Transaction data processing method and transaction data processing system

Abstract

The invention relates to a transaction data processing method and system. The method comprises the steps of obtaining target proportion vectors of a target in different project categories, and constructing a target proportion vector model sample based on the target proportion vectors; constructing an association model based on the target proportion vector model sample and the association probability of associating the target from one project category to another project category; constructing a change model based on the target proportion vector model sample, the association probability and the change association probability of changing the target from one project category to another project category; performing parameter fitting on parameters of a change model based on the target proportion vector model sample to obtain the change association probability; and calculating target-related decisions of different item categories based on the change association probability. Therefore, the change of the proportion of the target in different project categories caused by the participation of the action object can be efficiently and accurately found out, and the resource configuration can be optimized and the effective risk management can be supported by accurately finding out the change of the proportion.

Description

Technical Field

[0001] The present invention relates to computer information technology, and in particular to a transaction data processing method and a transaction data processing system. Background Art

[0002] With the development of computer Internet technology, there are more and more types and methods of transactions. In order to stimulate consumption, the government, financial institutions or enterprises will release resources with certain rights and interests, such as consumer coupons. Users can enjoy the rights and interests of the resources by collecting resources and going to designated places for transaction verification, which can stimulate user consumption to a certain extent.

[0003] However, the funds for issuing consumption vouchers are limited, and the effects of distributing consumption vouchers in different proportions to different industries are mixed. Therefore, how to distribute consumption vouchers, especially in which industries to distribute them and in what proportion to distribute them, in order to achieve improved and optimized effects of consumption voucher distribution is also a key technical issue that needs to be solved.

[0004] On the other hand, the existing methods for issuing consumer vouchers are mainly aimed at a single customer group, which makes it difficult to meet the government's industry-oriented allocation needs. The use of manual methods to determine the target industries and the allocation of issuance quotas to each industry based on relevant technologies has led to poor results in the issuance of consumer vouchers. No effective solution has been proposed so far. Summary of the invention

[0005] In order to solve the above-mentioned problems in the prior art, the present invention aims to provide a transaction data processing method and a transaction data processing system that can efficiently and accurately find out the changes in the proportion of the target in different project categories caused by the participation of the object of action.

[0006] Furthermore, the present invention aims to provide a transaction data processing method and a transaction data processing system that can efficiently and accurately release resources with rights and interests (such as consumption coupons).

[0007] A transaction data processing method according to one aspect of the present invention comprises:

[0008] A model sample construction step, obtaining target proportion vectors of targets in different project categories, and constructing a target proportion vector model sample based on the target proportion vectors;

[0009] The association model building step is to build an association model based on the target proportion vector model sample and the association probability of the target from one project category to another project category;

[0010] A change model building step, building a change model based on the target proportion vector model sample, the association probability and the change association probability of the target changing from one project category to another project category;

[0011] a parameter fitting step, performing parameter fitting on the parameters of the change model based on the target proportion vector model sample to obtain the change association probability; and

[0012] The decision calculation step is to calculate the decisions related to the target for different project categories based on the change association probability.

[0013] Optionally, in the change model building step, a change model is built based on the target proportion vector model sample, the association probability, the change association probability and an effector object identifier for identifying whether to participate in a transaction activity.

[0014] Optionally, in the association model building step, the association model is built based on a Markov process.

[0015] In the step of constructing the change model, the active samples participating in the transaction activities and the inactive samples not participating in the transaction activities are incorporated into the Markov process to construct the change model.

[0016] Optionally, in the parameter fitting step, any one of the following fitting models is used for parameter fitting:

[0017] Least squares estimation fitting model, maximum likelihood estimation fitting model, moment estimation fitting model, least absolutes estimation fitting model, generalized moment estimation fitting model.

[0018] Optionally, when there are P items of different categories, the association probability is represented by a P×P dimensional association matrix, and the variable association probability is represented by a P×P dimensional association matrix, where P is a natural number.

[0019] Optionally, in the decision calculation step, as the decision related to the target, the sum of each column of the P columns of the change association probability is calculated and sorted to obtain a priority ranking.

[0020] Optionally, in the decision calculation step, as the decision related to the target, n category items from P categories are selected, and the proportions of the n category items are calculated respectively.

[0021] Optionally, in the model sample construction step, the consumption proportion vector of consumers in different industry categories is obtained as the target proportion vector and the model sample is constructed according to the following formula (1), where it is assumed that there are P different industries:

[0022] Y it =(yit,1 ,y it,2 ,…,y it,P ) T (1)

[0023] Among them, Y it is the consumption proportion vector, i represents the i-th sample, t represents the t-th moment, t-1 represents the t-1-th moment,

[0024] In the association model construction step, the association model is constructed based on the Markov process according to the following formula (2):

[0025] Y it =WY it-1 +∈ it (2)

[0026] Where W = (w p1p2 )∈R P×P is the association probability, w p1p2 represents the transfer probability of consumers’ consumption tendency in industry p2 to industry p1, ∈ it is the noise term,

[0027] In the change model construction step, the transition probability change model is constructed according to the following formula (3):

[0028] Y it =WY it-1 +Treat i *ΩY it-1 +∈ it (3)

[0029] Where, Ω=(τ p1p2 )∈R P×P is the probability change matrix, τ p1p2 represents the probability of consumers changing their consumption from industry p2 to industry p1, ∈ it is the noise term, Treat i Indicates the object identifier.

[0030] Optionally, in the association model building step, by training W so that WY it-1 +∈ it With Y it The error is minimal.

[0031] Optionally, in the decision calculation step, the probability change matrix is ​​column-added and sorted, and the resulting sorting is used as the priority sorting of the consumer voucher issuance industry.

[0032] Optionally, in the decision-making calculation step, n industries are selected to calculate the proportion of the n industries and use it as the proportion of the industry for the consumption voucher issuance.

[0033] A transaction data processing system according to one aspect of the present invention comprises:

[0034] A model sample construction module, used to obtain target proportion vectors of targets in different project categories, and to construct a target proportion vector model sample based on the target proportion vectors;

[0035] An association model building module, used to build an association model based on the target proportion vector model sample and the association probability of the target from one project category to another project category;

[0036] A change model building module, used to build a change model based on the target proportion vector model sample, the association probability and the change association probability of the target changing from one project category to another project category;

[0037] a parameter fitting module, configured to perform parameter fitting on the parameters of the change model based on the target proportion vector model sample to obtain the change association probability; and

[0038] The decision calculation module is used to calculate the decisions related to the target of different project categories based on the change association probability.

[0039] Optionally, in the change model construction module, a change model is constructed based on the target proportion vector model sample, the association probability, the change association probability and an effect object identifier for identifying whether to participate in the transaction activity.

[0040] In the step of constructing the change model, the active samples participating in the transaction activities and the inactive samples not participating in the transaction activities are incorporated into the Markov process to construct the change model.

[0041] Optionally, in the association model building module, the association model is built based on a Markov process.

[0042] Optionally, in the parameter fitting module, any one of the following fitting models is used for parameter fitting:

[0043] Least squares estimation fitting model, maximum likelihood estimation fitting model, moment estimation fitting model, least absolutes estimation fitting model, generalized moment estimation fitting model.

[0044] Optionally, when there are P items of different categories, the association probability is represented by a P×P dimensional association matrix, and the variable association probability is represented by a P×P dimensional association matrix, where P is a natural number.

[0045] Optionally, in the decision calculation module, as the decision related to the target, the sum of each column of the P columns of the change association probability is calculated and sorted to obtain a priority ranking.

[0046] Optionally, in the decision calculation module, as the decision related to the target, n category items from P categories are selected, and the proportions of the n category items are calculated respectively.

[0047] Optionally, in the model sample construction module, the consumption proportion vector of consumers in different industry categories is obtained as the target proportion vector and the model sample is constructed according to the following formula (1), where it is assumed that there are P different industries:

[0048] Y it =(y it,1 ,y it,2 ,…,y it,P ) T (1)

[0049] Among them, Y it is the consumption proportion vector, i represents the i-th sample, t represents the t-th moment, t-1 represents the t-1-th moment,

[0050] In the association model building module, the association model is built based on the Markov process according to the following formula (2):

[0051] Y it =WY it-1 +∈ it (2)

[0052] Where W = (w p1p2 )∈R P×P is the association probability, w p1p2 represents the transfer probability of consumers’ consumption tendency in industry p2 to industry p1, ∈ it is the noise term,

[0053] In the change model construction module, the transition probability change model is constructed according to the following formula (3):

[0054] Y it =WY it-1 +Treat i *ΩY it-1 +∈ it (3)

[0055] Where, Ω=(τ p1p2 )∈R P×P is the probability change matrix, τ p1p2 represents the probability of consumers changing their consumption from industry p2 to industry p1, ∈ it is the noise term, Treat i Indicates the object identifier.

[0056] Optionally, in the association model building module, by training W so that WY it-1 +∈ it With Y it The error is minimal.

[0057] Optionally, in the decision calculation module, the probability change matrix is ​​column-summed and sorted, and the obtained priority ranking is used as the ranking of the consumer voucher issuance industry.

[0058] Optionally, in the decision-making calculation module, n industries are selected to calculate the proportion of the n industries and use it as the proportion of the industry for the consumption voucher issuance.

[0059] A computer-readable medium according to one aspect of the present invention stores a computer program, which implements the above-mentioned transaction data processing method when executed by a processor.

[0060] A computer device according to one aspect of the present invention comprises a storage module, a processor, and a computer program stored in the storage module and executable on the processor, wherein the processor implements the above-mentioned transaction data processing method when executing the computer program.

[0061] A computer program product according to one aspect of the present invention comprises a computer program, which implements the above-mentioned transaction data processing method when executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a flow chart of the transaction data processing method of the present invention.

[0063] Figure 2 It is a structural block diagram of the transaction data processing system of the present invention.

[0064] The above and other objects and advantages of the present application will become more fully apparent from the following detailed description in conjunction with the accompanying drawings, in which the same or similar elements are represented by the same reference numerals. DETAILED DESCRIPTION

[0065] The following describes some of the various embodiments of the present invention, which are intended to provide a basic understanding of the present invention, but are not intended to identify the key or decisive elements of the present invention or to limit the scope of protection.

[0066] For the purpose of brevity and illustration, the principles of the present invention are described herein mainly with reference to its exemplary embodiments. However, those skilled in the art will readily recognize that the same principles are equally applicable to all types of transaction data processing methods and transaction data processing systems, and these same principles can be implemented therein, and any such changes do not depart from the true spirit and scope of the present patent application.

[0067] Moreover, in the following description, reference is made to the accompanying drawings, which illustrate specific exemplary embodiments. Electrical, mechanical, logical and structural changes may be made to these embodiments without departing from the spirit and scope of the present invention. In addition, although the features of the present invention are disclosed in conjunction with only one of several implementations / embodiments, this feature may be combined with one or more other features of other implementations / embodiments as may be desirable and / or advantageous for any given or identifiable function. Therefore, the following description should not be regarded as limiting, and the scope of the present invention is defined by the appended claims and their equivalents.

[0068] Terms such as “having” and “including” indicate that in addition to the units (modules) and steps directly and explicitly stated in the specification and claims, the technical solution of the present invention does not exclude the situation where it has other units (modules) and steps that are not directly or explicitly stated.

[0069] The transaction data processing method and transaction data processing system of the present invention can be applied not only to the issuance of consumption vouchers, but also to various aspects such as risk prevention and control and e-commerce. In the following description, for the convenience of explanation, consumption vouchers are taken as an example.

[0070] Figure 1 It is a flow chart of the transaction data processing method of the present invention.

[0071] like Figure 1 As shown, the transaction data processing method of the present invention comprises the following steps:

[0072] Model sample construction step S100: Obtain the consumption proportion vector Y of consumers in different industry categories it , build a model sample:

[0073] Assume there are P different industries, Y it That is a P-dimensional vector:

[0074] Y it =(y it,1 ,y it,2 ,…,y it,P ) T ,

[0075] Among them, i represents the i-th sample, and t represents the t-th time;

[0076] Association model construction step S200: constructing an industry consumption association model based on the Markov process:

[0077] Y it =WY it-1 +∈ it ,

[0078] Where W = (w p1p2 )∈R P×P is a P×P dimensional transfer matrix, w p1p2 represents the transfer probability of consumers’ consumption tendency in industry p2 to industry p1, ∈ it is the noise term;

[0079] Change model construction step S300: integrating the identification of whether to participate in the activity into the model fitting process to construct a transition probability change model:

[0080] Y it =WY it-1 +Treat i *ΩY it-1 +∈ it ,

[0081] Where, Ω=(τ p1p2 )∈R P×P is a P×P dimensional probability change matrix, τ p1p2 represents the probability of consumers changing their consumption from industry p2 to industry p1, ∈ it is the noise term, Treat i Indicates whether the sample is an object of action participating in the activity. This is an important inventive point of the present invention, that is, the object of action participating in the activity is integrated into the model fitting process;

[0082] Parameter fitting step S400: fitting the parameters of the transition probability change model incorporating intervention measures constructed in the transition probability change model construction step S300 based on the samples constructed in the model sample construction step S100, where the parameter estimation model includes but is not limited to an OLS least squares estimation fitting model; and

[0083] Decision calculation step S500: Calculate the priority ranking and proportion of consumption voucher placement decisions. Specifically, based on the probability change matrix Ω obtained in the parameter fitting step S400, calculate the sum of each column of the matrix and sort them to obtain the priority ranking of consumption voucher placement, and select the top n (where n is determined by the number of possible industries), calculate the investment proportion of n industries.

[0084] According to the transaction data processing method of the present invention, it is possible to efficiently and accurately find out the changes in the proportion of targets in different project categories due to the participation of the objects of action, and by accurately finding out the changes in the proportions, it is possible to optimize resource allocation and support effective risk management.

[0085] Next, an embodiment is listed below for specific description, in which the example of issuing consumption coupons in a certain area is used for description.

[0086] The transaction data processing method of this embodiment includes the following steps:

[0087] (1) Obtain the consumption proportion vector Y of consumers in different industry categories based on the indicator statistical method it , construct model training samples, and assign labels to users whether they participated in the consumption voucher activity, with participation as 1 and otherwise as 0. For ease of explanation, the three industries of catering, retail, and electronic products are used as examples. For example, the consumption proportion vector of the i-th user in the catering, retail, and electronic products industries at time t is Y it =[0.3,0.2,0.5], the consumption proportion vector Y of the i-th user in the catering, retail, and electronic products industries at the t+1th moment it+1 =[0.2,0.2,0.6].

[0088] (2) The industry consumption correlation model based on Markov process is constructed as follows:

[0089] Y it =WY it-1 +∈ it ,

[0090] Among them, Markov Process is an important type of random process, and its main characteristic is "no memory" or "Markov property". Specifically, if the future state of a system depends only on the current state and has nothing to do with the past state, then the system or process has Markov property and can be called a Markov process. Markov process is memoryless (the future state is only related to the current state and has nothing to do with the past state). Markov process can describe the transition probability (describes the probability of transitioning from one state to another). The present invention utilizes the characteristic that Markov process can describe the transition probability.

[0091] Where W = (w p1p2 )∈R 3×3 is a 3×3 dimensional incidence matrix, w p1p2 represents the probability of consumers' consumption in industry p2 being associated with industry p1, ∈ it is the noise term. Here, W is the parameter to be fitted, ∈ it is a random number term, including but not limited to a normal distribution, and has a normal distribution characteristic with a mean of 0 and a variance of σ2. Here, Y it-1 With Y it All are inputs to the model. By training W, WY it-1 +∈ it With Y itThe error is the smallest. The fitting process will be implemented in (5), where W reflects the consumption correlation between industries.

[0092] (3) The transition probability change model of the integrated participation activity identification is constructed as follows:

[0093] Y it =WY it-1 +Treat i *ΩY it-1 +∈ it

[0094] Where, Ω=(τ p1p2 )∈R3 ×3 is a 3×3 dimensional probability change matrix, τ p1p2 represents the probability of consumers' consumption in industry p2 being associated with industry p1, ∈ it is the noise term, Treat i Indicates whether the sample is the target of the intervention measure, that is, whether it participates in the transaction activity. If it has acted, it is marked as 1, and if it has not acted, it is marked as 0. Here, Ω is the parameter to be fitted, and ∈_it is a random number item, including but not limited to a normal distribution, and a random number with a normal distribution characteristic of 0 and a variance of σ2. Compared with (3), this is to incorporate the target identification of whether to participate in the activity. Therefore, Ω can better reflect the changes in the consumption correlation relationship between industries brought about by the release of consumer coupons.

[0095] Here, by incorporating multiple types of samples (i.e., active samples that participate in trading activities and inactive samples that do not participate in trading activities) into the Markov fitting process, the dynamic changes of consumer behavior can be revealed more accurately. In this way, by incorporating the Markov fitting idea of ​​multiple types of samples, not only the accuracy and comprehensiveness of parameter fitting are improved, but also strong support is provided for a deep understanding of consumer behavior and its impact by activities.

[0096] (4) Implement parameter fitting of the transition probability change model for integrated participation activity identification, including:

[0097] Based on the sample constructed in (1), the parameters of the transition probability change model of the integrated intervention measures constructed in (3) are fitted. The parameter estimation model here includes but is not limited to the OLS least squares estimation fitting model. For example: Y it To Y it +1 for the fit,

[0098] [0.2,0.2,0.6]≈W[0.3,0.2,0.5]+1*Ω[0.3,0.2,0.5]+0.25, where W and Ω are 3×3 dimensional matrix parameters.

[0099] (5) Decision analysis based on probability change (substitution effect) mining matrix

[0100] Based on the probability change matrix obtained by fitting in (4)

[0101]

[0102] As an example, suppose we need to recommend the top 2 industries (i.e. the top two industries), add the columns of the matrix and perform the recommendation sorting to get [1.00, 1.39, 1.14]->[1.39, 1.14, 1.00]->[retail, electronic products, catering]->[retail, electronic products], and the retail placement ratio is: Proportion of electronic products

[0103] As described above, according to the transaction data processing method of this embodiment, the consumption association matrix between industries is brought about by the automatic model fitting of the consumption voucher placement, and the priority ranking and placement ratio of the consumption voucher placement are calculated and output based on the association matrix, thereby solving the problem that the related technology uses manual methods to determine the target industry and the allocation of the issuance quota for each industry, resulting in poor results in the issuance of consumption vouchers. Specifically, the transaction data processing method according to this embodiment provides efficient and accurate support for actual consumption voucher placement decisions. Under the same amount of consumption voucher placement, when the transaction data processing method of this embodiment is used for placement, the overall consumption transaction volume is increased by 4.5% compared to the method of evenly spreading the placement.

[0104] Moreover, according to the transaction data processing method of this embodiment, by constructing a probability change matrix Ω between industries based on a Markov process and integrating the characteristics of the change values, the fitting effect of the model can be further improved.

[0105] Furthermore, the transaction data processing method of the present invention is not only applicable to the field of issuance of consumer vouchers, but can also be applied to risk prevention and control, e-commerce and other aspects. It has good application effect and a wide range of applications and has strong universality.

[0106] Figure 2 It is a structural block diagram of the transaction data processing system of the present invention.

[0107] like Figure 2 As shown, the transaction data processing system 100 of the present invention includes:

[0108] A model sample construction module 110 is used to obtain target proportion vectors of targets in different project categories, and to construct a target proportion vector model sample based on the target proportion vectors;

[0109] An association model building module 120, configured to build an association model based on the target proportion vector model sample and the association probability of the target from one project category to another project category;

[0110] A change model building module 130, configured to build a change model based on the target proportion vector model sample, the association probability, and the change association probability of the target changing from one project category to another project category;

[0111] A parameter fitting module 140 is used to perform parameter fitting on the parameters of the change model based on the target proportion vector model sample to obtain the change association probability; and

[0112] The decision calculation module 150 is used to calculate the decisions related to the target of different project categories based on the change association probability.

[0113] Among them, in the change model construction module 130, a change model is constructed based on the target proportion vector model sample, the association probability, the change association probability and the role object identifier used to identify whether to participate in the transaction activity.

[0114] In the correlation model building module 120 , a correlation model is built based on a Markov process.

[0115] In the parameter fitting module 140, any one of the following fitting models is used for parameter fitting:

[0116] Least squares estimation fitting model, maximum likelihood estimation fitting model, moment estimation fitting model, least absolutes estimation fitting model, generalized moment estimation fitting model.

[0117] Among them, when there are P items of different categories, the association probability is represented by a P×P dimensional association matrix, and the variable association probability is represented by a P×P dimensional association matrix, where P is a natural number.

[0118] Among them, in the decision calculation module 150, as the decision related to the target, the sum of each column of the P columns of the change association probability is calculated and sorted to obtain a priority ranking.

[0119] In the decision calculation module 150 , as the decision related to the target, n category items from the P categories are selected, and the proportions of the n category items are calculated respectively.

[0120] As an embodiment of a consumption voucher placement application, in the model sample construction module 110, the consumption proportion vector of consumers in different industry categories is obtained as the target proportion vector and the model sample is constructed according to the following formula (1), where it is assumed that there are P different industries:

[0121] Yit =(y it,1 ,y it,2 ,…,y it,P ) T (1)

[0122] Among them, Y it is the consumption proportion vector, i represents the i-th sample, t represents the t-th moment, t-1 represents the t-1-th moment,

[0123] In the association model building module 120, an association model is built based on the Markov process according to the following formula (2):

[0124] Y it =WY it-1 +∈ it (2)

[0125] Where W = (w p1p2 )∈R P×P is the association probability, w p1p2 represents the transfer probability of consumers’ consumption tendency in industry p2 to industry p1, ∈ it is the noise term,

[0126] In the change model building module 130, a transition probability change model is built according to the following formula (3):

[0127] Y it =WY it-1 +Treat i *ΩY it-1 +∈ it (3)

[0128] Where, Ω=(τ p1p2 )∈R P×P is the probability change matrix, τ p1p2 represents the probability of consumers changing their consumption from industry p2 to industry p1, ∈ it is the noise term, Treat i Indicates the object identifier.

[0129] In the association model building module 120, by training W so that WY it-1 +∈ it With Y it The error is minimal.

[0130] In the decision calculation module 150, the probability change matrix is ​​summed and sorted, and the obtained priority ranking is used as the ranking of the consumption voucher distribution industry, or n industries are selected to calculate the proportion of the n industries and use it as the proportion of the consumption voucher distribution industry.

[0131] The present invention also provides a computer-readable medium on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned transaction data processing method is implemented.

[0132] The present invention also provides a computer device, comprising a storage module, a processor, and a computer program stored in the storage module and executable on the processor, wherein the processor implements the above-mentioned transaction data processing method when executing the computer program.

[0133] The present invention also provides a computer program product, comprising a computer program, which implements the above-mentioned transaction data processing method when executed by a processor.

[0134] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Those skilled in the art can think of other feasible changes or substitutions based on the technical scope disclosed in the present application, and such changes or substitutions are all included in the protection scope of the present application. In the absence of conflict, the implementation modes of the present application and the features in the implementation modes can also be combined with each other. The protection scope of the present application shall be subject to the description of the claims.

Claims

1. A transaction data processing method, characterized in that: include: A model sample construction step, obtaining target proportion vectors of targets in different project categories, and constructing a target proportion vector model sample based on the target proportion vectors; The association model building step is to build an association model based on the target proportion vector model sample and the association probability of the target from one project category to another project category; A change model building step, building a change model based on the target proportion vector model sample, the association probability and the change association probability of the target changing from one project category to another project category; A parameter fitting step, performing parameter fitting on the parameters of the change model based on the target proportion vector model sample to obtain the change association probability; as well as The decision calculation step is to calculate the decisions related to the target for different project categories based on the change association probability.

2. The transaction data processing method according to claim 1, characterized in that: In the change model construction step, a change model is constructed based on the target proportion vector model sample, the association probability, the change association probability, and an effector object identifier for identifying whether to participate in a transaction activity.

3. The transaction data processing method according to claim 2, characterized in that: In the association model building step, the association model is built based on the Markov process. In the step of constructing the change model, the active samples participating in the transaction activities and the inactive samples not participating in the transaction activities are incorporated into the Markov process to construct the change model.

4. The transaction data processing method according to claim 3, characterized in that: In the parameter fitting step, any one of the following fitting models is used for parameter fitting: Least squares estimation fitting model, maximum likelihood estimation fitting model, moment estimation fitting model, least absolutes estimation fitting model, generalized moment estimation fitting model.

5. The transaction data processing method according to claim 4, characterized in that: When there are P items of different categories, the association probability is represented by a P×P dimensional association matrix, and the variable association probability is represented by a P×P dimensional association matrix, where P is a natural number.

6. The transaction data processing method according to claim 5, characterized in that: In the decision calculation step, as the decision related to the target, the sum of each column of the P columns of the change association probability is calculated and sorted to obtain a priority ranking.

7. The transaction data processing method according to claim 6, characterized in that: In the decision calculation step, as the decision related to the target, n category items are selected from P categories, and the proportions of the n category items are calculated respectively.

8. The transaction data processing method according to any one of claims 2 to 7, characterized in that: In the model sample construction step, the consumption proportion vector of consumers in different industry categories is obtained as the target proportion vector and the model sample is constructed according to the following formula (1), where it is assumed that there are P different industries: AND it (and it，1 ,and it，2 ,…,and it，P ) T (1) Among them, Y it is the consumption proportion vector, i represents the i-th sample, t represents the t-th moment, t-1 represents the t-1-th moment, In the association model construction step, the association model is constructed based on the Markov process according to the following formula (2): Y it =WY it-1 +∈ it (2) Where W = (w p1p2 )∈R P×P is the association probability, w p1p2 represents the transfer probability of consumers’ consumption tendency in industry p2 to industry p1, ∈ it is the noise term, In the change model construction step, the transition probability change model is constructed according to the following formula (3): Y it =WY it-1+ Treat i *ΩY it-1 +∈ it (3) Where, Ω=(τ p1p2 )∈R P×P is the probability change matrix, τ p1p2 represents the probability of consumers changing their consumption from industry p2 to industry p1, ∈ it is the noise term, Treat i Indicates the object identifier.

9. The transaction data processing method according to claim 8, characterized in that: In the association model building step, by training W so that WY it-1 +∈ it With Y it The error is minimal.

10. The transaction data processing method according to claim 9, characterized in that: In the decision calculation step, the probability change matrix is ​​column-added and sorted, and the resulting sorting is used as the priority sorting of the consumer voucher distribution industry.

11. The transaction data processing method according to claim 8, characterized in that: In the decision-making calculation step, n industries are selected to calculate the proportion of the n industries and use it as the proportion of the industry for the consumption voucher issuance.

12. A transaction data processing system, characterized in that: include: A model sample construction module, used to obtain target proportion vectors of targets in different project categories, and to construct a target proportion vector model sample based on the target proportion vectors; An association model building module, used to build an association model based on the target proportion vector model sample and the association probability of the target from one project category to another project category; A change model building module, used to build a change model based on the target proportion vector model sample, the association probability and the change association probability of the target changing from one project category to another project category; A parameter fitting module, used for performing parameter fitting on the parameters of the change model based on the target proportion vector model sample to obtain the change association probability; as well as The decision calculation module is used to calculate the decisions related to the target of different project categories based on the change association probability.

13. The transaction data processing system according to claim 12, characterized in that: In the change model construction module, a change model is constructed based on the target proportion vector model sample, the association probability, the change association probability and the role object identifier for identifying whether to participate in the transaction activity. In the step of constructing the change model, the active samples participating in the transaction activities and the inactive samples not participating in the transaction activities are incorporated into the Markov process to construct the change model.

14. The transaction data processing system according to claim 13, characterized in that: In the association model building module, an association model is built based on a Markov process.

15. The transaction data processing system according to claim 14, characterized in that: In the parameter fitting module, any one of the following fitting models is used for parameter fitting: Least squares estimation fitting model, maximum likelihood estimation fitting model, moment estimation fitting model, least absolutes estimation fitting model, generalized moment estimation fitting model.

16. The transaction data processing system according to claim 15, characterized in that: When there are P items of different categories, the association probability is represented by a P×P dimensional association matrix, and the variable association probability is represented by a P×P dimensional association matrix, where P is a natural number.

17. The transaction data processing system according to claim 16, characterized in that: In the decision calculation module, as the decision related to the target, the sum of each column of the P columns of the change association probability is calculated and sorted to obtain a priority ranking.

18. The transaction data processing system according to claim 17, characterized in that: In the decision calculation module, as the decision related to the target, n category items from P categories are selected, and the proportions of the n category items are calculated respectively.

19. The transaction data processing system according to any one of claims 13 to 18, characterized in that: In the model sample construction module, the consumption proportion vector of consumers in different industry categories is obtained as the target proportion vector and the model sample is constructed according to the following formula (1), where it is assumed that there are P different industries: AND it (and it,1 ,and it,2 ,...,and it,P ) T (1) Among them, Y it is the consumption proportion vector, i represents the i-th sample, t represents the t-th moment, t-1 represents the t-1-th moment, In the association model building module, the association model is built based on the Markov process according to the following formula (2): Y it =WY it-1 +∈ it (2) Where W = (w p1p2 )∈R P×P is the association probability, w p1p2 represents the transfer probability of consumers’ consumption tendency in industry p2 to industry p1, ∈ it is the noise term, In the change model construction module, the transition probability change model is constructed according to the following formula (3): Y it =WY it-1 +Treat i *QY it-1 +∈ it (3) Where, Ω=(τ p1p2 )∈R P×P is the probability change matrix, τ p1p2 represents the probability of consumers changing their consumption from industry p2 to industry p1, ∈ it is the noise term, Treat i Indicates the object identifier.

20. The transaction data processing system according to claim 19, characterized in that: In the association model building module, by training W so that WY it-1 +∈ it With Y it The error is minimal.

21. The transaction data processing system according to claim 20, characterized in that: In the decision calculation module, the probability change matrix is ​​column-summed and sorted, and the obtained priority ranking is used as the ranking of the consumer voucher distribution industry.

22. The transaction data processing system according to claim 21, characterized in that: In the decision-making calculation module, n industries are selected to calculate the proportion of the n industries and use it as the proportion of the industry for the consumption voucher issuance.

23. A computer readable medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the transaction data processing method according to any one of claims 1 to 11 is implemented.

24. A computer device comprising a storage module, a processor, and a computer program stored in the storage module and executable on the processor, characterized in that: When the processor executes the computer program, the transaction data processing method according to any one of claims 1 to 11 is implemented.

25. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the transaction data processing method according to any one of claims 1 to 11 is implemented.

Images