed in an embodiment of the present application.

Detailed ways

The embodiments of the present application provide a capital screening method, a capital screening device, and a computer-readable storage medium, which are used to screen out a target capital for processing a current order. The capital screening device is a capital risk control device.

Please refer to FIG. 1. FIG. 1 is a schematic flowchart of a method for screening an employer disclosed in an embodiment of the present application. The method includes:

101. Obtain the order risk value and preset risk threshold of the current order.

In this embodiment, when performing capital screening, the order risk value and preset risk threshold of the current order can be obtained.

102. Determine multiple candidate capitals, and the multiple candidate capitals have corresponding capital risk values.

Multiple alternative capitals may be determined, and multiple alternative capitals have corresponding capital risk values.

103. For each candidate employer, use the fuzzy logic inference algorithm to perform fuzzy logic processing on the risk value of the current order, the preset risk threshold and the employer risk value of the candidate employer, and obtain the rejection probability corresponding to the candidate employer; It indicates the possibility that the alternative employer will reject the current order.

After multiple candidate capitals are determined, fuzzy logic can be used for each candidate capital to use fuzzy logic inference algorithm to perform fuzzy logic processing on the risk value of the current order, the preset risk threshold and the capital risk value of the candidate capital, and obtain the corresponding candidate capital. The rejection probability of ; the rejection probability is used to indicate the possibility of the alternative employer rejecting the current order. It is understandable that by using fuzzy logic reasoning algorithm to perform fuzzy logic processing on the risk value of the current order, the preset risk threshold and the capital risk value of the alternative capital, the rejection probability that can better reflect the possibility of the alternative capital rejecting the current order can be obtained. , the rejection probability can be used as a screening basis for employer screening. Therefore, it is of certain significance to use fuzzy logic reasoning algorithm to perform fuzzy logic processing on the risk value of the current order, the preset risk threshold and the capital risk value of the alternative capital.

104. Take the candidate employer whose rejection probability does not exceed the preset rejection probability threshold as the target employer for processing the current order.

After obtaining the rejection probability corresponding to the candidate employer, the candidate employer whose rejection probability does not exceed the preset rejection probability threshold may be used as the target employer for processing the current order.

In the embodiment of the present application, fuzzy logic processing can be performed on each candidate capital by using a fuzzy logic inference algorithm on the risk value of the current order, the preset risk threshold, and the capital risk value of the candidate capital, and the corresponding rejection of the candidate capital can be obtained. The probability of rejection is used to indicate the possibility of the candidate employer rejecting the current order, and the candidate employer whose rejection probability does not exceed the preset rejection probability threshold is used as the target employer for processing the current order. The requirement that the alternative funder can be used as the target funder for processing the current order has been improved. After the loan is successful, the probability that the funder risk of the target funder is greater than the preset risk threshold is reduced, and the need for the loan assistance platform to pay the penalty fee to the target funder is reduced. The probability increases the profit of the broader market profit.

In the embodiment of the present application, using the fuzzy logic reasoning algorithm, the risk value of the employer, the risk value of the current order and the preset risk threshold are subjected to fuzzy logic processing, and there are many methods for obtaining the rejection probability of the candidate employer rejecting the current order, as follows. One of these methods is described.

In this embodiment, there are multiple candidate capitals in the capital pool of the loan assistance platform, and the candidate capitals can be either an alternative institution that can provide funds, or other capital providers that can provide funds, which is not limited here. . When the loan assistance platform receives the current order, the artificial intelligence matching model can filter the candidate capital in the capital pool before deciding the matching order of the candidate capital, and use the candidate capital that meets the conditions as the target capital to process the current order, in order to prevent The risk value of the capital exceeds the preset risk threshold, resulting in losses. When performing capital screening, the order risk value and preset risk threshold of the current order can be obtained. The current order can be formed based on information such as user information, order amount, account period and interest rate. The specific method of forming the current order is not described here. limited.

Multiple alternative capitals may be determined, and multiple alternative capitals have corresponding capital risk values. Among them, the capital risk value of multiple alternative capital parties is predetermined, and the determination method may be: obtaining the historical order amount and historical order risk value of each candidate capital party, and for each candidate capital party, according to the historical order risk value of the alternative capital party The order amount and historical order risk value are obtained from the capital risk value of the alternative capital, which can also be determined according to other relevant information of the historical orders processed by the alternative capital, or can be determined according to other relevant information of the alternative capital. There is no limitation on the method of selecting the capital's capital value at risk. Among them, the method of obtaining the capital risk value of the alternative capital party according to the historical order amount and historical order risk value of the alternative capital party may be weighted average of the historical order amount and historical order risk value of the alternative capital party to obtain the capital capital of the alternative capital party. The value at risk, the specific formula is as follows:

Formula 1

Among them, Zr is the capital risk value, od_amt is the historical order amount, and od_lc is the historical order risk. It can be understood that, in addition to the above methods of obtaining the capital risk value, other methods can also be used to obtain the capital risk value of the alternative capital based on the historical order amount and historical order risk value of the alternate capital, which is not limited here. .

After obtaining the capital risk value of the alternative capital, the fuzzy logic inference algorithm can be used for each candidate capital to perform fuzzy logic processing on the risk value of the current order, the preset risk threshold and the capital risk value of the alternative capital, and obtain the alternative capital. The rejection probability corresponding to the employer; the rejection probability is used to indicate the possibility of the alternative employer rejecting the current order. There are various methods for obtaining the rejection probability by performing fuzzy logic processing. One of the methods will be described below. Please refer to FIG. 2 , which is a schematic flowchart of a fuzzy logic processing method disclosed in an embodiment of the present application. , methods include:

201. Use the difference between the capital risk value and the preset risk threshold as the first input quantity, and use the difference between the current order risk value and the capital risk value as the second input quantity.

The difference between the capital risk value and the preset risk threshold may be used as the first input quantity, and the difference between the current order risk value and the capital risk value may be used as the second input quantity.

It can be understood that, the first input quantity and the second input quantity may have corresponding numerical ranges, and the specific numerical ranges are not limited. For this embodiment, for example, the numerical range of the first input quantity is (-0.5˜0.5), and the numerical range of the second input quantity is (-1˜1).

202. Input the first input quantity into the first membership function to obtain the membership of the first input, and input the second input into the second membership function to obtain the membership of the second input.

After obtaining the first input quantity and the second input quantity, the first input quantity can be input into the first membership function to obtain the membership degree of the first input quantity, and the second input quantity can be input into the second membership function to obtain the second input quantity. affiliation.

It is worth mentioning that the first membership function and the second membership function are predetermined. Specifically, the numerical range of the first input quantity and the numerical range of the second input quantity can be obtained in advance. For the first input quantity and For the second input amount, a preset number of fuzzy marks can be set. For example, the first input amount and the second input amount are both set with 5 fuzzy marks, which are much higher, higher, almost equal, lower and lower. Please refer to the figure 3. FIG. 3 is a schematic diagram of a first membership function disclosed in an embodiment of the application. Different lines in FIG. 3 represent different fuzzy marks, the x-axis represents the first input quantity, and the y-axis represents the membership degree of the first input quantity. , please refer to FIG. 4, FIG. 4 is a schematic diagram of a second membership function disclosed in an embodiment of the application, different lines in FIG. 4 represent different fuzzy marks, the x-axis represents the second input quantity, and the y-axis represents the second input Quantity of membership. For example, the first input quantity is 0.015, and 0.015 is input into the first membership function, the membership degree of the first input quantity can be obtained. Please refer to Table 1. Table 1 is a membership degree of a first input quantity disclosed in an embodiment of the present application.

The first input amount: capital risk value - preset risk threshold 0.015

Table 1

The second input is -0.01, and -0.01 is input into the second membership function to obtain the membership of the second input. Please refer to Table 2. Table 2 is a membership degree of a second input quantity disclosed in this embodiment of the present application.

The second input amount: order risk value - capital risk value - 0.01

Table 2

203. Determine, according to the membership degree of the first input quantity and the membership degree of the second input quantity, the target rejection willingness ambiguous mark corresponding to the first input quantity and the second input quantity and the membership degree of each target rejection willingness ambiguous mark.

After obtaining the membership degree of the first input quantity and the membership degree of the second input quantity, the target rejection corresponding to the first input quantity and the second input quantity can be determined according to the membership degree of the first input quantity and the membership degree of the second input quantity Willful ambiguity markers and membership of each target's rejection willingness ambiguity markers. Specifically, according to the membership degree of the first input quantity and the membership degree of the second input quantity, the method of determining the target rejection willingness ambiguous mark corresponding to the first input quantity and the second input quantity and the membership degree of each target rejection willingness ambiguous mark There are many kinds. It can be that in the preset fuzzy rule base, the rejection willingness fuzzy mark corresponding to the membership degree of the first input quantity and the membership degree of the second input quantity is used as the target rejection willingness fuzzy mark, according to the target rejection willingness fuzzy mark corresponding The membership degree of the first input quantity and the membership degree of the second input quantity are determined, the intensity value of the target rejection willing ambiguous mark is determined, and the membership degree of each target rejection willing ambiguous mark is determined according to the strength value of at least one target rejection willing ambiguous mark. .

It is worth mentioning that the preset fuzzy rule base is pre-determined, and the determination method is based on the fuzzy mark of the first input amount, the fuzzy mark of the second input amount, the predetermined fuzzy mark of rejection willingness, business experience and actual value distribution. definite. The method for determining the vagueness of rejection willingness may be to obtain the numerical range of rejection willingness in advance, the numerical range of rejection willingness may be (0-1), and a preset number of vague rejection willingness marks may be set, such as three rejection willingnesses. The fuzzy marks are respectively strong rejection intention, medium rejection intention and weak rejection intention. Please refer to FIG. 5 , which is a schematic diagram of a rejection intention membership function disclosed in an embodiment of the present application. Different lines in Figure 5 represent different fuzzy marks, the x-axis represents rejection willingness, and the y-axis represents the degree of membership of the first input. Please refer to Table 3. Table 3 is a preset fuzzy rule base disclosed by the embodiment of the present application. For Table 3, the first input amount is the difference between the capital risk value and the preset risk threshold, and the second input amount is the difference between the current order risk value and the capital risk value. If the capital risk of the alternative capital exceeds the preset risk threshold, the order The VaR is much lower than the capital's VaR, and it can still be used as the target capital for processing the current order, because the order VaR is lower than the capital's VaR. After processing the order, the asset VaR can be lowered.

table 3

Please refer to FIG. 6. FIG. 6 is a schematic diagram of a fuzzy logic inference surface disclosed in an embodiment of the present application. The coordinate system of FIG. 6 takes the first input quantity as the x-axis, the second input quantity as the y-axis, and the rejection intention as the z-axis. 3D coordinate system.

It should be understood that, according to the membership degree of the first input quantity and the membership degree of the second input quantity corresponding to the target rejection intention fuzzy mark, the method for determining the intensity value of the target rejection intention fuzzy mark may be: The smaller of the membership degree of the first input quantity corresponding to the mark and the membership degree of the second input quantity, as the intensity value of the target rejection willingness fuzzy mark, can also be the first according to the target rejection willingness fuzzy mark. The membership degree of the input quantity and the membership degree of the second input quantity are the method for determining the intensity value of the fuzzy mark of the target rejection willingness, which is not specifically limited here. Wherein, according to the strength value of the at least one target rejection willingness ambiguous mark, the method for determining the membership degree of each target rejection willingness ambiguous mark may be to determine the target rejection willingness ambiguous mark belonging to the same type in the at least one target rejection willingness ambiguous mark, The maximum value of the intensity values of the same kind of target rejection willing ambiguous marks is used as the membership degree of this kind of target rejection willing ambiguous marks, and it can also be other values according to at least one target rejection willing ambiguous mark to determine each target rejection willingness. The method for blurring the membership degree of the mark is not specifically limited here. For example, the first input is 0.015, the second input is -0.01, and the membership of the first input is "a much higher degree of membership is 0.0102, a higher degree of membership is 0.9444, and an almost equal degree of membership. is 0, the lower membership is 0, and the much lower membership is 0", the membership of the second input is "much higher membership is 0, higher membership is 0, almost equal membership is 0.9800, a lower degree of membership is 0.0200, and a much lower degree of membership is 0", according to the degree of membership of the first input quantity and the degree of membership of the second input quantity, please continue to refer to Table 1, which can be found in the preset fuzzy rule base , the target fuzzy rules corresponding to the membership degree of the first input quantity of 0.015 and the membership degree of the second input quantity of -0.01 can be determined, and the target fuzzy rules are as follows:

Target fuzzy rule a: The membership degree of the risk value of the employer is much higher than the preset risk threshold is 0.0102, the membership degree of the order risk value is almost equal to the risk value of the employer is 0.9800, and the target rejection willingness is vaguely marked as strong rejection willingness;

Target fuzzy rule b: the membership degree of the risk value of the employer is much higher than the preset risk threshold is 0.0102, the membership degree of the order risk value is a little lower than the risk value of the employer is 0.0200, and the target rejection willingness is vaguely marked as rejection willingness;

Target fuzzy rule c: the membership degree of the risk value of the employer is higher than the preset risk threshold is 0.9444, the membership degree of the order risk value and the risk value of the employer is almost equal to 0.9800, and the target rejection willingness is vaguely marked as the rejection willingness;

Target fuzzy rule d: the membership degree of the risk value of the employer is much higher than the preset risk threshold is 0.9444, the membership degree of the order risk value is almost equal to the risk value of the employer is 0.0200, and the target rejection willingness is vaguely marked as weak rejection willingness.

After obtaining the target fuzzy rules, it can be determined that the target rejection willingness corresponding to the membership degree of the first input quantity of 0.015 and the membership degree of the second input quantity of -0.01 is marked with strong rejection intention, medium rejection intention and weak rejection intention. The smaller of the membership degree of the first input quantity and the membership degree of the second input quantity corresponding to the target rejection willingness fuzzy mark is taken as the strength value of the target rejection willingness fuzzy mark. Therefore, the strength of the target fuzzy rule a is 0.0102 , the strength of target fuzzy rule b is 0.0102, the strength of target fuzzy rule c is 0.9444, and the strength of target fuzzy rule d is 0.0200. Among these four target fuzzy rules, the rejection willingness of target fuzzy rule b and target fuzzy rule c is fuzzy If the marks are all medium, the maximum value of the intensity values can be taken as the membership degree of medium. Therefore, for the first input amount of 0.015 and the second input amount of -0.01, the membership degree of the rejection willingness fuzzy marked medium is 0.9444, The membership degree of rejection willingness vaguely marked as high is 0.0102, and the membership degree of rejection willingness vaguely marked as low is 0.0200. It can be understood that, in addition to the above-described method for determining the fuzzy marks of target rejection willingness corresponding to the first input quantity and the second input quantity and the membership degree of each target rejection willingness ambiguous mark, there may also be other methods based on the first input. The membership degree of the quantity and the membership degree of the second input quantity, the method of determining the target rejection willingness fuzzy mark corresponding to the first input quantity and the second input quantity and the membership degree of each target rejection willingness fuzzy mark, which is not limited here. .

204. Perform a weighted average of the membership degree of each target rejection willingness ambiguous mark and the preset weight corresponding to each target rejection willingness ambiguous mark to obtain a rejection probability.

After determining the target rejection willingness ambiguous mark corresponding to the first input quantity and the second input quantity and the membership degree of each target rejection willingness ambiguous mark, the membership degree of each target rejection willingness ambiguous mark and each target rejection willingness ambiguous mark can be calculated. The corresponding preset weights are weighted and averaged to obtain the rejection probability. Continue to take the example that the first input is 0.015 and the second input is -0.01, and the membership degree in the rejection willingness is 0.9444, the membership degree with high rejection willingness is 0.0102, and the membership degree with low rejection willingness is 0.0200. The preset weights corresponding to the fuzzy marks of the target rejection willingness, for example, the weight of strong rejection intention is 1, the weight of rejection intention is 0.5, and the weight of weak rejection intention is 0.001, the first input amount can be calculated as 0.015, and the second input amount is The probability of rejection of -0.01 is calculated as follows:

Formula 2

The rejection probability that the first input amount is 0.015 and the second input amount is -0.01 is 0.4950. It can be understood that, in addition to the fuzzy logic processing method shown in FIG. 2 , there may also be other methods for performing fuzzy logic processing to obtain the rejection probability, which is not specifically limited here.

After obtaining the rejection probability corresponding to the candidate employer, the candidate employer whose rejection probability does not exceed the preset rejection probability threshold may be used as the target employer for processing the current order. Continue to take the example that the first input amount is 0.015 and the second input amount is -0.01, and the rejection probability that the first input amount is 0.015 and the second input amount is -0.01 is 0.4950, for example, the preset rejection probability threshold is 0.6, Because 0.4950 is less than 0.6, this alternative capital can be used as the target capital for processing the current order.

It can be understood that the alternative capital may be the profit-sharing capital under the profit-sharing model, or may be the capital in other models, which is not specifically limited here. It can also be understood that the target employer that processes the current order may be the employer that directly processes the current order, or may be another alternative employer for processing the current order, which is not specifically limited here. It is also understandable that the fuzzy logic reasoning algorithm is transferable, and can be used for other business scenarios that require decision-making in addition to screening the business scenarios of employers, which is not limited here.

In this embodiment, the fuzzy logic inference algorithm can be used for each candidate employer to perform fuzzy logic processing on the risk value of the current order, the preset risk threshold, and the employer risk value of the candidate employer, to obtain the rejection probability corresponding to the candidate employer , the rejection probability is used to indicate the possibility of the alternative employer rejecting the current order, and the alternative employer whose rejection probability does not exceed the preset rejection probability threshold is used as the target employer for processing the current order. The requirement that the alternative funder can be used as the target funder for processing the current order has been improved. After the loan is successful, the probability that the funder risk of the target funder is greater than the preset risk threshold is reduced, and the need for the loan assistance platform to pay the penalty fee to the target funder is reduced. The probability increases the profit of the broader market profit. Secondly, it can be understood that the introduced fuzzy logic inference algorithm is applied to the capital risk control in asset allocation. By inputting variables, the entire capital risk control becomes simple and effective. Based on the capital risk value, order risk value and forecast If the risk threshold is input to the control module, the final decision strength value can be output, which can not only fuzzify the precise numerical value for reasoning, but also output the accurate decision strength numerical value to assist decision-making. Furthermore, the first membership function, the second membership function and the preset fuzzy rule base are all obtained through business experience and continuous online debugging. The whole process is controlled, and the control module is effective in any case. The real-time nature of capital screening and capital risk control has been improved.

The capital screening method in the embodiment of the present application has been described above, and the capital screening device in the embodiment of the present application is described below. Referring to FIG. 7 , an embodiment of the capital screening device in the embodiment of the present application includes:

Obtaining unit 701, configured to obtain the order risk value and the preset risk threshold of the current order;

A determination unit 702, configured to determine a plurality of candidate capitals, and the plurality of candidate capitals have corresponding capital risk values;

The processing unit 703 is configured to, for each candidate employer, use a fuzzy logic inference algorithm to obtain the current order risk value obtained by the obtaining unit 701, the preset risk threshold value, and the candidate determined by the determining unit 702. The capital risk value of the capital is processed by fuzzy logic to obtain the rejection probability corresponding to the candidate capital; the rejection probability is used to indicate the possibility of the candidate capital rejecting the current order;

As unit 704, it is configured to use the candidate employer whose rejection probability obtained by the processing unit 703 does not exceed the preset rejection probability threshold as the target employer for processing the current order.

In the embodiment of the present application, fuzzy logic processing can be performed on each candidate capital by using a fuzzy logic inference algorithm on the risk value of the current order, the preset risk threshold, and the capital risk value of the candidate capital, and the corresponding rejection of the candidate capital can be obtained. The probability of rejection is used to indicate the possibility of the candidate employer rejecting the current order, and the candidate employer whose rejection probability does not exceed the preset rejection probability threshold is used as the target employer for processing the current order. The requirement that the alternative funder can be used as the target funder for processing the current order has been improved. After the loan is successful, the probability that the funder risk of the target funder is greater than the preset risk threshold is reduced, and the need for the loan assistance platform to pay the penalty fee to the target funder is reduced. The probability increases the profit of the broader market profit.

The following describes the capital screening device in the embodiment of the present application in detail. Please refer to FIG. 8 . Another embodiment of the capital screening device in the embodiment of the present application includes:

Obtaining unit 801, configured to obtain the order risk value and the preset risk threshold of the current order;

A determination unit 802, configured to determine a plurality of candidate capitals, and the plurality of candidate capitals have corresponding capital risk values;

The processing unit 803 is configured to, for each of the candidate employers, use a fuzzy logic inference algorithm to obtain the current order risk value obtained by the obtaining unit 801, the preset risk threshold, and the candidate determined by the determining unit 802. The capital risk value of the capital is processed by fuzzy logic to obtain the rejection probability corresponding to the candidate capital; the rejection probability is used to indicate the possibility of the candidate capital rejecting the current order;

As the unit 804, it is configured to use the candidate employer whose rejection probability obtained by the processing unit 803 does not exceed the preset rejection probability threshold as the target employer for processing the current order.

The capital screening equipment further includes: a obtaining unit 805;

The obtaining unit 801 is further configured to obtain the historical order amount and historical order risk value of each of the alternative capitals;

The obtaining unit 805 is specifically configured to obtain the capital risk value of the candidate capital according to the historical order amount and historical order risk value of the candidate capital obtained by the obtaining unit 801 for each candidate capital.

The determining unit 802 is specifically configured to perform a weighted average of the historical order amount and the historical order risk value of the candidate capital obtained by the obtaining unit 801 to obtain the capital risk value of the candidate capital.

The obtaining unit 805 is specifically configured to use the difference between the capital risk value determined by the determining unit 802 and the preset risk threshold obtained by the obtaining unit 801 as the first input quantity, and compare the current order risk value with the selected risk value. The difference between the risk value of the capital and the capital is used as the second input, and the first input is input into the first membership function to obtain the membership of the first input, and the second input is input into the second membership function to obtain the first input. The membership degree of two input quantities, according to the membership degree of the first input quantity and the membership degree of the second input quantity, determine the target rejection willingness fuzzy flag corresponding to the first input quantity and the second input quantity; The membership degree of each target rejection willingness ambiguous mark is weighted and averaged by the membership degree of each target rejection willingness ambiguous mark and the preset weight corresponding to each target rejection willingness ambiguous mark to obtain the rejection probability.

The determining unit 802 is specifically configured to use, in the preset fuzzy rule base, the rejection intention fuzzy marks corresponding to the membership degree of the first input quantity and the membership degree of the second input quantity as the target rejection intention fuzzy mark, According to the membership degree of the first input quantity corresponding to the target rejection willingness ambiguous mark and the membership degree of the second input quantity, the strength value of the target rejection willingness ambiguous mark is determined, and according to the strength of at least one of the target rejection willingness ambiguous marks value to determine the membership of each target's refusal willingness ambiguous marker.

The determining unit 802 is specifically configured to use the smaller degree of membership among the degree of membership of the first input quantity and the degree of membership of the second input quantity obtained by the obtaining unit 805 corresponding to the target rejection willingness ambiguous mark as the minimum degree of membership. The intensity value of the ambiguous marker of the target's rejection willingness.

Specifically, the determining unit 802 determines, in at least one of the target rejection intention ambiguous marks, belonging to the same type of target rejection intention ambiguous marks, and sets the same type of target rejection intention ambiguous marks in the intensity value determined by the determining unit 802. The maximum value is taken as the membership degree of this kind of target rejection willingness ambiguous mark.

In this embodiment, each unit in the capital screening device performs the operations of the capital screening device in the embodiments shown in FIG. 1 and FIG. 2 , and details are not repeated here.

Referring to FIG. 9 below, another embodiment of the vehicle sharing device 900 in the embodiment of the present application includes:

Central processing unit 901, memory 905, input/output interface 904, wired or wireless network interface 903 and power supply 902;

The memory 905 is a short-term storage memory or a persistent storage memory;

The central processing unit 901 is configured to communicate with the memory 905 and execute the operations of the instructions in the memory 905 to perform the methods in the aforementioned embodiments shown in FIG. 1 and FIG. 2 .

Embodiments of the present application also provide a computer-readable storage medium, where the computer-readable storage medium includes instructions, when the instructions are executed on the computer, the computer is made to execute the methods in the foregoing embodiments shown in FIG. 1 and FIG. 2 .

Embodiments of the present application also provide a computer program product containing instructions, when the computer program product runs on a computer, the computer causes the computer to execute the methods in the foregoing embodiments shown in FIG. 1 and FIG. 2 .

It should be understood that, although the steps in the flowcharts involved in the above embodiments are sequentially displayed according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, the execution of these steps is not strictly limited to the order, and these steps may be performed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above embodiments may include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but may be performed at different times The execution order of these steps or phases is not necessarily sequential, but may be performed alternately or alternately with other steps or at least a part of the steps or phases in the other steps.

Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system, device and unit described above may refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

In the several embodiments provided in this application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the apparatus embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored, or not implemented. On the other hand, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.

The integrated unit, if implemented in the form of a software functional unit and sold or used as an independent product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the present application can be embodied in the form of software products in essence, or the parts that contribute to the prior art, or all or part of the technical solutions, and the computer software products are stored in a storage medium , including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: U disk, removable hard disk, read-only memory (ROM, read-only memory), random access memory (RAM, random access memory), magnetic disk or optical disk and other media that can store program codes.

Claims (10)

1. a method of screening for capital, is characterized in that, is applied to screening equipment of capital, comprising: Obtain the order risk value and preset risk threshold of the current order; determining a plurality of alternative capital parties, the plurality of alternative capital parties have corresponding capital party risk values; For each candidate capital, fuzzy logic is used to perform fuzzy logic processing on the risk value of the current order, the preset risk threshold and the capital risk value of the candidate capital, to obtain the candidate capital The corresponding rejection probability; the rejection probability is used to indicate the possibility that the alternative employer rejects the current order; The candidate employer whose rejection probability does not exceed a preset rejection probability threshold is used as the target employer for processing the current order. 2. The method according to claim 1, characterized in that, before performing fuzzy logic processing on the capital risk value, the current order risk value and the preset risk threshold by using a fuzzy logic inference algorithm, the The method also includes: Obtain the historical order amount and historical order risk value of each said alternative capital; For each candidate capital, the capital risk value of the candidate capital is obtained according to the historical order amount and historical order risk value of the candidate capital. 3. The method according to claim 2, wherein, determining the capital risk value of the alternative capital according to the historical order amount and historical order risk of the alternative capital, comprising: A weighted average of the historical order amount and the historical order risk value of the candidate capital is performed to obtain the capital risk value of the candidate capital. 4. The method according to claim 1, characterized in that, by using a fuzzy logic inference algorithm, fuzzy logic processing is performed on the capital risk value, the current order risk value and the preset risk threshold to obtain the result. The probability of rejection by the alternative employer to reject the current order, including: Taking the difference between the capital risk value and the preset risk threshold as the first input quantity, and taking the difference between the current order risk value and the capital risk value as the second input quantity; Inputting the first input into the first membership function to obtain the membership of the first input, and inputting the second input into the second membership function to obtain the membership of the second input; According to the membership degree of the first input quantity and the membership degree of the second input quantity, the target rejection willingness ambiguous mark corresponding to the first input quantity and the second input quantity and each target rejection willingness ambiguous mark are determined. affiliation; The rejection probability is obtained by performing a weighted average of the membership degree of each target rejection willingness ambiguous mark and the preset weight corresponding to each target rejection willingness ambiguous mark. 5 . The method according to claim 4 , wherein the first input quantity and the second input quantity are determined according to the membership degree of the first input quantity and the membership degree of the second input quantity. 6 . The fuzzy marks of target rejection willingness corresponding to the input amount and the membership degree of each target rejection willingness fuzzy mark, including: In the preset fuzzy rule base, the rejection intention fuzzy marks corresponding to the membership degree of the first input quantity and the membership degree of the second input quantity are used as the target rejection intention fuzzy mark; According to the membership degree of the first input quantity corresponding to the target rejection willingness ambiguous mark and the membership degree of the second input quantity, determine the intensity value of the target rejection willingness ambiguous mark; According to the strength value of at least one of the target rejection intention ambiguous marks, the membership degree of each target rejection intention ambiguous mark is determined. 6 . The method according to claim 5 , wherein the target rejection willingness ambiguity is determined according to the membership degree of the first input quantity and the membership degree of the second input quantity corresponding to the target rejection willingness ambiguous mark. 7 . Intensity values for markers, including: The smaller of the membership degree of the first input quantity corresponding to the target rejection willingness ambiguous mark and the membership degree of the second input quantity is taken as the strength value of the target rejection willingness ambiguous mark. 7 . The method according to claim 5 , wherein, according to the intensity value of at least one ambiguous marker of target rejection willingness, determining the membership degree of each target rejection willingness ambiguous marker, comprising: 8 . In at least one of the target rejection intention ambiguous marks, determine the target rejection intention ambiguous marks belonging to the same type; The maximum value of the intensity values of the same kind of target rejection willing ambiguous marks is taken as the membership degree of this kind of target rejection willing ambiguous marks. 8. A capital screening equipment, characterized in that, comprising: Obtaining unit, used to obtain the order risk value and preset risk threshold of the current order; a determining unit, configured to determine a plurality of candidate capitals, the plurality of candidate capitals have corresponding capital risk values; The processing unit is configured to perform fuzzy logic processing on the current order risk value, the preset risk threshold and the capital risk value of the alternative capital by using a fuzzy logic inference algorithm for each candidate capital, to obtain The rejection probability corresponding to the alternative capital; the rejection probability is used to indicate the possibility of the alternative capital rejecting the current order; As a unit, it is used for taking the candidate employer whose rejection probability does not exceed the preset rejection probability threshold as the target employer for processing the current order. 9. A capital screening equipment, characterized in that, comprising: Central processing unit, memory, input and output interface, wired or wireless network interface and power supply; the memory is a short-term storage memory or a persistent storage memory; The central processing unit is configured to communicate with the memory and execute operations of instructions in the memory to perform the method of any one of claims 1-7. 10. A computer-readable storage medium, characterized in that the computer-readable storage medium comprises instructions that, when executed on a computer, cause the computer to perform the method according to any one of claims 1 to 7. method.

Images