CN117952741A - Order distribution method, order distribution device, computer equipment and storage medium - Google Patents
Order distribution method, order distribution device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN117952741A CN117952741A CN202410144819.9A CN202410144819A CN117952741A CN 117952741 A CN117952741 A CN 117952741A CN 202410144819 A CN202410144819 A CN 202410144819A CN 117952741 A CN117952741 A CN 117952741A
- Authority
- CN
- China
- Prior art keywords
- order
- target
- sponsor
- candidate
- samples
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 61
- 238000012545 processing Methods 0.000 claims abstract description 31
- 238000012550 audit Methods 0.000 claims description 88
- 238000012549 training Methods 0.000 claims description 27
- 238000004590 computer program Methods 0.000 claims description 17
- 238000010586 diagram Methods 0.000 description 14
- 238000009736 wetting Methods 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 235000019800 disodium phosphate Nutrition 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000008685 targeting Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06316—Sequencing of tasks or work
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Accounting & Taxation (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Technology Law (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the application discloses an order distribution method, an order distribution device, computer equipment and a storage medium. The method comprises the following steps: acquiring a target order; determining the matching probability of each target candidate sponsor and each target order according to the order basic feature of the target order and the candidate sponsor feature of each target candidate sponsor in a plurality of preset target candidate sponsors based on a preset global allocation decision model, wherein the global allocation decision model is trained based on a target sample set, the target sample set comprises a plurality of order allocation samples, and the order allocation samples are samples for sponsor allocation of the order with preset indexes as targets in a historical preset statistical period; sorting all the target candidate sponsors based on the matching probability to obtain target sorting; and carrying out the sponsor allocation processing on the target orders according to the target ordering. The method for distributing orders can improve the profit margin of a large disc.
Description
Technical Field
The present application relates to the field of financial technologies, and in particular, to an order allocation method, an order allocation device, a computer device, and a storage medium.
Background
In the case of staged consumption or other lending operations, it is often necessary to carry out the lending by means of an intermediate lending platform, which often cooperates with a plurality of parties, and when a customer places an order at the front end, the customer gives the order as an asset to a group of parties for polling and auditing until the asset is audited by a party and payment is successful, i.e. the financing matching process.
The same order has different profit amounts in different sponsors, and different sponsors have different requirements of pricing, period number, risk and the like, so how to maximize profit amounts becomes a critical problem in capital asset matching under the condition of meeting the requirements of different sponsors.
In the prior art, after acquiring an order, the profit margin of the order at each sponsor is calculated, then each sponsor is ranked according to the profit margin, the order is preferentially allocated to the sponsor ranked in front, if the sponsor is ranked according to the profit margin only, the best ranking mode under the current order can not be considered at most, and the global situation can not be considered, each sponsor has the corresponding sponsor allowance, for example, the sponsor A has a limit of ten million per month, if the limit of ten million is used up in front, the order with high profit for the sponsor is subsequently received, and since the limit of the sponsor is used up, the order is not allocated to the sponsor, so a method is needed for allocating the order based on the global profit maximization.
Disclosure of Invention
The embodiment of the application provides an order distribution method, an order distribution device, computer equipment and a storage medium, which can be used for order distribution based on global profit maximization.
In a first aspect, an embodiment of the present application provides an order allocation method, including:
Acquiring a target order;
Acquiring order basic characteristics from the target order based on a preset global allocation decision model, determining the matching probability of each target candidate sponsor and the target order respectively according to the order basic characteristics and the candidate sponsor characteristics of each target candidate sponsor in a plurality of preset target candidate sponsors, wherein the global allocation decision model is trained based on a target sample set, the target sample set comprises a plurality of order allocation samples, and the order allocation samples are samples for sponsor allocation of the order with preset indexes as targets in a historical preset statistical period;
sorting all the target candidate sponsors based on the matching probability to obtain target sorting;
and carrying out the sponsor allocation processing on the target orders according to the target ordering.
In a second aspect, an embodiment of the present application further provides an order allocation apparatus, including:
The receiving and transmitting unit is used for acquiring a target order;
The processing unit is used for acquiring order basic characteristics from the target order based on a preset global allocation decision model, determining the matching probability of each target candidate sponsor and the target order respectively according to the order basic characteristics and the candidate sponsor characteristics of each target candidate sponsor in a plurality of preset target candidate sponsors, wherein the global allocation decision model is trained based on a target sample set, the target sample set comprises a plurality of order allocation samples, and the order allocation samples are samples for sponsor allocation of the order with preset indexes as targets in a historical preset statistical period; sorting all the target candidate sponsors based on the matching probability to obtain target sorting; and carrying out the sponsor allocation processing on the target orders according to the target ordering.
In a third aspect, an embodiment of the present application further provides a computer device, which includes a memory and a processor, where the memory stores a computer program, and the processor implements the method when executing the computer program.
In a fourth aspect, embodiments of the present application also provide a computer readable storage medium storing a computer program comprising program instructions which, when executed by a processor, implement the above-described method.
The embodiment of the application provides an order distribution method, an order distribution device, computer equipment and a storage medium. Wherein the method comprises the following steps: acquiring a target order; acquiring order basic characteristics from the target order based on a preset global allocation decision model, determining the matching probability of each target candidate sponsor and the target order respectively according to the order basic characteristics and the candidate sponsor characteristics of each target candidate sponsor in a plurality of preset target candidate sponsors, wherein the global allocation decision model is trained based on a target sample set, the target sample set comprises a plurality of order allocation samples, and the order allocation samples are samples for sponsor allocation of the order with preset indexes as targets in a historical preset statistical period; sorting all the target candidate sponsors based on the matching probability to obtain target sorting; and carrying out the sponsor allocation processing on the target orders according to the target ordering. After the embodiment of the application acquires the order, the matching probability of each sponsor is determined through the global allocation decision model, and because the training sample (order allocation sample) of the global allocation decision model provided by the embodiment is a sample for distributing the sponsor to the order by taking the preset index as the target, the closer the sponsor is to the preset index (such as maximizing the profit of the large disc), the larger the calculated matching probability is, so that the matching probability of each sponsor calculated according to the embodiment distributes the sponsor to the order, thereby improving the profit rate of the large disc.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present application, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic diagram of an application scenario of an order allocation method according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a training process for auditing a pass rate model in an order distribution method according to an embodiment of the present application;
FIG. 3 is a schematic diagram of a historical order in an order distribution method according to an embodiment of the present application;
FIG. 4 is a schematic diagram of a training flow of a global allocation decision model in an order allocation method according to an embodiment of the present application;
FIG. 5 is a schematic diagram illustrating a process of historical orders in an order distribution method according to an embodiment of the present application;
FIG. 6 is a flow chart of an order distribution method according to an embodiment of the present application;
FIG. 7 is a schematic block diagram of an order distribution device provided by an embodiment of the present application;
fig. 8 is a schematic block diagram of a computer device according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
It should be understood that the terms "comprises" and "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should be further understood that the term "and/or" as used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
The embodiment of the application provides an order distribution method, an order distribution device, computer equipment and a storage medium.
The execution main body of the order distribution method can be the order distribution device provided by the embodiment of the application or computer equipment integrated with the order distribution device, wherein the order distribution device can be realized in a hardware or software mode, and the computer equipment can be a terminal or a server.
Referring to fig. 1, fig. 1 is a schematic diagram of an application scenario of an order allocation method according to an embodiment of the present application. The order allocation method is applied to computer equipment in fig. 1, the computer equipment can be hardware equipment of an intermediate lending platform, in some embodiments, after the computer equipment acquires a target order, a target candidate sponsor which can pass the target order is determined based on an audit passing rate model, then an order basic feature is acquired from the target order based on a preset global allocation decision model, and the matching probability of each target candidate sponsor and each target order is determined according to the order basic feature and the candidate sponsor feature of each target candidate sponsor in a plurality of preset target candidate sponsors, wherein the global allocation decision model is trained based on a target sample set, the target sample set comprises a plurality of order allocation samples, and the order allocation samples are samples for sponsor allocation of the order with a preset index as a target in a history preset statistical period; then sorting all the target candidate sponsors based on the matching probability to obtain target sorting; and finally, carrying out sponsor allocation processing on the target orders according to the target ordering.
In this embodiment, an audit passing rate model and a global allocation decision model are preset in the computer device, where the audit passing rate model may specifically be a neural network model based on a Gate circulation unit (Gate RecurrentUnit, GRU) and is used to determine whether an order can be audited by a party; the global allocation decision model may be a two-class model based on a gradient lifting framework (LIGHT GRADIENT Boosting Machine, lightGBM), and the process of training the audit passing rate model and the global allocation decision model is described in detail below:
Auditing a passing rate model:
Fig. 2 is a schematic diagram of a training flow for auditing a pass rate model in an order allocation method according to an embodiment of the present application. As shown in fig. 2, the method includes the following steps S110 to S130.
S110, acquiring a plurality of historical orders in a historical preset statistical period, wherein each historical order comprises an order sample and a plurality of candidate sponsor samples.
Each candidate sponsor sample carries an audit identifier, wherein the audit identifier comprises an audit passing identifier, an audit failing identifier or an unverified identifier.
For example, the historical preset statistical period (such as the historical large-disc profit statistical period) is one month, and the embodiment can acquire historical orders in a plurality of historical preset statistical periods, then the training sample of each historical preset statistical period is used as a training sample of a training period, and the audit passing rate model and the global allocation decision model are trained through a plurality of training periods.
As shown in fig. 3, fig. 3 is a schematic diagram of a plurality of historical orders in a historical preset statistical period, wherein "0" in a column of "whether the audit passes" in fig. 3 indicates that the audit does not pass an identification, "1" indicates that the audit passes an identification, "null" indicates that the audit does not pass an identification, "sponsor id" is a unique identification of each candidate sponsor sample, and N is the total number of orders.
S120, determining the candidate sponsor sample carrying the audit passing mark and the corresponding order sample as an audit passing sample, and determining the candidate sponsor sample carrying the audit failing mark and the corresponding order sample as an audit failing sample.
Specifically, the order sample in the passing sample is an order basic characteristic of the corresponding order, and the order basic characteristic comprises characteristics such as order amount, order period number, order user information and the like; the candidate sponsor sample is the sponsor characteristics of the corresponding sponsor, including the sponsor credit limit of the sponsor, profit calculation rules and other characteristics.
S130, training the audit passing rate model according to the audit passing sample and the audit non-passing sample.
In this embodiment, after an audit passing sample and an audit failing sample are determined from a historical order, an audit passing rate model is trained according to the audit passing sample and the audit failing sample, and then audit results of all parties to the acquired order are predicted by using the audit passing rate model.
After the audit passing rate model in this embodiment is trained by a plurality of historical orders in a historical preset statistical period, in a subsequent use process, further optimization training is further required to be performed on the model, for example, every 3 days, the model is optimally trained by using the historical orders of the first 3 days.
Global allocation decision model:
Fig. 4 is a schematic diagram of a training flow of a global allocation decision model in an order allocation method according to an embodiment of the present application. As shown in fig. 4, the method includes the following steps S210 to S250.
S210, acquiring a plurality of historical orders in the historical preset statistical period, wherein each historical order comprises an order sample and a plurality of candidate sponsor samples.
This step is similar to step S110, and is not described here in detail, and in some embodiments, this step may be the same step as step S110.
S220, determining target candidate sponsor samples corresponding to the order samples respectively from a plurality of candidate sponsor samples according to a preset auditing passing rate model.
The audit passing rate model in this embodiment is an audit passing rate model trained by the corresponding embodiment in fig. 2, and the target candidate sponsor sample is a sponsor sample in the historical order that can pay successfully (i.e. pass the audit) based on the audit passing rate model.
Specifically, each candidate sponsor sample in the historical order carries an audit identifier, wherein the audit identifier comprises an audit passing identifier, an audit non-passing identifier or an unverified identifier; the candidate sponsor samples with the audit passing mark and the audit non-passing mark are sponsor samples with known audit results, so that the application can train the audit passing rate model by using the two candidate sponsor samples, then input the candidate sponsor samples with the non-audit mark and the corresponding order samples into the trained audit passing rate model to determine the audit result of the sponsor samples with the non-audit mark, and finally determine the audit result as the candidate sponsor sample with the audit passing mark and the candidate sponsor sample with the audit passing mark as the target candidate sponsor sample.
For example, as shown in fig. 5, "null" in the column, which is originally "whether or not to audit pass," is determined as "0" or "1".
S230, determining optimal sponsor samples corresponding to the order samples respectively from the corresponding target candidate sponsor samples by taking a preset index as a target and a preset constraint rule.
The preset constraint rule comprises a sponsor allocation quantity constraint and a sponsor quota constraint, and the preset index can be used for maximizing the profit of a large disc.
In this embodiment, it is specified in the constraint of the amount allocated by the sponsor that each order can be allocated to only one sponsor, that is, each order has only one optimal sponsor sample, and it is specified in the constraint of the amount of the sponsor that the sum of the amounts allocated to the orders in the historical preset statistics period cannot exceed the corresponding amount of the sponsor.
In this embodiment, OR-tools are used as a solver to calculate optimal tariff samples corresponding to each order sample.
It should be noted that, the optimal sponsor sample in this embodiment is a sponsor sample to be allocated to the corresponding order, which is obtained by targeting the preset index in the historical preset statistics period, and is not the sponsor actually allocated in the historical order (the same sponsor may be the same sponsor or different sponsors may be different).
In some embodiments, the preset constraint rules further include a sponsor allocation identification constraint, a split duty cycle constraint, a minimum sponsor level constraint, and the like.
For easy understanding, the following detailed description will be given of the objects and some constraints in this embodiment with specific formulas:
goal-maximize large disc profit:
Wherein N is the number of historical orders in a historical preset statistical period, i is an order sequence number, j is a sponsor sequence number, a ij is the distribution condition of a jth sponsor sample in an ith order sample, a ij has a value of 1 or 0, wherein when the value is 1, the corresponding sponsor sample is the optimal sponsor sample of the order, when the value is 0, the corresponding sponsor sample is not the optimal sponsor sample of the order, Profit for the jth sponsor sample in the ith order sample.
Constraint rules:
the above formula s.t is a constraint rule that includes a plurality of constraints.
Wherein, the formula (1) is an allocation rule constraint, and the value of a ij is 0 (not allocated) or 1 (allocated); the formula (2) allocates quantity constraint for the sponsors, each order can only be allocated to one sponsor, and C i is the number of the sponsors corresponding to the ith order; formula (3) is a partial wetting duty ratio constraint, the partial wetting duty ratio is specified to be more than 18%, s ij is the partial wetting condition in the jth sponsor of the ith order, the value of 0 is not partial wetting, the value of 1 is partial wetting, and amt i is the order amount of the ith order; equation (4) is the lowest magnitude constraint for a certain sponsor, e.g., the sponsor magnitude constraint for a sponsor with a sponsor id 1056 in fig. 3 is above 100 w.
In addition, constraint regulation is carried out according to actual needs in constraint rules, and for example, risk constraint, regional constraint and the like are included.
S240, determining the optimal sponsor sample as a positive sponsor sample corresponding to the order sample, and determining sponsor samples except the optimal sponsor sample in the target candidate sponsor sample as negative sponsor samples corresponding to the order sample, so as to determine the order distribution samples respectively corresponding to the order samples, and obtaining the target sample set.
In this embodiment, for each historical order, the resulting optimal sponsor sample is determined to be the positive sponsor sample for the corresponding order sample, and the other sponsor samples corresponding to that order sample are determined to be the negative sponsor samples for the corresponding order sample.
Specifically, when the optimal sponsor sample is determined to be a positive sponsor sample corresponding to the order sample, the optimal sponsor sample needs to be labeled with a positive sample, and other sponsor samples need to be labeled with negative samples.
It should be noted that the target sample set includes a plurality of order allocation samples, each of which includes an order sample, a positive party sample, and one or more negative party samples, where the order sample includes an order base feature of a corresponding order, and the party sample includes a party feature.
S250, training a global allocation decision model according to the target sample set.
In this embodiment, after the target sample set is obtained, the global allocation decision model is trained using the target sample set.
In some embodiments, before the training of the global allocation decision model from the set of target samples, the method further comprises: acquiring profit amounts of the target candidate sponsor samples on the corresponding order samples; for each historical order, determining in-order statistical sample characteristics of each target candidate sponsor sample according to profit amounts of each target candidate sponsor sample.
Wherein the intra-order statistical sample characteristics include at least one of a profit ranking, a profit difference, a maximum profit, and a minimum profit, the profit ranking being a profit ranking of the target candidate sponsor sample within a corresponding historical order, the profit difference being a profit margin difference of the target candidate sponsor sample and a maximum profit within the corresponding historical order, the maximum profit corresponding to the profit margin having a maximum value within the historical order, the minimum profit being the smallest of the profit margins in the corresponding historical order.
At this time, the global allocation decision model needs to be trained according to the target sample set and the intra-order statistical sample features.
After the global allocation decision model in this embodiment is trained by a plurality of historical orders in a historical preset statistical period, in a subsequent use process, further optimization training is further required to be performed on the model, for example, every month, the model is optimized and trained by using the historical orders of the previous month.
In the following, a detailed description is given of the order allocation method provided by the embodiment of the present application, and fig. 6 is a schematic flow chart of the order allocation method provided by the embodiment of the present application. As shown in fig. 6, the method includes the following steps S310 to S340.
S310, acquiring a target order.
In this embodiment, the target order is a currently acquired order, and after the target order is acquired, the order basic features of the target order including the features of order amount, order period number, order user information and the like are extracted.
S320, acquiring order basic features from the target order based on a preset global allocation decision model, and determining the matching probability of each target candidate sponsor and the target order according to the order basic features and the candidate sponsor features of each target candidate sponsor in a plurality of preset target candidate sponsors.
The global allocation decision model is trained based on a target sample set, the target sample set comprises a plurality of order allocation samples, and the order allocation samples are samples for performing sponsor allocation on orders by taking preset indexes as targets in a historical preset statistical period.
The candidate sponsor features of the target candidate sponsor comprise the features of sponsor quota, profit calculation rule and the like.
In some embodiments, a plurality of candidate sponsors are initially assigned to the target order, and the embodiment screens out target candidate sponsors that can be payed out for the target order from among the plurality of candidate sponsors, and then calculates a matching probability for each target candidate sponsor.
Specifically, the screening of target candidate sponsors is performed from a plurality of candidate sponsors by:
Determining a first audit result of each candidate sponsor on the target order according to the order basic characteristics and the candidate sponsor characteristics of a plurality of preset candidate sponsors based on a preset audit passing rate model; and determining the candidate sponsor with the first checking result passing the checking as the target candidate sponsor.
The audit passing rate model is a model trained by the corresponding embodiment of fig. 2, and the first audit result may be an audit passing probability. The threshold of the passing probability may be set to 0.5, or may be set to another value according to needs, and the embodiment of the present application is not limited to the specific value.
In some examples, step S320 specifically includes: acquiring the order basic characteristics from the target order; acquiring profit amounts of the target candidate sponsors on the target orders; determining the intra-order statistical characteristics of each target candidate sponsor according to the profit margin of each target candidate sponsor; inputting the basic features of the order, the features of each candidate sponsor and the statistical features in each order into the global allocation decision model for matching processing, and obtaining the matching probability of each target candidate sponsor and the target order respectively.
Wherein the in-order statistics include at least one of a profit ranking, a profit difference, a maximum profit, and a minimum profit, the profit ranking being a profit ranking of the target candidate sponsor sample in a corresponding historical order, the profit difference being a profit margin difference of the target candidate sponsor sample and a maximum profit in the corresponding historical order, the maximum profit corresponding to the profit margin having a maximum value in the historical order, the minimum profit being the smallest profit margin in the corresponding historical order.
S330, sorting all the target candidate sponsors based on the matching probability to obtain target sorting.
In this embodiment, the target candidate sponsors may be ranked according to the magnitude of the matching probability, where in the target ranking, the larger the value of the matching probability, the more advanced the ranking.
Specifically, in some embodiments, step S330 includes: determining the target candidate sponsor with the matching probability larger than or equal to a preset probability threshold as a first candidate sponsor, and determining the target candidate sponsor with the matching probability smaller than the probability threshold as a second candidate sponsor; sorting all the first candidate sponsors according to the matching probability to obtain a first sorting; ranking each second candidate sponsor according to the profit margin of each second candidate sponsor on the target order to obtain a second ranking; and determining the target sequence according to the first sequence and the second sequence.
That is, in ranking the sponsors, only the sponsors with higher matching probability are ranked from large to small according to the matching probability, and the sponsors with lower matching probability may be ranked from large to small according to profit margin, wherein the profit margin is the profit margin of the corresponding sponsor on the target order, specifically, the first ranking is before and the second ranking is after.
In some embodiments, for candidate sponsors that are not predicted to pass through the audit pass rate model, the target sequence may also be added, in particular, the rank may be added to the second rank together with the second candidate sponsor according to the size of the profit margin, i.e., the second rank may include candidate sponsors that are not predicted to pass through the audit pass rate model in addition to the second candidate sponsor.
S340, performing sponsor allocation processing on the target orders according to the target ordering.
In this embodiment, after the target ordering is obtained, the target order is preferentially allocated to the front-ranked sponsor for processing, and if the front-ranked sponsor refuses the target order, the target order is sequentially allocated to the next sponsor for processing until the sponsor accepts the target order. If the top ranking sponsor receives the target order, the target order is not required to be further distributed.
In summary, after the embodiment of the present application obtains the order, the matching probability of each sponsor is determined by the global allocation decision model, and since the training sample (order allocation sample) of the global allocation decision model provided in the present embodiment is a sample for performing sponsor allocation on the order with the preset index as the target, the closer the sponsor is to the target of maximizing the large-disc profit, the larger the calculated matching probability is, so that the matching probability of each sponsor calculated according to the present embodiment performs sponsor allocation on the order, and the large-disc profit rate can be improved.
Fig. 7 is a schematic block diagram of an order distribution device according to an embodiment of the present application. As shown in fig. 7, the present application further provides an order allocation device corresponding to the above order allocation method. The order allocation device comprises a unit for executing the order allocation method described above, which device may be configured in a terminal or a server. Specifically, referring to fig. 7, the order distribution device 700 includes a transceiver unit 701 and a processing unit 702.
A transceiver unit 701, configured to obtain a target order;
The processing unit 702 is configured to obtain an order basic feature from the target order based on a preset global allocation decision model, determine a matching probability of each target candidate sponsor and the target order according to the order basic feature and a candidate sponsor feature of each target candidate sponsor in a preset plurality of target candidate sponsors, where the global allocation decision model is trained based on a target sample set, and the target sample set includes a plurality of order allocation samples, where the order allocation samples are samples for sponsor allocation of the order with a preset index as a target in a historical preset statistical period; sorting all the target candidate sponsors based on the matching probability to obtain target sorting; and carrying out the sponsor allocation processing on the target orders according to the target ordering.
In some embodiments, before executing the step of determining the matching probabilities of the target candidate sponsors and the target order according to the order basic feature and the candidate sponsor feature of each target candidate sponsor of the target candidate sponsors, the processing unit 702 is further configured to:
Determining a first audit result of each candidate sponsor on the target order according to the order basic characteristics and the candidate sponsor characteristics of a plurality of preset candidate sponsors based on a preset audit passing rate model; and determining the candidate sponsor with the first checking result passing the checking as the target candidate sponsor.
In some embodiments, before executing the step of determining the matching probabilities of the target candidate sponsors and the target order according to the order basic feature and the candidate sponsor feature of each target candidate sponsor of the target candidate sponsors, the processing unit 702 is further configured to:
Acquiring a plurality of historical orders in the historical preset statistical period through the transceiver unit 701, wherein each historical order comprises an order sample and a plurality of candidate sponsor samples; determining target candidate sponsor samples corresponding to each order sample from a plurality of candidate sponsor samples according to a preset auditing passing rate model; determining optimal sponsor samples corresponding to the order samples respectively from the corresponding target candidate sponsor samples by taking a preset index as a target and a preset constraint rule, wherein the preset constraint rule comprises sponsor allocation quantity constraint and sponsor limit constraint; determining the optimal party samples as positive party samples corresponding to the order samples, and determining party samples except the optimal party samples in the target candidate party samples as negative party samples corresponding to the order samples to determine the order distribution samples respectively corresponding to the order samples, so as to obtain the target sample set; and training the global allocation decision model according to the target sample set.
In some embodiments, the processing unit 702 is further configured to, prior to performing the step of training the global allocation decision model based on the set of target samples:
acquiring profit amounts of the target candidate sponsor samples on the corresponding order samples; for each historical order, determining in-order statistical sample characteristics of each target candidate sponsor sample according to profit amounts of each target candidate sponsor sample;
at this time, the processing unit 702 is specifically configured to, when executing the step of training the global allocation decision model according to the target sample set:
and training the global allocation decision model according to the target sample set and the statistical sample characteristics in each order.
In some embodiments, each candidate sponsor sample carries an audit identifier, where the audit identifier includes an audit passing identifier, an audit failing identifier, or an unverified identifier; the processing unit 702 is further configured to, before executing the step of determining, from a plurality of candidate sponsor samples, a target candidate sponsor sample corresponding to each of the order samples according to the preset audit passing rate model:
Determining a candidate sponsor sample carrying the audit passing mark and a corresponding order sample as an audit passing sample, and determining a candidate sponsor sample carrying the audit non-passing mark and a corresponding order sample as an audit non-passing sample; training the audit passing rate model according to the audit passing sample and the audit non-passing sample;
At this time, when the step of determining, from a plurality of candidate sponsor samples, the target candidate sponsor sample corresponding to each of the order samples according to the preset audit passing rate model is performed, the processing unit 702 is specifically configured to:
inputting the candidate sponsor sample carrying the non-audited mark and the corresponding order sample into the audit passing rate model, and determining a second audit result of the candidate sponsor sample carrying the non-audited mark; and determining the candidate sponsor sample with the second checking result as the candidate sponsor sample passing the checking and carrying the checking mark as the target candidate sponsor sample.
In some embodiments, the processing unit 702 is specifically configured to, when executing the step of determining the matching probabilities of each target candidate sponsor and the target order according to the order basic feature and the candidate sponsor feature of each target candidate sponsor in the preset plurality of target candidate sponsors, obtain the order basic feature from the target order based on the preset global allocation decision model:
Acquiring the order basic characteristics from the target order; acquiring profit amounts of the target candidate sponsors on the target orders; determining the intra-order statistical characteristics of each target candidate sponsor according to the profit margin of each target candidate sponsor; inputting the basic features of the order, the features of each candidate sponsor and the statistical features in each order into the global allocation decision model for matching processing, and obtaining the matching probability of each target candidate sponsor and the target order respectively.
In some embodiments, when the processing unit 702 performs the ranking of the target candidate sponsors based on the matching probabilities, the target ranking step is specifically configured to:
Determining the target candidate sponsor with the matching probability larger than or equal to a preset probability threshold as a first candidate sponsor, and determining the target candidate sponsor with the matching probability smaller than the probability threshold as a second candidate sponsor; sorting all the first candidate sponsors according to the matching probability to obtain a first sorting; ranking each second candidate sponsor according to the profit margin of each second candidate sponsor on the target order to obtain a second ranking; and determining the target sequence according to the first sequence and the second sequence.
In summary, after the order allocation device 700 in the embodiment of the present application obtains the order, the matching probability of each sponsor is determined by the global allocation decision model, and since the training sample (order allocation sample) of the global allocation decision model provided in the embodiment is a sample for performing sponsor allocation on the order with the preset index as the target, the closer the sponsor is to the target of maximizing the profit on the large disc, the greater the calculated matching probability is, so that the matching probability of each sponsor calculated according to the embodiment performs sponsor allocation on the order, and the profit on the large disc can be improved.
It should be noted that, as those skilled in the art can clearly understand the specific implementation process of the order distribution device and each unit, reference may be made to the corresponding description in the foregoing method embodiments, and for convenience and brevity of description, details are not repeated herein.
The order distribution device described above may be implemented in the form of a computer program which is executable on a computer device as shown in fig. 8.
Referring to fig. 8, fig. 8 is a schematic block diagram of a computer device according to an embodiment of the present application. The computer device 800 may be a terminal or a server.
With reference to FIG. 8, the computer device 800 includes a processor 802, memory, and a network interface 805 connected by a system bus 801, wherein the memory may include a non-volatile storage medium 803 and an internal memory 804.
The nonvolatile storage medium 803 may store an operating system 8031 and a computer program 8032. The computer program 8032 includes program instructions that, when executed, cause the processor 802 to perform an order distribution method.
The processor 802 is used to provide computing and control capabilities to support the operation of the overall computer device 800.
The internal memory 804 provides an environment for the execution of the computer program 8032 in the non-volatile storage medium 803, which computer program 8032, when executed by the processor 802, causes the processor 802 to perform an order allocation method.
The network interface 805 is used for network communication with other devices. It will be appreciated by those skilled in the art that the architecture shown in fig. 8 is merely a block diagram of some of the architecture associated with the present inventive arrangements and is not limiting of the computer device 800 to which the present inventive arrangements may be applied, and that a particular computer device 800 may include more or less components than those shown, or may combine some of the components, or have a different arrangement of components.
Wherein the processor 802 is configured to execute a computer program 8032 stored in the memory, so as to implement the following steps:
Acquiring a target order;
Acquiring order basic characteristics from the target order based on a preset global allocation decision model, determining the matching probability of each target candidate sponsor and the target order respectively according to the order basic characteristics and the candidate sponsor characteristics of each target candidate sponsor in a plurality of preset target candidate sponsors, wherein the global allocation decision model is trained based on a target sample set, the target sample set comprises a plurality of order allocation samples, and the order allocation samples are samples for sponsor allocation of the order with preset indexes as targets in a historical preset statistical period;
sorting all the target candidate sponsors based on the matching probability to obtain target sorting;
and carrying out the sponsor allocation processing on the target orders according to the target ordering.
It should be appreciated that in embodiments of the present application, the Processor 802 may be a central processing unit (Central Processing Unit, CPU), the Processor 802 may also be other general purpose processors, digital signal processors (DIGITAL SIGNAL processors, DSPs), application SPECIFIC INTEGRATED Circuits (ASICs), off-the-shelf Programmable gate arrays (Field-Programmable GATEARRAY, FPGA) or other Programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. Wherein the general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
Those skilled in the art will appreciate that all or part of the flow in a method embodying the above described embodiments may be accomplished by computer programs instructing the relevant hardware. The computer program comprises program instructions, and the computer program can be stored in a storage medium, which is a computer readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the flow steps of the embodiments of the method described above.
Accordingly, the present application also provides a storage medium. The storage medium may be a computer readable storage medium. The storage medium stores a computer program, wherein the computer program includes program instructions. The program instructions, when executed by the processor, cause the processor to perform the steps of:
Acquiring a target order;
Acquiring order basic characteristics from the target order based on a preset global allocation decision model, determining the matching probability of each target candidate sponsor and the target order respectively according to the order basic characteristics and the candidate sponsor characteristics of each target candidate sponsor in a plurality of preset target candidate sponsors, wherein the global allocation decision model is trained based on a target sample set, the target sample set comprises a plurality of order allocation samples, and the order allocation samples are samples for sponsor allocation of the order with preset indexes as targets in a historical preset statistical period;
sorting all the target candidate sponsors based on the matching probability to obtain target sorting;
and carrying out the sponsor allocation processing on the target orders according to the target ordering.
The storage medium may be a U-disk, a removable hard disk, a Read-Only Memory (ROM), a magnetic disk, or an optical disk, or other various computer-readable storage media that can store program codes.
Those of ordinary skill in the art will appreciate that the elements and algorithm steps described in connection with the embodiments disclosed herein may be embodied in electronic hardware, in computer software, or in a combination of the two, and that the elements and steps of the examples have been generally described in terms of function in the foregoing description to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
In the several embodiments provided by the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. For example, the device embodiments described above are merely illustrative. For example, the division of each unit is only one logic function division, and there may be another division manner in actual implementation. For example, multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed.
The steps in the method of the embodiment of the application can be sequentially adjusted, combined and deleted according to actual needs. The units in the device of the embodiment of the application can be combined, divided and deleted according to actual needs. In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The integrated unit may be stored in a storage medium if implemented in the form of a software functional unit and sold or used as a stand-alone product. Based on such understanding, the technical solution of the present application is essentially or a part contributing to the prior art, or all or part of the technical solution may be embodied in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a terminal, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application.
While the application has been described with reference to certain preferred embodiments, it will be understood by those skilled in the art that various changes and substitutions of equivalents may be made and equivalents will be apparent to those skilled in the art without departing from the scope of the application. Therefore, the protection scope of the application is subject to the protection scope of the claims.
Claims (10)
1. An order allocation method, comprising:
Acquiring a target order;
Acquiring order basic characteristics from the target order based on a preset global allocation decision model, determining the matching probability of each target candidate sponsor and the target order respectively according to the order basic characteristics and the candidate sponsor characteristics of each target candidate sponsor in a plurality of preset target candidate sponsors, wherein the global allocation decision model is trained based on a target sample set, the target sample set comprises a plurality of order allocation samples, and the order allocation samples are samples for sponsor allocation of the order with preset indexes as targets in a historical preset statistical period;
sorting all the target candidate sponsors based on the matching probability to obtain target sorting;
and carrying out the sponsor allocation processing on the target orders according to the target ordering.
2. The method of claim 1, wherein the method further comprises, prior to obtaining order base features from the target order based on a preset global allocation decision model, determining a probability of matching each of the target candidate sponsors with the target order based on the order base features and candidate sponsor features of each of a plurality of target candidate sponsors:
Determining a first audit result of each candidate sponsor on the target order according to the order basic characteristics and the candidate sponsor characteristics of a plurality of preset candidate sponsors based on a preset audit passing rate model;
And determining the candidate sponsor with the first checking result passing the checking as the target candidate sponsor.
3. The method of claim 1, wherein the method further comprises, prior to obtaining order base features from the target order based on a preset global allocation decision model, determining a probability of matching each of the target candidate sponsors with the target order based on the order base features and candidate sponsor features of each of a plurality of target candidate sponsors:
Acquiring a plurality of historical orders in the historical preset statistical period, wherein each historical order comprises an order sample and a plurality of candidate sponsor samples;
Determining target candidate sponsor samples corresponding to each order sample from a plurality of candidate sponsor samples according to a preset auditing passing rate model;
determining optimal sponsor samples corresponding to the order samples respectively from the corresponding target candidate sponsor samples by taking a preset index as a target and a preset constraint rule, wherein the preset constraint rule comprises sponsor allocation quantity constraint and sponsor limit constraint;
Determining the optimal party samples as positive party samples corresponding to the order samples, and determining party samples except the optimal party samples in the target candidate party samples as negative party samples corresponding to the order samples to determine the order distribution samples respectively corresponding to the order samples, so as to obtain the target sample set;
and training the global allocation decision model according to the target sample set.
4. A method according to claim 3, wherein prior to said training said global allocation decision model from said set of target samples, said method further comprises:
Acquiring profit amounts of the target candidate sponsor samples on the corresponding order samples;
for each historical order, determining in-order statistical sample characteristics of each target candidate sponsor sample according to profit amounts of each target candidate sponsor sample;
the training the global allocation decision model according to the target sample set includes:
and training the global allocation decision model according to the target sample set and the statistical sample characteristics in each order.
5. A method according to claim 3, wherein each candidate sponsor sample carries audit identifiers including audit pass identifiers, audit fail identifiers or non-audit identifiers; before determining the target candidate sponsor samples corresponding to the order samples respectively from the plurality of candidate sponsor samples according to a preset auditing passing rate model, the method further comprises:
determining a candidate sponsor sample carrying the audit passing mark and a corresponding order sample as an audit passing sample, and determining a candidate sponsor sample carrying the audit non-passing mark and a corresponding order sample as an audit non-passing sample;
Training the audit passing rate model according to the audit passing sample and the audit non-passing sample;
The determining, according to a preset audit passing rate model, a target candidate sponsor sample corresponding to each order sample from a plurality of candidate sponsor samples includes:
Inputting the candidate sponsor sample carrying the non-audited mark and the corresponding order sample into the audit passing rate model, and determining a second audit result of the candidate sponsor sample carrying the non-audited mark;
And determining the candidate sponsor sample with the second checking result as the candidate sponsor sample passing the checking and carrying the checking mark as the target candidate sponsor sample.
6. The method of claim 1, wherein the obtaining order base features from the target order based on the preset global allocation decision model, determining a matching probability of each target candidate sponsor with the target order according to the order base features and candidate sponsor features of each target candidate sponsor of a preset plurality of target candidate sponsors, comprises:
acquiring the order basic characteristics from the target order;
Acquiring profit amounts of the target candidate sponsors on the target orders;
determining the intra-order statistical characteristics of each target candidate sponsor according to the profit margin of each target candidate sponsor;
Inputting the basic features of the order, the features of each candidate sponsor and the statistical features in each order into the global allocation decision model for matching processing, and obtaining the matching probability of each target candidate sponsor and the target order respectively.
7. The method according to any one of claims 1 to 6, wherein said ranking each of said target candidate sponsors based on said matching probabilities, resulting in a target ranking, comprises:
Determining the target candidate sponsor with the matching probability larger than or equal to a preset probability threshold as a first candidate sponsor, and determining the target candidate sponsor with the matching probability smaller than the probability threshold as a second candidate sponsor;
Sorting all the first candidate sponsors according to the matching probability to obtain a first sorting;
ranking each second candidate sponsor according to the profit margin of each second candidate sponsor on the target order to obtain a second ranking;
And determining the target sequence according to the first sequence and the second sequence.
8. An order dispensing device, comprising:
The receiving and transmitting unit is used for acquiring a target order;
The processing unit is used for acquiring order basic characteristics from the target order based on a preset global allocation decision model, determining the matching probability of each target candidate sponsor and the target order respectively according to the order basic characteristics and the candidate sponsor characteristics of each target candidate sponsor in a plurality of preset target candidate sponsors, wherein the global allocation decision model is trained based on a target sample set, the target sample set comprises a plurality of order allocation samples, and the order allocation samples are samples for sponsor allocation of the order with preset indexes as targets in a historical preset statistical period; sorting all the target candidate sponsors based on the matching probability to obtain target sorting; and carrying out the sponsor allocation processing on the target orders according to the target ordering.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the order allocation method of any of claims 1-7 when the computer program is executed.
10. A storage medium storing a computer program comprising program instructions which, when executed by a processor, cause the processor to perform the order allocation method of any of claims 1-7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410144819.9A CN117952741A (en) | 2024-02-01 | 2024-02-01 | Order distribution method, order distribution device, computer equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202410144819.9A CN117952741A (en) | 2024-02-01 | 2024-02-01 | Order distribution method, order distribution device, computer equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117952741A true CN117952741A (en) | 2024-04-30 |
Family
ID=90791928
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202410144819.9A Pending CN117952741A (en) | 2024-02-01 | 2024-02-01 | Order distribution method, order distribution device, computer equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117952741A (en) |
-
2024
- 2024-02-01 CN CN202410144819.9A patent/CN117952741A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109389431B (en) | Method and device for distributing coupons, electronic equipment and readable storage medium | |
Puterman | Markov decision processes: discrete stochastic dynamic programming | |
Kobren et al. | Paper matching with local fairness constraints | |
CN107679946A (en) | Fund Products Show method, apparatus, terminal device and storage medium | |
CN110443715B (en) | Fund product recommendation method, device, equipment and computer readable storage medium | |
CN107705211A (en) | A kind of adjustment method and apparatus of settlement of insurance claim case | |
CN112381428A (en) | Business allocation method, device, equipment and storage medium based on reinforcement learning | |
CN110009417B (en) | Target customer screening method, device, equipment and computer readable storage medium | |
CN111523748B (en) | Case distribution method, device and system | |
CN111881158A (en) | Management report data processing method and device, computer system and readable storage medium | |
US20240212041A1 (en) | Automatic data segmentation system | |
CN108416684A (en) | A kind of credibility appraisal procedure, device and the server of account main body | |
CN118134176A (en) | Task allocation method and device, electronic equipment and storage medium | |
CN111061948A (en) | User label recommendation method and device, computer equipment and storage medium | |
CN114625957A (en) | Payment method, device and equipment based on personalized recommendation of payment mode | |
CN111598632B (en) | Method and device for determining equity shares and equity share sequence | |
US7974903B2 (en) | System and method for analyzing distributions for taxation analysis | |
CN113377823A (en) | Value data processing method, device, equipment and storage medium | |
CN117236941A (en) | Enterprise grading-based payment settlement method, device, equipment and medium | |
CN117952741A (en) | Order distribution method, order distribution device, computer equipment and storage medium | |
CN114925261A (en) | Keyword determination method, apparatus, device, storage medium and program product | |
CN112116459B (en) | Data processing method and device based on lending | |
CN115277856B (en) | Flow screening method and system | |
CN113538157B (en) | Annuity cost management method and annuity cost management device | |
CN112862602B (en) | User request determining method, storage medium and electronic device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |