CN114065722A - System, method and device for generating transaction report and electronic equipment - Google Patents

Abstract

The embodiment of the specification discloses a system, a method, a device and electronic equipment for generating a transaction report, wherein the method comprises the following steps: extracting a feature set of the target transaction of a target user according to transaction data of the target transaction of the target user, wherein the feature set of the target transaction of the target user at least comprises an identity feature of the target user, a transaction feature of the target transaction and a risk field feature of the target transaction; generating a transaction report for the target transaction of the target user based on the set of features for the target transaction for the target user and a transaction report generation model; the transaction report generation model is obtained by training feature sets of a plurality of transactions operated by a plurality of users.

Description

System, method and device for generating transaction report and electronic equipment

Technical Field

The present disclosure relates to the field of computer software technologies, and in particular, to a system, a method, an apparatus, and an electronic device for generating a transaction report.

Background

At present, the quality requirements of the transaction reports by the regulatory authorities are increasing. The existing transaction report is generated by either manual writing or splicing based on a plurality of established rules. Where manually written transaction reports may be ragged at the quality level due to lack of uniform specifications. The transaction reports generated by splicing based on a plurality of customized rules often cannot cover all suspicious transaction conditions because of relying on the rules formulated by manual experience, and the transaction reports generated by splicing in the mode are scattered and often have logic lower than that of manually written transaction reports.

Disclosure of Invention

The embodiments of the present specification provide a system, a method, a device and an electronic device for generating a transaction report, so as to avoid personalized differences and unsmooth and logical connections of rule splicing in manual report writing, and further improve the quality of writing a transaction report.

In order to solve the above technical problem, the embodiments of the present specification are implemented as follows:

in a first aspect, a method for generating a transaction report is provided, including:

extracting a feature set of a target transaction of a target user according to transaction data of the target transaction of the target user, wherein the feature set of the target transaction of the target user at least comprises an identity feature of the target user, a transaction feature of the target transaction and a risk field feature of the target transaction;

generating a transaction report for the target transaction of the target user based on the set of features for the target transaction for the target user and a transaction report generation model;

wherein the transaction report generation model is trained based on feature sets of a plurality of transactions operated by a plurality of users.

In a second aspect, a system for generating a transaction report is provided, including:

the data storage module is used for acquiring and storing transaction data of target transaction of a target user;

the feature extraction module is used for extracting a feature set related to the target transaction of the target user according to the transaction data of the target transaction of the target user in the data storage module, wherein the feature set related to the target transaction of the target user at least comprises an identity feature of the target user, a transaction feature of the target transaction and a risk field feature of the target transaction;

the message generation module is used for generating a transaction report of the target transaction of the target user based on the feature set of the target transaction of the target user and the transaction report generation model which are extracted by the feature extraction module;

wherein the transaction report generation model is trained based on feature sets of a plurality of transactions operated by a plurality of users.

In a third aspect, an apparatus for generating a transaction report is provided, including:

the feature set extraction unit is used for extracting a feature set related to the target transaction of a target user according to transaction data of the target transaction of the target user, wherein the feature set related to the target transaction of the target user at least comprises an identity feature of the target user, a transaction feature of the target transaction and a risk field feature of the target transaction;

a report generation unit which generates a transaction report of the target transaction of the target user based on the feature set of the target transaction of the target user and a transaction report generation model;

wherein the transaction report generation model is trained based on feature sets of a plurality of transactions operated by a plurality of users.

In a fourth aspect, an electronic device is presented, comprising:

a processor; and

a memory arranged to store computer executable instructions that, when executed, cause the processor to:

extracting a feature set of a target transaction of a target user according to transaction data of the target transaction of the target user, wherein the feature set of the target transaction of the target user at least comprises an identity feature of the target user, a transaction feature of the target transaction and a risk field feature of the target transaction;

generating a transaction report for the target transaction of the target user based on the set of features for the target transaction for the target user and a transaction report generation model;

wherein the transaction report generation model is trained based on feature sets of a plurality of transactions operated by a plurality of users.

In a fifth aspect, a computer-readable storage medium is presented, which stores one or more programs that, when executed by an electronic device comprising a plurality of application programs, cause the electronic device to perform operations comprising:

extracting a feature set of a target transaction of a target user according to transaction data of the target transaction of the target user, wherein the feature set of the target transaction of the target user at least comprises an identity feature of the target user, a transaction feature of the target transaction and a risk field feature of the target transaction;

generating a transaction report for the target transaction of the target user based on the set of features for the target transaction for the target user and a transaction report generation model;

wherein the transaction report generation model is trained based on feature sets of a plurality of transactions operated by a plurality of users.

As can be seen from the technical solutions provided in the embodiments of the present specification, the embodiments of the present specification have at least one of the following technical effects:

one or more embodiments provided in the present specification can extract a feature set related to a target transaction of a target user according to transaction data of the target transaction of the target user, where the feature set related to the target transaction of the target user at least includes an identity feature of the target user, a transaction feature of the target transaction, and a risk field feature of the target transaction; generating a transaction report for the target transaction of the target user based on the set of features regarding the target transaction of the target user and a transaction report generation model; the transaction report generation model is obtained by training a feature set of a plurality of transactions operated by a plurality of users. The transaction report generation model is obtained through machine learning training based on the identity characteristics of a plurality of users, the transaction characteristics of a plurality of transactions operated by the users and the feature set of the risk field characteristics of the transactions. Because the artificial intelligence model can learn the manual writing mode of the transaction report to write the transaction report, the generated transaction report not only avoids the individual difference of different manually written transaction reports, but also avoids the problem that the transaction reports generated by the regular splicing mode are scattered and unsmooth, and simultaneously effectively improves the quality of the generated transaction report.

Drawings

In order to more clearly illustrate the embodiments of the present specification or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings described below are only some embodiments described in the specification, and other drawings can be obtained by those skilled in the art without inventive labor.

Fig. 1 is a schematic implementation flow diagram of a transaction report generation method according to an embodiment of the present disclosure.

Fig. 2 is a schematic flow chart of a method for generating a transaction report, which is provided by an embodiment of the present disclosure and is applied in an actual scenario.

Fig. 3 is a schematic structural diagram of a transaction report generation system according to an embodiment of the present specification.

Fig. 4 is a schematic structural diagram of a transaction report generation device according to an embodiment of the present specification.

Fig. 5 is a schematic structural diagram of an electronic device provided in an embodiment of the present specification.

Detailed Description

In order to make the objects, technical solutions and advantages of the present specification clearer, the technical solutions in the present specification will be clearly and completely described below with reference to specific embodiments of the present specification and corresponding drawings. It is to be understood that the embodiments described are only a few embodiments of this document, and not all embodiments. All other embodiments obtained by a person skilled in the art without making creative efforts based on the embodiments in this document belong to the protection scope of this document.

The technical solutions provided by the embodiments of the present description are described in detail below with reference to the accompanying drawings.

Faced with increasingly severe forms of supervision, there is often a need to provide higher quality transaction reports of suspicious risk transactions to regulatory authorities. However, the existing transaction report for generating suspicious risk transactions is either directly written by manual work, the method is laggard, the efficiency is low, the quality written by different people is different, some personalized differences exist, or the transaction report is matched and spliced with a plurality of rules to generate a transaction report.

In view of the above, one or more embodiments of the present specification provide a method for generating a transaction report, which can extract a feature set related to a target transaction of a target user according to transaction data of the target transaction of the target user, where the feature set related to the target transaction of the target user at least includes an identity feature of the target user, a transaction feature of the target transaction, and a risk domain feature of the target transaction; generating a transaction report for the target transaction of the target user based on the set of features regarding the target transaction of the target user and a transaction report generation model; the transaction report generation model is obtained by training feature sets of a plurality of transactions operated by a plurality of users.

The method comprises the steps of generating a model through a transaction report obtained through machine learning training based on identity characteristics of a plurality of users, transaction characteristics of a plurality of transactions operated by the plurality of users and a characteristic set of risk field characteristics of the plurality of transactions. Because the artificial intelligence model can learn the manual writing mode of the transaction report to write the transaction report, the generated transaction report not only avoids the individual difference of different manual writing transaction reports, but also avoids the problem that the transaction reports generated by the regular splicing mode are scattered and unsmooth, and simultaneously effectively improves the quality of the generated transaction report.

It should be understood that the executing body of the method for generating a transaction report provided in the embodiments of the present specification may be, but is not limited to, a server, a computer, etc. that can be configured to execute at least one of the user terminals of the method provided in the embodiments of the present specification, or the executing body of the method may also be the client itself that can execute the method.

For convenience of description, the following description will be made of an embodiment of the method, taking an execution subject of the method as a server capable of executing the method as an example. It is understood that the implementation of the method by the server is merely an exemplary illustration and should not be construed as a limitation of the method.

Fig. 1 is a schematic implementation flow diagram of a transaction report generation method according to an embodiment of the present disclosure. The method of fig. 1 may include:

s110, extracting a feature set related to the target transaction of the target user according to the transaction data of the target transaction of the target user.

It should be understood that the core of the generated transaction report is often to summarize the various features involved in the suspected risk transaction, and therefore, accurately summarizing and generalizing the features of the target transaction for the target user is key to generating a high quality transaction report. Generally, the feature set of the target transaction related to the target user at least includes the identity feature of the target user, the transaction feature of the target transaction, and the risk domain feature of the target transaction.

The identity of the target user may include basic information of the target user, such as the name, age, sex, work, industry, residence address, and the like of the target user. It will be appreciated that in the area of backwash money, the basic information that identifies the target user is the core of due diligence, which is the first step forward to determine if the target user is at risk.

The transaction characteristics of the target transaction may include the amount of money flowing in the target transaction, the amount of money flowing out of the target transaction, the total amount of transactions of the target user in the last month when operating the target transaction, the total number of transactions of the target user in the last month when operating the target transaction, and the like, which reflect the financial behavior of the target user.

The risk domain characteristics of the transaction may include risk domain characteristics of gambling transactions, risk domain characteristics of cash-out transactions, and the like. The different risk transactions often have different manifestations, such as gambling transactions, which have obvious risk characteristics that the early morning transactions are more and the whole hundred transactions are more; for cash register transactions, the user often purchases goods from the merchant and receives a merchant refund slightly less than the loan amount. The embodiments of the present disclosure may summarize corresponding risk domain features in advance based on different risk transactions, so as to describe different risk transactions. When the target transaction is a gambling transaction, the risk area characteristics may be summarized including more transactions in the morning and more transactions in the whole hundred. When the target transaction is a cash-out transaction, its risk area characteristics may be summarized including loan payments, merchant cash-back, and cash-back amounts slightly less than the payment amount.

Optionally, the feature set may also include features of other dimensions in order to improve the quality of the generated transaction report. Specifically, the set of features for a target transaction with a target user may further include:

the target user operates the timing characteristics of the target transaction;

a target user operating a graph feature of a target transaction;

the target user manipulates the relational features of the target transaction.

The time sequence characteristics of the target user operating the target transaction can be compiled from behavior operation records within about 24 hours (or within about one week and other time periods) when the target user operates the target transaction. The graph characteristics of the target user operating the target transaction may include characteristics in a logic graph when the target user operates the target transaction, and the logic graph may be formed by association relations between set rules that are met when the target user operates the target transaction. The relationship characteristic of the target user operating the target transaction may include a characteristic of an associative relationship between other transactions (including dimensions of time, place, ID, etc.) operated by other users generated when the target user operates the target transaction.

And S120, generating a transaction report of the target transaction of the target user based on the feature set of the target transaction of the target user and the transaction report generation model.

The transaction report generation model is obtained by training feature sets of a plurality of transactions operated by a plurality of users.

In order to solve the problems of insufficient fluency and logic incapability of messages generated by the conventional rule splicing, the embodiment of the specification adopts a machine learning mode, namely, a machine learns the transaction report written manually, learns the logic and method behind the transaction report written manually, namely, trains to obtain a transaction report generation model. When the feature set of the target transaction of the target user is input into the transaction report generation model, the transaction report generation model can generate an auxiliary message according to each specific feature in the feature set of the target transaction of the target user by simulating human thinking, and compared with a transaction report generated by simple rule splicing, the generated transaction report is smoother and more logical, and meanwhile, different types of transaction reports can be generated aiming at different risk transactions.

Alternatively, a expert dialog library of transaction reports may incorporate a retrieval mechanism into the model input to address the problem of insufficient long text information of the generated transaction reports. Specifically, the transaction report generation model is trained on a feature set of a plurality of transactions operated by a plurality of users and a expert talk skill base of transaction reports. The transaction report generation model obtained by training the expert dialog library combined with the transaction report can be called KAIN (fully known as Knowledge Aware Interaction Network) algorithm, that is, deep learning and expert Knowledge are combined to generate the transaction report of the user.

Specifically, when the expert speech library of the transaction report can be introduced into the model input by a retrieval mechanism, when a feature set of one transaction operated by a user is received each time, the feature set can be used as a keyword to retrieve the first k expert speech related to the keyword in the expert speech library of the transaction report, and then an artificial four-dimensional learning process is simulated to select the best matching expert speech from the first k expert speech, so that the generation model of the transaction report is trained and obtained.

Optionally, the trade report generated during the training of the trade report generation model may be modified based on the expert dialog library of the trade report to modify the less accurate conclusions in the trade report generated during the training. The transaction report generation model is trained based on a feature set of a plurality of transactions operated by a plurality of users and a loss function constructed by a expert conversational library of transaction reports.

The loss function constructed by the expert speech library of the transaction report can be composed of the loss function of the generation model of the transaction report and the message probability corresponding to each expert speech in the expert speech library of the transaction report.

Optionally, a logic map can be constructed for the transaction operated by the user, and complex logic is added to the model training process in a meta path (meta path) form, so that the problem that the learning of the transaction report generation model is insufficient in complex logic is solved. Specifically, the transaction report generation model is obtained by training a logic map of a plurality of transactions based on a plurality of user operations;

the logic map of the transaction operated by one user is constructed by a feature set of the transaction operated by the user and a plurality of set rules;

the logic map of the transaction operated by a user is used for indicating the incidence relation between at least two rules matched in the set rules when the user operates the transaction.

For example, if a user-operated transaction satisfies both rule 1 and rule 2, and some parameters between rule 1 and rule 2 have an association relationship, a logical map of rule 1 and rule 2 satisfied by the user-operated transaction may be constructed.

Then, generating a transaction report for the target transaction of the target user based on the set of features regarding the target transaction of the target user and the transaction report generation model, comprising:

determining a rule matched by the target user when the target user operates the target transaction from the set plurality of rules based on the feature set of the target transaction related to the target user;

determining a logic map of the target transaction operated by the target user based on a rule matched by the target user when the target transaction is operated;

and generating a transaction report of the target transaction of the target user based on the logic map of the target transaction operated by the target user and the transaction report generation model.

Optionally, in order to continuously optimize the transaction report generation model and effectively improve the quality of the transaction report generated by the transaction report generation model, the method provided in the embodiment of the present specification can generate the transaction report based on the transaction report generation model, and can also receive an audit result of a service supervisor on the generated transaction report, and dynamically optimize the transaction report generation model according to the transaction report which does not meet the quality and the newly issued supervision requirement. Specifically, the method provided by the embodiment of the present specification further includes:

acquiring an auditing result of a transaction report of a target transaction of a target user by a service supervisor;

optimizing a transaction report of the target transaction of the target user based on the auditing result of the service supervisor;

and adding the optimized transaction report into a training sample set of a transaction report generation model, and optimizing the transaction report generation model based on the training sample set of the transaction report generation model.

The auditing result of the transaction report of the target transaction of the target user by the service supervisor can be realized based on the following modes: algorithm correction, message review and supervision evaluation.

Wherein the algorithm is modified to detect the quality of the generated transaction report from multiple aspects, such as wrong words, grammar, compliance, logic, etc., by using the algorithm, and simultaneously modify the corresponding errors in the transaction report. The message review is to input the Transaction report after the algorithm correction into a Transaction supervision System (TMS) and manually review the quality of the Transaction report by a professional. And the supervision evaluation is to report the transaction report audited by the TMS system to a service supervisor, the service supervisor evaluates the transaction report in a sampling manner, and feeds back some new supervision requirements and transaction reports which do not meet the requirements to the transaction report generation model.

Fig. 2 is a schematic flow chart of a method for generating a transaction report, which is provided by an embodiment of the present disclosure and is applied in an actual scenario. In fig. 2, the implementation of the generation method of the transaction report first depends on the underlying support, including the mass storage database, the distributed computing platform and the algorithm platform. The distributed computing platform is used for extracting a corresponding characteristic set based on the massive transaction data stored in the massive storage database, and the algorithm platform is used for supporting training and prediction of a transaction report generation model. And then, extracting a characteristic set from the massive transaction data stored in the massive storage database through a distributed computing platform to obtain characteristics such as identity characteristics, transaction characteristics, risk field characteristics and the like. And inputting the extracted feature set into a generation model of the transaction report to generate the transaction report. And finally, auditing the transaction report, namely, sequentially carrying out algorithm correction, message reexamination and supervision evaluation on the transaction report generated by the transaction report generation model, and feeding the auditing result of the transaction report back to the transaction report generation model so as to optimize the transaction report generation model.

One or more embodiments provided in the present specification can extract a feature set related to a target transaction of a target user according to transaction data of the target transaction of the target user, where the feature set related to the target transaction of the target user at least includes an identity feature of the target user, a transaction feature of the target transaction, and a risk field feature of the target transaction; generating a transaction report for the target transaction of the target user based on the set of features regarding the target transaction of the target user and a transaction report generation model; the transaction report generation model is obtained by training a feature set of a plurality of transactions operated by a plurality of users. And generating a model of the transaction report obtained by machine learning training based on the identity characteristics of a plurality of users, the transaction characteristics of a plurality of transactions operated by the plurality of users and the characteristic set of the risk field characteristics of the plurality of transactions. Because the artificial intelligence model can learn the manual writing mode of the transaction report to write the transaction report, the generated transaction report not only avoids the individual difference of different manually written transaction reports, but also avoids the problem that the transaction reports generated by the regular splicing mode are scattered and unsmooth, and simultaneously effectively improves the quality of the generated transaction report.

Fig. 3 is a schematic structural diagram of a system 300 for generating a transaction report according to an embodiment of the present disclosure. Referring to FIG. 3, in one software implementation, a system 300 for generating a transaction report may include:

the data storage module 301 is used for acquiring and storing transaction data of target transactions of target users;

a feature extraction module 302, configured to extract a feature set related to the target transaction of the target user according to the transaction data of the target transaction of the target user in the data storage module, where the feature set related to the target transaction of the target user at least includes an identity feature of the target user, a transaction feature of the target transaction, and a risk domain feature of the target transaction;

the message generation module 303 is configured to generate a transaction report of the target transaction of the target user based on the feature set about the target transaction of the target user and the transaction report generation model extracted by the feature extraction module;

wherein the transaction report generation model is trained based on feature sets of a plurality of transactions operated by a plurality of users.

Optionally, in an embodiment, the system further includes a model training module configured to:

acquiring feature sets of a plurality of transactions operated by a plurality of users;

and training to obtain the transaction report generation model based on the feature sets of a plurality of transactions operated by the plurality of users.

Optionally, in an embodiment, the model training module is configured to:

acquiring a specialist talk skill library of a transaction report;

and training a specialist talk library of the transaction report based on the feature sets of the transactions operated by the users and the transaction report to obtain the transaction report generation model.

Optionally, in an embodiment, the model training module is configured to:

constructing a loss function based on the expert conversation library of the transaction report, wherein the loss function is used for correcting the transaction report generated by the transaction report generation model;

training to obtain the transaction report generation model based on the feature sets of the transactions operated by the users and the loss function constructed by the expert conversational library of the transaction report.

Optionally, in an embodiment, the model training module is configured to:

constructing a logic map of a plurality of transactions operated by the plurality of users based on the feature sets of the transactions operated by the plurality of users and a plurality of set rules; the logic diagram spectrum of the transaction operated by one user is used for indicating the incidence relation between at least two rules matched in the set rules when the user operates the transaction;

and training to obtain the transaction report generation model based on the logic maps of the transactions operated by the users.

Optionally, in an implementation manner, the message generating module 303 is configured to:

determining, from the set plurality of rules, a rule that the target user matches when operating the target transaction based on a set of features for the target transaction for the target user;

determining a logical graph of the target transaction operated by the target user based on rules matched by the target user when operating the target transaction;

generating a transaction report for the target transaction of the target user based on the logical graph of the target transaction operated by the target user and the transaction report generation model.

Optionally, in an implementation manner, the message generating module 303 further includes:

the manual writing unit is used for acquiring a feature set of the target transaction of the target user from the feature extraction module so as to enable a business person to write a transaction report of the target transaction of the target user based on the feature set of the target transaction of the target user;

and the rule splicing unit is used for matching the feature set about the target transaction of the target user, which is extracted by the feature extraction module, with a plurality of preset message generation rules to obtain a plurality of transaction messages, and splicing the transaction messages to obtain a transaction report of the target transaction of the target user.

Optionally, in an embodiment, the system further includes an optimization module configured to:

acquiring an auditing result of a transaction report of a target transaction of the target user by a service supervisor;

optimizing the transaction report of the target transaction of the target user based on the auditing result of the service supervisor;

adding the optimized transaction report into a training sample set of the transaction report generation model, and optimizing the transaction report generation model based on the training sample set of the transaction report generation model.

Optionally, in one embodiment, the set of features for the target transaction with respect to the target user further comprises:

the target user operates the timing characteristics of the target transaction;

the target user operating a graph feature of the target transaction;

the target user operates a relationship feature of the target transaction.

The system 300 for generating a transaction report can implement the method of the embodiment of the method shown in fig. 1 to fig. 2, and specifically refer to the method for generating a transaction report shown in the embodiment shown in fig. 1 to fig. 3, which is not described again.

Fig. 4 is a schematic structural diagram of a transaction report generation apparatus 400 according to an embodiment of the present disclosure. Referring to fig. 4, in one software implementation, the transaction report generation apparatus 400 may include:

the feature set extracting unit 401 is configured to extract a feature set of a target transaction of a target user according to transaction data of the target transaction of the target user, where the feature set of the target transaction of the target user at least includes an identity feature of the target user, a transaction feature of the target transaction, and a risk domain feature of the target transaction;

a report generation unit 402 for generating a transaction report of the target transaction of the target user based on the feature set of the target transaction and a transaction report generation model for the target user;

wherein the transaction report generation model is trained based on feature sets of a plurality of transactions operated by a plurality of users.

Optionally, in one embodiment, the transaction report generation model is trained for a expert conversational library of transaction reports and feature sets of transactions based on the plurality of user actions.

Optionally, in one embodiment, the transaction report generation model is trained based on a feature set of a plurality of transactions operated by the plurality of users and a loss function constructed from a expert conversational library of transaction reports.

Optionally, in an embodiment, the transaction report generation model is trained for a logical graph of a plurality of transactions based on the plurality of user operations;

the logic map of the transaction operated by one user is constructed by a feature set of the transaction operated by the user and a plurality of set rules;

the logic map of the transaction operated by the user is used for indicating the incidence relation between at least two rules matched in the set multiple rules when the user operates the transaction.

Optionally, in an embodiment, the report generating unit 402 is configured to:

determining, from the set plurality of rules, a rule that the target user matches when operating the target transaction based on a set of features for the target transaction for the target user;

determining a logical graph of the target transaction operated by the target user based on rules matched by the target user when operating the target transaction;

generating a transaction report for the target transaction of the target user based on the logical graph of the target transaction operated by the target user and the transaction report generation model.

Optionally, in an embodiment, the apparatus further includes:

the acquisition unit is used for acquiring the auditing result of the business supervisor on the transaction report of the target transaction of the target user;

the report optimization unit is used for optimizing the transaction report of the target transaction of the target user based on the auditing result of the service supervisor;

and the model optimization unit is used for adding the optimized transaction report into a training sample set of the transaction report generation model and optimizing the transaction report generation model based on the training sample set of the transaction report generation model.

Optionally, in one embodiment, the set of features for the target transaction with respect to the target user further comprises:

the target user operates the timing characteristics of the target transaction;

the target user operating a graph feature of the target transaction;

the target user operates a relationship feature of the target transaction.

The transaction report generation apparatus 400 can implement the method of the embodiment of the method shown in fig. 1 to fig. 2, and specifically refer to the method of generating the transaction report shown in the embodiment shown in fig. 1 to fig. 2, which is not described again.

Fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present specification. Referring to fig. 5, at the hardware level, the electronic device includes a processor, and optionally an internal bus, a network interface, and a memory. The Memory may include a Memory, such as a Random-Access Memory (RAM), and may further include a non-volatile Memory, such as at least 1 disk Memory. Of course, the electronic device may also include hardware required for other services.

The processor, the network interface, and the memory may be connected to each other via an internal bus, which may be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, an EISA (Extended Industry Standard Architecture) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one double-headed arrow is shown in FIG. 5, but this does not indicate only one bus or one type of bus.

And the memory is used for storing programs. In particular, the program may include program code comprising computer operating instructions. The memory may include both memory and non-volatile storage and provides instructions and data to the processor.

The processor reads the corresponding computer program from the nonvolatile memory into the memory and then runs the computer program to form a generating device of the transaction report on a logic level. The processor is used for executing the program stored in the memory and is specifically used for executing the following operations:

extracting a feature set of a target transaction of a target user according to transaction data of the target transaction of the target user, wherein the feature set of the target transaction of the target user at least comprises an identity feature of the target user, a transaction feature of the target transaction and a risk field feature of the target transaction;

generating a transaction report for the target transaction of the target user based on the set of features for the target transaction for the target user and a transaction report generation model;

wherein the transaction report generation model is trained based on feature sets of a plurality of transactions operated by a plurality of users.

The method performed by the transaction report generation device disclosed in the embodiments of fig. 1 to 2 in the present specification can be applied to or implemented by a processor. The processor may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be performed by integrated logic circuits of hardware in a processor or instructions in the form of software. The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The various methods, steps and logic blocks disclosed in the embodiments of this specification may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present specification may be directly implemented by a hardware decoding processor, or may be implemented by a combination of hardware and software modules in the decoding processor. The software module may be located in ram, flash memory, rom, prom, or eprom, registers, etc. storage media as is well known in the art. The storage medium is located in a memory, and a processor reads information in the memory and completes the steps of the method in combination with hardware of the processor.

The electronic device may also execute the method shown in fig. 1, and implement the functions of the transaction report generation apparatus in the embodiment shown in fig. 1, which are not described herein again in this specification.

Embodiments of the present specification also provide a computer-readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by a portable electronic device comprising a plurality of application programs, enable the portable electronic device to perform the method of the embodiment shown in fig. 1, and in particular to perform the following operations:

extracting a feature set of a target transaction of a target user according to transaction data of the target transaction of the target user, wherein the feature set of the target transaction of the target user at least comprises an identity feature of the target user, a transaction feature of the target transaction and a risk field feature of the target transaction;

generating a transaction report for the target transaction of the target user based on the set of features for the target transaction for the target user and a transaction report generation model;

wherein the transaction report generation model is trained based on feature sets of a plurality of transactions operated by a plurality of users.

Of course, besides the software implementation, the electronic device in the present specification does not exclude other implementation manners, such as a logic device or a combination of software and hardware, and the like, that is, the execution subject of the following processing flow is not limited to each logic unit, and may also be hardware or a logic device.

The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

In short, the above description is only a preferred embodiment of the present disclosure, and is not intended to limit the scope of the present disclosure. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present specification shall be included in the protection scope of the present specification.

The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement the information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Disks (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

The embodiments in the present specification are described in a progressive manner, and portions that are similar to each other in the embodiments are referred to each other, and each embodiment focuses on differences from other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.

Claims (11)

1. A method of generating a transaction report, comprising:

extracting a feature set of the target transaction of a target user according to transaction data of the target transaction of the target user, wherein the feature set of the target transaction of the target user at least comprises an identity feature of the target user, a transaction feature of the target transaction and a risk field feature of the target transaction;

generating a transaction report for the target transaction of the target user based on the set of features for the target transaction for the target user and a transaction report generation model;

the transaction report generation model is obtained by training feature sets of a plurality of transactions operated by a plurality of users.

2. The method of claim 1, wherein the transaction report generation model is trained for a expert conversational library of transaction reports and a set of features for a plurality of transactions based on the plurality of user actions.

3. The method of claim 2, wherein the transaction report generation model is trained based on a set of features for a plurality of transactions operated by the plurality of users and a loss function constructed from a expert conversational library of transaction reports.

4. The method of claim 1, the transaction report generation model trained for a logical graph of a plurality of transactions based on the plurality of user operations;

the logic map of the transaction operated by a user is constructed by a set of characteristics of the transaction operated by the user and a plurality of set rules;

the logic map of the transaction operated by a user is used for indicating the association relationship between at least two rules matched in the set multiple rules when the user operates the transaction.

5. The method of claim 4, generating a transaction report for the target transaction of the target user based on the set of features for the target transaction for the target user and a transaction report generation model, comprising:

determining, from the set plurality of rules, a rule that the target user matches when operating the target transaction based on a set of features for the target transaction for the target user;

determining a logic map of the target transaction operated by the target user based on rules matched by the target user when operating the target transaction;

generating a transaction report for the target transaction of the target user based on the logical graph of the target transaction operated by the target user and the transaction report generation model.

6. The method of claim 1, further comprising:

acquiring an auditing result of a transaction report of a target transaction of the target user by a service supervisor;

optimizing a transaction report of the target transaction of the target user based on the auditing result of the service supervisor;

adding the optimized transaction report into a training sample set of the transaction report generation model, and optimizing the transaction report generation model based on the training sample set of the transaction report generation model.

7. The method of claim 1, the set of features for the target transaction with the target user further comprising:

the target user operates the timing characteristics of the target transaction;

the target user operating a graph feature of the target transaction;

the target user operates a relationship feature of the target transaction.

8. A system for generating a transaction report, comprising:

the data storage module is used for acquiring and storing transaction data of target transaction of a target user;

the feature extraction module is used for extracting a feature set related to the target transaction of the target user according to the transaction data of the target transaction of the target user in the data storage module, wherein the feature set related to the target transaction of the target user at least comprises an identity feature of the target user, a transaction feature of the target transaction and a risk field feature of the target transaction;

the message generation module is used for generating a transaction report of the target transaction of the target user based on the feature set of the target transaction of the target user and the transaction report generation model which are extracted by the feature extraction module;

the transaction report generation model is obtained by training feature sets of a plurality of transactions operated by a plurality of users.

9. An apparatus for generating a transaction report, comprising:

the feature set extraction unit is used for extracting a feature set of the target transaction of a target user according to transaction data of the target transaction of the target user, wherein the feature set of the target transaction of the target user at least comprises an identity feature of the target user, a transaction feature of the target transaction and a risk field feature of the target transaction;

a report generation unit which generates a transaction report of the target transaction of the target user based on the feature set of the target transaction of the target user and a transaction report generation model;

the transaction report generation model is obtained by training feature sets of a plurality of transactions operated by a plurality of users.

10. An electronic device, comprising:

a processor; and

a memory arranged to store computer executable instructions that, when executed, cause the processor to:

extracting a feature set of the target transaction of a target user according to transaction data of the target transaction of the target user, wherein the feature set of the target transaction of the target user at least comprises an identity feature of the target user, a transaction feature of the target transaction and a risk field feature of the target transaction;

generating a transaction report for the target transaction of the target user based on the set of features for the target transaction for the target user and a transaction report generation model;

the transaction report generation model is obtained by training feature sets of a plurality of transactions operated by a plurality of users.

11. A computer-readable storage medium storing one or more programs that, when executed by an electronic device including a plurality of application programs, cause the electronic device to:

extracting a feature set of the target transaction of a target user according to transaction data of the target transaction of the target user, wherein the feature set of the target transaction of the target user at least comprises an identity feature of the target user, a transaction feature of the target transaction and a risk field feature of the target transaction;

generating a transaction report for the target transaction of the target user based on the set of features for the target transaction for the target user and a transaction report generation model;

the transaction report generation model is obtained by training feature sets of a plurality of transactions operated by a plurality of users.

Images