CN106875156B

CN106875156B - Universal intelligent auditing platform and auditing method thereof

Info

Publication number: CN106875156B
Application number: CN201710078245.XA
Authority: CN
Inventors: 李晓瑜; 王航; 桑楠; 郑德生; 胡晗
Original assignee: University of Electronic Science and Technology of China
Current assignee: University of Electronic Science and Technology of China
Priority date: 2017-02-14
Filing date: 2017-02-14
Publication date: 2020-09-08
Anticipated expiration: 2037-02-14
Also published as: CN106875156A

Abstract

The invention discloses a generalized intelligent auditing platform and an auditing method thereof, wherein the auditing platform comprises a project construction module and an intelligent auditing module; the project building module is used for managing a project library; the project library comprises at least one instantiation project; the project building module comprises a service data unit, a service database unit, a service JavaBean unit and a service rule unit. The invention carries out irrelevant processing on the service data and the service auditing logic when the auditing platform is developed and deployed, improves the availability and stability of the system, and reduces the development, operation and maintenance cost of the auditing platform; furthermore, the invention supports different auditing application scenes, the auditing platform supports two auditing modes of in-service auditing and after-service auditing, and the corresponding processing modes of the service auditing modes are different according to different auditing requirements of real-time performance, response speed and data volume.

Description

Universal intelligent auditing platform and auditing method thereof

Technical Field

The invention relates to the technical field of intelligent auditing of business data, and also relates to the field of data auditing such as medical insurance reimbursement auditing, enterprise reimbursement auditing and administrative auditing, in particular to a universal intelligent auditing platform and an auditing method thereof.

Background

Under the environment that national economy downlink pressure is continuously increased, market competition is more intense, and enterprise operation risks are continuously increased, enterprises and organizations win the market competition of the enterprises and the organizations with the advantages and the disadvantages in order to improve the market competitiveness of the enterprises and the organizations, and the common method is to open sources and throttle, reduce unnecessary daily operation expenses, improve the resource management level and improve the output rate of input resources. Under the promotion of market demand, the auditing of business data enters a rapid growth period. However, the existing business data auditing mode has the following problems:

in the face of a large number of business data auditing tasks, the traditional manual auditing mode is long in auditing time, low in auditing efficiency and low in abnormal data discrimination capability, and meanwhile, the workload of auditors is too large, so that physical and psychological health is affected. With the coming of the data era, the auditing requirement of business data is higher and higher, and the problem of the traditional manual auditing mode is highlighted.

At present, some service data auditing platforms exist in the market, for example, the invention patent with the application number of CN201510464287.8 and the name of 'an intelligent auditing system and auditing algorithm for medical insurance of major illness', the invention provides an intelligent auditing system for medical insurance of major illness, which comprises a database module, an intelligent auditing engine and a basic information library management module, wherein the database module is used for acquiring medical insurance data of major illness; the intelligent auditing engine is used for auditing the correlation matching degree of each attribute item of the medical insurance data of the major illness; the basic information library management module stores various attribute information of basic data, wherein the intelligent auditing engine is respectively connected with the database module and the basic information library management module and gives the data of the database module and the basic information library management module to complete the auditing treatment. However, the auditing platform has a single service auditing range and limited auditing fields, can only process specific auditing fields, and cannot expand the auditing range and the auditing fields. Although the current business data auditing requirements can be met, the market is dynamic and constantly changes, the actual business requirements are constantly changed, and the data auditing platforms with single auditing range and limited auditing fields are finally eliminated in the harsh market competition.

Meanwhile, in the prior art, actually processed service data are concerned when the audit platform is developed, specifically: most of the existing auditing platforms already set a service data structure to be processed by default during development, and can only analyze the service data corresponding to the service data structure set by default after deployment, so that the limitation of the auditing platform in processing the service data range is highlighted, and the auditing platform has defects. Since the actual business data structure changes dynamically with the changing market, the business data structure of the auditing platform needs to be changed accordingly, but the auditing platform solidifies the business data structure provided by the demander in the code when being developed. When the platform is deployed, the service data structure needs to be modified, the source code of the auditing platform needs to be modified, and the maintenance cost and risk of the auditing platform are increased. After the auditing platform is deployed and released, the risk of modifying the source code is extremely high, and other functional modules are possibly influenced in the code modifying process, so that hidden bugs are left. To circumvent these risks, considerable effort is required by the testers to test the stability of the audit platform, which undoubtedly increases the development, operation and maintenance costs of the audit platform.

In addition, most auditing platforms in the market need to determine specific business auditing logic during development and be solidified in codes during development. Therefore, the expansibility and flexibility of the system are greatly reduced, and the actual service auditing logic is not invariable but changes along with the change of market demands. The design mode of solidifying the business auditing logic into the system in advance is finally to expose huge defects in the market with high or low quality, thereby being slowly eliminated by the harsh market. If the source code is modified to make up for this deficiency, not only are development, testing, and maintenance costs increased, but the availability and stability of the system is reduced.

Finally, most of the auditing platforms on the market only support a single auditing mode, and there should be different auditing application scenarios, and since no selective auditing mode is considered during design and development, a single auditing mode is adopted for all auditing requests, which is obviously unreasonable.

Disclosure of Invention

The invention aims to overcome the defects of the prior art and provides a universal intelligent auditing platform and an auditing method thereof, which are used for carrying out irrelevant processing on service data and service auditing logic during development and deployment of the auditing platform, improving the availability and stability of a system and reducing the development, operation and maintenance costs of the auditing platform.

The purpose of the invention is realized by the following technical scheme: a universal intelligent auditing platform comprises a project construction module and an intelligent auditing module;

the project building module is used for managing a project library; the project library comprises at least one instantiation project; the project construction module comprises a service data unit, a service database unit, a service JavaBean unit and a service rule unit; the service data unit is used for constructing a service data structure of an instantiated project, the service data structure consists of data table information and field information, and provides support for analyzing and verifying a JSON data packet uploaded by a requester and simultaneously provides support for a service database unit, a service JavaBean unit and a service rule unit; the business database unit is used for constructing a business database based on the business data structure of the instantiated project and providing physical storage support for persistent audit data; the service JavaBean unit is used for constructing a service JavaBean set based on the service data structure of the instantiated project, and when an audit request of the intelligent audit module arrives, the audit data is converted into an instantiated service JavaBean object; the business rule unit is used for constructing a business rule base based on the business data structure of the instantiated project, and the business rule base comprises rules and rule flows which are supported by a Drools rule engine and are converted from business auditing logic of the instantiated project;

the intelligent auditing module is used for realizing the auditing of the service data of the auditing platform.

The intelligent auditing module comprises an in-affair auditing unit and a post-affair auditing unit; the in-process auditing unit is used for auditing an application scene with high real-time performance, high response speed and small data volume, and the after-process auditing unit is used for auditing an application scene with low real-time performance, high response speed and large data volume;

the in-service auditing unit comprises a first auditing data interface subunit, a first task management subunit and an in-service auditing engine, wherein the first auditing data interface subunit is responsible for managing data communication between the auditing platform and the auditing requester, and comprises auditing request monitoring, auditing request analysis and auditing result return; the first task management subunit is responsible for managing the audit tasks, and the audit tasks comprise maintaining a task table, creating the audit tasks, changing task states and finishing the audit tasks, wherein all audit task information is recorded in the task table; the affair auditing engine is responsible for auditing business data audited in the affair;

the post-audit unit comprises a second audit data interface subunit, a second task management subunit and a post-audit engine, wherein the second audit data interface subunit is responsible for managing data communication between the audit platform and the audit requester, and comprises monitoring an audit request, analyzing the audit request and returning an audit result; the second task management subunit is responsible for managing the audit tasks, and the audit tasks comprise maintaining a task table, creating the audit tasks, changing the task state and finishing the audit tasks, wherein all audit task information is recorded in the task table; the post audit engine is responsible for processing the post audit business data audit.

The in-service auditing engine is an auditing engine of a linear execution non-pipeline mode; the business audit engine comprises a business data audit component, an audit result assembly component and an audit task persistence component, wherein the business data audit component is used for auditing business data, the audit result assembly component is used for assembling audit results according to a return result standard format, and the audit task persistence component is used for persisting audit task information, business data and audit results to a physical database.

The post-event auditing engine is an auditing engine of a parallel execution pipeline mode; the in-service audit engine comprises a business data audit component, an audit result assembly component, a result return component and an audit task persistence component, wherein the business data audit component is used for auditing business data, the audit result assembly component is used for assembling audit results according to a returned result standard format, the result return component returns the assembled audit results to an audit requester based on a second audit data interface subunit, and the audit task persistence component is used for persisting audit task information, business data and audit results to a physical database.

A generalized intelligent auditing method comprises a project construction step and an intelligent auditing step; the project building step comprises adding instantiated projects, and the adding of the instantiated projects comprises the following sub-steps:

s11: adding instantiation items, and adding basic information of the instantiation items;

s12: adding at least one service data structure, including the addition of data table information and field information; after the addition is completed, the system adds a task number field and a task data index field to the service data structure;

s13: after step S12 is completed, respectively constructing a service database, a service JavaBean set, and a service rule base; the construction of the business database comprises the following sub-steps:

s1311: constructing a data table SQL statement according to the data table number of the service data structure and the field information bound with the data table number;

s1312: executing the structured SQL statement in the instantiated project business database, completing the creation of a business data table, updating a generation mark of the business data structure data table information, and indicating that the business data table is generated;

s1313: generating corresponding service data tables for all service data structures of the instantiated project one by one;

the construction of the service JavaBean set comprises the following sub-steps:

s1321: constructing a java file corresponding to the service JavaBean according to the data table number of the service data structure and the field information bound with the data table number;

s1322: compiling the java file to generate a class file by using a java language programming compiler java, and updating a generation mark of the service data structure data table information to indicate that a service JavaBean is generated;

s1323: generating corresponding service JavaBean by all service data structures of the instantiated project one by one, and constructing a service JavaBean set;

the construction of the business rule base comprises the following sub-steps:

s1331: constructing a rule part, wherein the rule part comprises an attribute part, a condition part and a structure part; the attribute part is used for defining some attributes of the current rule execution, including whether the current rule execution can be repeatedly executed or not, expiration time and effective time; the condition part is used for defining single/multiple conditions of the current rule; the structure part is used for defining the operation executed after the current condition part is met;

s1332: constructing a rule flow, wherein the rule flow is realized based on a business flow chart, and the business flow chart describes a series of auditing business sequences needing to be executed and comprises a rule part, a conditional branch and a branch aggregation part;

the intelligent auditing step is used for realizing the auditing of the business data of the auditing platform.

The basic information of the instantiated project comprises a project name, a project number, a project introduction, a service database account and a password; after the adding application is submitted, a corresponding database account is established in the database, the established database account is used as a business database of an instantiated project, and business data of the instantiated project is stored;

the data table information comprises a data table number and a generation mark; the generation mark indicates whether the data table information generates a corresponding service JavaBean and whether a corresponding service data table is generated in a database; the serial number of the data table is used as the table name of the service data table and the class name of the service JavaBean;

the field information comprises a field number and an concerned identification, the field number is used as a field name of a business data table and a class member variable name of a business JavaBean, the concerned identification represents whether the field is concerned data, when business audit, the data of the data table triggers a business rule, the field data of the concerned identification needs to be spliced on an audit result and returned to an audit requester.

The project building step also comprises modification, viewing and deletion of the instantiated project.

The intelligent auditing step comprises an in-affair auditing sub-step and an after-affair auditing sub-step; the in-process auditing sub-step is in a non-pipeline mode of linear execution, and the after-process auditing sub-step is in a pipeline mode of parallel execution.

The in-process auditing substep comprises the following substeps:

s211: the audit data interface captures an audit request; the auditing request is an auditing request of auditing a data interface in the affair of an auditing platform, wherein the auditing request is an auditing request of an auditing platform, which is to encapsulate verification information, auditing requirements and service data into a JSON data packet by an auditing requester, then encrypt and compress the JSON data packet, set the JSON data packet as interface input data and transmit the interface input data to the auditing platform;

s212: the audit data interface decompresses and decrypts the uploaded data, analyzes the verification information, compares and verifies the verification information and audit service registration information, and if the verification is passed, the step S213 is executed; otherwise, the audit data interface returns the audit service limited prompt message to the audit requester;

s213: the in-process audit data interface analyzes the audit requirement, if the audit requirement is successfully analyzed, an audit task is created through the task management subunit, the audit task is initialized, information of a requester is bound, the audit requirement is set, and the step S214 is carried out; otherwise, the audit data interface returns error information to the audit requester;

s214: the business audit data interface analyzes the business data based on the business data structure, if the business data is successfully analyzed, the business audit data interface converts the business data into an instantiated business JavaBean object, and the JavaBean object is transferred to a business audit engine for business audit; otherwise, the task management subunit cancels the audit task, and the audit data interface returns error information to the audit requester;

s215: the in-service auditing engine extracts a service rule packet related to the auditing from the compiled service rule base according to the auditing requirement of the auditing task, creates an auditing session based on the extracted service rule packet, loads a service data JavaBean object into the auditing session, and executes the auditing session;

s216: after the auditing session is finished, assembling an auditing result according to a returned result standard format;

s217: after the assembly is completed, the audit data interface module encrypts and compresses an assembly result according to the address of the requester bound by the audit task and sends the assembly result to the audit requester; meanwhile, after the data are successfully sent, the audit related data and the audit task information are backed up to the database, the audit task is persisted, the task management changes the task audit state into the audit success, the audit thread is finished and is returned to the thread pool, and the audit is finished in the whole process.

The post audit substep comprises the following substeps:

s221: the after-the-fact audit data interface captures an audit request; the verification request is a verification request which is transmitted to a verification platform by a verification requester after verification information, verification requirements and service data are packaged into JSON data packets, then encrypted and compressed and set as interface input data;

s222: the post audit data interface decompresses and decrypts the uploaded data, analyzes the verification information, compares and verifies the verification information and audit service registration information, and if the verification is passed, the step S223 is executed; otherwise, the audit data interface returns the audit service limited prompt message to the audit requester after the fact;

s223: the audit data interface analyzes the audit requirement afterwards, if the audit requirement is successfully analyzed, the task management subunit creates an audit task, initializes the audit task, binds information of a requester, sets the audit requirement, sets a task state as audit data to be analyzed, and then the step S224 is carried out; otherwise, the post-audit data interface returns error information to the audit requester, the thread is finished and the thread is returned to the thread pool;

s224: the post-audit data interface analyzes the service data based on the service data structure, if the service data is successfully analyzed, the post-audit data interface converts the service data into a service JavaBean object, the service JavaBean object binds an audit task number, stores the audit task number into a memory cache, mounts the audit task to a queue to be audited, changes the task state into to-be-audited, transfers to a post-audit engine for service audit, returns the audit task number to an audit requester, finishes the thread and returns the thread to a thread pool, and then the step S225 is executed; otherwise, changing the task state into failure of auditing data analysis, returning error information to an auditing requester, ending the thread and returning the thread to the thread pool;

s225: the auditing component, the assembling component, the result returning component and the backup component of the post-event auditing engine respectively process the corresponding queues:

(1) the auditing component monitors whether the queue to be audited is to be audited, when the queue to be audited is monitored to be audited to be awaiting audited, the task to be audited is taken out from the head of the queue, after taking out, an idle thread is taken out from the thread pool to execute the task to be audited, the task management subunit changes the state of the auditing task into auditing, and an auditing engine extracts a service rule packet related to the auditing from a compiled service rule base after the auditing according to the auditing requirement of the auditing task; creating an audit session based on the extracted service rule packet, extracting a service JavaBean object from a memory cache according to the task number, loading the service JavaBean object into the audit session, and executing the audit session; after the auditing session is finished, the auditing result is bound with a task number, the task number is stored in a memory for caching, the auditing task state is changed into a state to be assembled, the auditing task is mounted in a queue to be assembled, the thread is finished, and the thread is returned to a thread pool;

(2) the assembly component monitors whether a queue to be assembled is a task to be assembled or not, when the task to be assembled of the queue to be assembled is monitored, the task to be assembled is taken out from the head of the queue, an idle thread is taken out from the thread pool after the task to be assembled is taken out, the state of the audit task is changed into the assembly, the audit result is extracted from the memory cache according to the task number, the audit result is assembled according to a return result standard format, after the assembly is completed, the task number is bound, the assembly result is stored in the memory cache, the state of the audit task is changed into the state to be returned, the audit task is mounted to the queue to be returned, the thread is finished, and the thread is returned;

(3) the result is returned to the assembly and is monitored whether the queue to be returned is to be returned or not, when the queue to be returned is monitored to be returned is to be returned, the task to be returned is taken out from the head of the queue, an idle thread is taken out from the thread pool after the task to be returned is taken out, the state of the audit task is changed into the return, an assembly result is extracted from the memory cache according to the task number, an after-event audit data interface encrypts and compresses the assembly result according to the address of a requester bound by the audit task, the assembly result is sent to a requester, after the audit task is successfully sent, the state of the audit task is changed into the standby state, the audit task is mounted on the queue to be backed up, the thread is;

(4) the backup component monitors whether a queue to be backed up has a task to be backed up, when the queue to be backed up has the task to be backed up, the task to be backed up is taken out from the head of the queue, an idle thread is taken out from the thread pool after the task to be backed up is taken out, the state of the audit task is changed into backup, all data related to the audit task in the memory cache are persisted and stored in a database, after the data are stored, all data of the audit task in the memory cache are cleared, the pressure of the memory cache is reduced, the state of the audit task is changed into audit success, the thread is ended, and the thread is returned to the thread pool.

The invention has the beneficial effects that:

(1) the platform auditing field generalization is realized, the auditing field is not limited to specific fields (financial reimbursement, administrative approval and the like), the auditing field during the operation of the auditing platform is a field related to an instantiated project selected from a project library by a user, and the project library can contain instantiated projects in a plurality of fields, so that the scheme of the auditing platform supports a plurality of auditing fields, and the auditing field generalization is realized.

(2) The method has the advantages that the service data independence is realized, the actual processed service data is not concerned when the auditing platform is developed, and the problems that in the prior art, the service data structure needing to be processed is set by default during development, only the service data corresponding to the service data structure set by default can be analyzed after deployment, the limitation of the auditing platform for processing the service data range is highlighted, and the auditing platform has defects are solved.

(3) The independence of the service auditing logic, the actual processing service auditing logic is not concerned when the auditing platform is deployed, and the problems that the specific service auditing logic needs to be determined during development and is solidified in codes during the development process, and the expansibility and flexibility of the system are greatly reduced in the prior art are solved.

(4) Different auditing application scenes are supported, and the auditing platform supports two auditing modes of auditing in the process and auditing after the process. The service auditing modes are different according to different auditing requirements of real-time performance, response speed and data volume, and the corresponding processing modes are different.

Drawings

FIG. 1 is a schematic view of a hierarchical structure of a review platform module;

FIG. 2 is a schematic diagram of a project building block based on a B/S architecture;

FIG. 3 is a diagram of instantiated items;

FIG. 4 is a diagram illustrating a project library architecture and an implementation of an audit platform;

FIG. 5 is a schematic diagram of a business rule base unit;

FIG. 6 is a business flow diagram organizational chart diagram of a rule flow;

FIG. 7 is a flowchart of an audit in fact;

FIG. 8 is a flowchart of post audit.

Detailed Description

The technical scheme of the invention is further described in detail by combining the attached drawings:

the auditing platform of embodiment 1 is composed of a project construction module and an intelligent auditing module, wherein the project construction module realizes the generalization of auditing fields and provides support for intelligent auditing to execute auditing tasks; the intelligent auditing module realizes the auditing of the service data and provides an optional auditing application scene. The module hierarchy of the present embodiment is shown in fig. 1.

As shown in fig. 2, in this embodiment, the project building module is based on a B/S architecture, and a user logs in an intelligent audit platform through a terminal browser to manage a project library; by adopting the mode of the embodiment, the user can rapidly manage the project library. In another embodiment, the project building module is based on a C/S architecture, and a user logs in the intelligent auditing platform after installing a corresponding client through a terminal to manage the project library.

The project construction realizes platform audit field generalization, the audit field is not limited to specific fields (financial reimbursement, administrative approval and the like), the audit field during the operation of the audit platform is a field related to an instantiated project selected from a project library by a user, and the project library can contain instantiated projects of a plurality of fields, so that the scheme of the audit platform supports a plurality of audit fields, and realizes platform audit field generalization.

The project building module is responsible for managing a project library, which, as shown in FIG. 4, consists of zero or more instantiated projects.

As shown in fig. 3, the instantiated item is composed of a business data unit, a business database unit, a business JavaBean unit and a business rule base unit. Specifically, the method comprises the following steps:

the service data unit manages a service data structure. The service data structure consists of data table information and field information, provides an analysis scheme for analyzing JSON data packets uploaded by an audit request party, and provides support for management of a service database unit, a service JavaBean unit and a service rule base unit.

And the service database unit manages the service database based on the service data structure and provides physical storage support for the persistent audit data.

And the service JavaBean unit manages the service JavaBean set based on the service data structure. When the audit request arrives, support is provided for converting the audit data into the instantiated business JavaBean object. Through the conversion operation, the auditing platform can operate the auditing data and complete the data processing operation in the auditing process.

The business rule base unit converts the business audit logic of the instantiated items into rules and rule flows that the Drools rule engine can recognize. When the auditing platform audits the service data, the auditing data are audited according to the rules or rule flow defined by the user.

In this embodiment, managing the project library includes instantiating project additions, modifications, views, and deletions.

Specifically, for instantiated item addition:

in this embodiment, first, basic information such as a project name, a project number, a project introduction, a business database account, and a password is added under the guidance of an instantiated project adding template.

The project number is used as a verification platform to distinguish instantiation project identifiers and is composed of English, numbers and underlines, and the first position is English; and after the adding application is submitted, a corresponding database account is established in the database, the established database account is used as a service database of the instantiated project, and the service data of the instantiated project is stored.

And when the instantiation item is added successfully, the instantiation item management is entered. The instantiated project management comprises a business data module, a business database module, a business JavaBean module and a business rule module.

For the added service data structure, the data table information and the field information need to be added respectively:

under the guidance of a data table information adding template, adding data table information, wherein the data table information comprises a data table name, a data table number, a data table introduction and a generation mark, and the data table information comprises:

the name of the data table is Chinese and is the name of the Chinese of the data table;

the data table is introduced into Chinese and is introduced in detail;

generating a mark to indicate whether the data table information generates a corresponding service JavaBean or not, and whether a corresponding service data table is generated in a database or not;

the data table number is composed of English, number and underline, and the first position is English as the table name of the service data table and the class name of the service JavaBean.

After the data table information is added, all field information of the service data structure is added under the guidance of a field information adding template, wherein the field information comprises field names, field descriptions, field numbers, field data types and sizes, primary key marks, nullable marks, default values and concerned marks, and the method comprises the following steps:

the field name is Chinese and is the Chinese name of the field;

the field is described as Chinese, and the field is introduced in detail;

the field number is used as a field name of the service data table and a class member variable name of the service JavaBean, and is composed of English, numbers and underlines, and the first position is English;

the field data type and size correspond to the field data type and size of the database, such as varchar2(40), number (10), number (8,2) and the like, and serve as the data type and size of the field of the service data table;

the primary key mark indicates whether the field is a primary key in the service data table;

the nullable identifier indicates whether the field can be null in the service data table;

the default value represents a default value when the field is empty;

the concerned mark indicates whether the field is concerned data, when the data of the data table triggers the business rule during the business check, the field data which is the concerned data needs to be spliced on the check result and returned to the check requester.

In this embodiment, after all the field information of the service data structure is added, the service data module adds a task number field and a task data index field to the service data structure, where:

the task number field records which audit task the data belongs to, and the data are clustered and distinguished by the task number;

the task data index field identifies which service data of the audit task the data is, so that the service data of the audit task can be distinguished and positioned conveniently.

The task number field and the task data index field are invisible to a user, and are required for auditing the internal operation of the platform, and the two fields are automatically generated when each business data structure is added.

After the task number field and the task data index field are automatically generated, the business data module associates the data table information and the field information of the business data structure, and associates the data table information and the field information of the business data structure by binding the data table number to each field information.

After the series of operations, the service data structure addition is completed.

During the process of adding the service data structure, there may be modification and deletion of the data table information and the field information. And realizing the modification and deletion processes of the processes such as the information system under the condition that the corresponding service data table and the service JavaBean are not generated.

An instantiated project may include multiple business data structures, each with an add flow as described above.

On the premise of completing the construction of the service data structure, the construction sequence of the service database module, the service JavaBean set module and the service rule base module is not related and is carried out in any sequence.

In this embodiment, on the premise that the service data structure is successfully added, the service database module constructs and creates a data table SQL statement according to the data table number of the service data structure and the field information bound to the data table number, where the field information includes the field number, the field data type and size, the key sign, the nullable flag, and the default value. And after the composition is finished, executing the structured SQL statement in the instantiated project business database, finishing the creation of the business data table, and updating the generation mark of the business data structure data table information to indicate that the business data table is generated.

And the business database module generates corresponding business data tables from all business data structures of the instantiated project one by one to construct a business database.

In this embodiment, on the premise that the service data structure is successfully added, the service JavaBean module constructs a java file corresponding to the service JavaBean according to the data table number of the service data structure and the field information bound to the data table number, where the field information includes the field number, the field data type, and the size.

The construction scheme is as follows: the data table number is treated by a hump naming method, and the initial letter is capitalized to be used as a class name. After the field number is processed by a hump naming method, the field number is used as a class member variable name, and the data type of the class member variable is determined according to the data type and the size of the field, for example, number (4) is Integer, number (10) is Long, varchar2 is String, and the like.

And after constructing the class member variables, constructing corresponding get and set access methods for each class member variable. If the access method of the class member variable userName is setUserName and getUserName.

After the java file of the service JavaBean is constructed, compiling the java file to generate a class file by using a java language programming compiler java c, and updating a generation mark of the service data structure data table information to indicate that the service JavaBean is generated. And when the auditing platform is started, the generated class file is loaded to the JVM, uploaded auditing data can be converted into instantiated JavaBean objects, the platform can operate the auditing data in the JVM, and meanwhile, the auditing data JavaBean objects can be inserted into an auditing session to be used as fact objects in an auditing engine Work Memory to match with business rules for auditing.

And the service JavaBean module generates corresponding service JavaBean for all the service data structures of the instantiated project one by one, and constructs a service JavaBean set.

In this embodiment, after the adding of the business data structure of the instantiated project is completed, the business rule base is added.

As shown in FIG. 5, the business rules based on the Drools rules engine are divided into rules and rule flows.

Wherein, the structure of the rule is as follows:

rule "name" - -rule name

attributes-Attributes part

when

LHS- -Condition part

the then RHS- -results section

end

A rule typically includes three parts, an attribute part (attribute), a condition part (LHS), and a result part (RHS).

Attribute part (attribute) defining some attributes of the current rule execution, such as whether it can be repeatedly executed, expiration time, validation time, etc.

Condition part (LHS) conditions defining the current rule, which may be single or multiple. The service data structure provides support for the conditional part construction.

A result section (RHS) defines the operations performed after the current conditions section is satisfied. The service data structure provides support for the construction of the result part.

As shown in fig. 6, the rule flow is implemented based on a business flow diagram. The business flow chart describes a series of auditing business sequences needing to be executed, and the auditing business sequence consists of three parts of rules, conditional branches and branch aggregation.

The business rule base module is responsible for managing the business rule base, including adding, modifying, checking and deleting the business rules.

In this embodiment, the addition/modification/viewing/deletion of business rules: based on a Web browser mode, a user adds/modifies/checks/deletes rules or rule flows under the guidance of a business rule adding template, and after the addition/modification/checking/deletion is submitted, a checking platform generates business rules in a rule base.

And after the adding of the service data structure, the service database, the service JavaBean set and the service rule base is completed, the instantiation project adding process is completed.

In the present embodiment, for instantiated item modification:

under the guidance of the instantiated project modification template, the project name and the project introduction can be modified, and the project number, the account number of the service database and the password cannot be modified.

And modifying the service data structure through the service data module, wherein the modification comprises data table information modification and field information modification.

The data table name and the data table introduction of the data table information can be modified, and the data table number and the generation flag cannot be modified.

The field name, field description and concerned identification of the field information can be modified, and the service database and the service JavaBean set are not influenced.

The field number of the field information cannot be modified, and the field data type and size, the primary key flag, the nullable flag, and the default value can be modified.

When the type and the size of the field data are modified, the business database module is triggered to modify the business database and the business JavaBean module is triggered to modify the business JavaBean set.

When the primary key flag, nullable flag and default values are modified, the service database module is triggered to modify the service database.

And for the condition that the field data type and size, the nullable identifier and the default value are modified, the business database module acquires the latest field data type and size, the nullable identifier and the default value according to the bound data table number and the field number, then constructs a modified database field SQL statement, and executes the constructed SQL statement in the instantiated project business database to complete the modification of the corresponding business data table.

And for the condition that the main key mark is modified, the business database module traverses the bound field information according to the bound data table number, inquires all field numbers which are the main keys, constructs and creates a main key SQL statement, and executes the operation of deleting the existing main keys and then executes the constructed SQL statement in the instantiated project business database to complete the main key modification of the corresponding business data table.

For the case that the field data type and size are modified, the service JavaBean module constructs a java file corresponding to the service JavaBean according to the number of the bound data table and the field information bound with the number of the data table, wherein the field information comprises the field number, the field data type and the size, and the construction scheme is as above. And after the java file is constructed, replacing the existing java file in a covering manner, compiling the new java file by using a java language programming compiler java to generate a new class file, and replacing the existing class file in a covering manner.

The business rule base may be modified by the business rule module.

In the present embodiment, for instantiated item modification:

after the instantiated project deletion is initiated, the auditing platform deletes all data related to the project, including project information, a business data structure, a business rule base, a business JavaBean set, a business database and the like. And logging out the account number of the service database.

The core operation of the project building is as described above.

A generalized implementation scheme in the platform audit field is shown in fig. 4, where an audit platform selects an instantiated project from a project library to deploy and start according to user configuration information when starting.

The intelligent audit has two audit modes of audit in the prior and after-the-fact audits.

The in-service auditing is suitable for auditing tasks with high real-time requirement, high response speed and small data volume, and comprises a first auditing data interface, a first task management and an in-service auditing engine. Wherein: the first audit data interface is responsible for managing data communication between the audit platform and the audit requester, and comprises monitoring audit requests, analyzing the audit requests, returning audit results and the like; the first task management is responsible for maintaining the task table, creating the audit task, changing the task state, ending the audit task and the like, and all audit task information is recorded in the task table, so that the audit platform can manage the audit task conveniently; the in-office audit engine is responsible for processing in-office audit, including business data audit, audit result assembly and audit task persistence.

The audit engine comprises a business data audit component, an audit result assembly component and an audit task persistence component. The business data auditing component audits the business data; the auditing result assembly component assembles the auditing result according to a returned result standard format; and the nuclear task persistence component persists the audit task information, the service data, the audit result and the like to the physical database.

In this embodiment, the auditing process of auditing is a non-pipelined manner, and is implemented linearly and synchronously. The auditing process in the world is shown in FIG. 7:

s11: and the verification requester encapsulates the verification information, the verification requirement and the service data into JSON data, then encrypts and compresses the JSON data, sets the JSON data as interface input data, calls the verification interface in affairs to initiate a request, and transmits the verification request to the verification platform through the Internet.

S12: the first audit data interface captures an audit request, decompresses and decrypts the uploaded data, analyzes verification information, and compares and verifies the verification information and audit service registration information: if the verification is passed, go to step S13; and if the verification fails, the first audit data interface returns audit service limited prompt information and ends.

S13: the first audit data interface analyzes the audit requirement: if the audit request is successfully analyzed, the first task manages and creates the audit task, initializes the audit task, binds the information of the requester, sets the audit request, and goes to step S14; and if the auditing requirement analysis is unsuccessful, the first auditing data interface returns error information to the requester.

S14: the first audit data interface analyzes the service data based on the service data structure, if the service data is successfully analyzed, the data interface converts the service data into an instantiated service JavaBean object, and the JavaBean object is transferred to an in-service audit engine for service audit, namely step S15; otherwise, the first task management cancels the audit task, and the first audit data interface returns error information to the requester;

s15: and the in-process auditing engine extracts the business rule packet related to the auditing from the compiled business rule base according to the auditing requirement of the auditing task. And creating an auditing session based on the extracted service rule packet, loading the service data JavaBean object into the auditing session, and executing the auditing session.

S16: and after the auditing session is finished, the auditing result is assembled according to the returned result standard format, and after the assembly is finished, the first auditing data interface encrypts and compresses the assembly result according to the address of the requester bound by the auditing task and sends the encrypted and compressed assembly result to the auditing requester. After the successful sending, the related audit data and the audit task information are backed up to the database, the persistence of the audit task is realized, the first task management changes the task audit state to be successful, the audit thread is finished and returns the audit thread to the thread pool, and the audit is finished in the whole process.

The post-audit is suitable for audit tasks with low real-time requirement, high response speed and large data volume, and comprises a second audit data interface, a second task management and a post-audit engine.

Wherein: the second audit data interface is responsible for managing data communication between the audit platform and the audit requester, including monitoring audit requests, analyzing the audit requests, returning audit results and the like; the second task management is responsible for maintaining the task table, creating the audit task, changing the task state, ending the audit task and the like, and all audit task information is recorded in the task table, so that the audit platform can manage the audit task conveniently; the in-office audit engine is responsible for processing in-office audit, including business data audit, audit result assembly and audit task persistence.

The post audit engine comprises a business data audit component, an audit result assembly component, a result return component and an audit task persistence component. The business data auditing component audits the business data; the auditing result assembly component assembles the auditing result according to a returned result standard format; the result returning component returns the assembled auditing result to the auditing requester based on the second auditing data interface; and the task persistence component persists the audit task information, the service data, the audit result and the like to the physical database.

In this embodiment, the audit process of post audit is a pipeline mode, parallel execution and asynchronous process. The specific flow is shown in fig. 8:

s201: and the verification requester encapsulates the verification information, the verification requirement and the service data into JSON data, encrypts and compresses the JSON data, sets the JSON data as interface input data, calls a post-event verification interface to initiate a request, and transmits the verification request to the verification platform through the Internet.

S202: the second audit data interface module captures the audit request, decompresses and decrypts the uploaded data, analyzes the verification information, and compares and verifies the verification information and the audit service registration information: if the verification is passed, the step S203 is entered; otherwise, the verification fails, and the data interface returns prompt information about limited access and audit service;

s203: and the second audit data interface analyzes the audit requirement: if the audit request is successfully analyzed, the task management establishes an audit task, initializes the audit task, binds information of a requester, sets the audit request, sets the task state as audit data to be analyzed, and enters step S204; otherwise, the data interface returns error information to the requester, the thread is finished and the thread is returned to the thread pool.

S204: the second audit data interface analyzes the service data based on the service data structure: if the analysis of the service data is successful, the second audit data interface converts the service data into a service JavaBean object, the service JavaBean object binds an audit task number, stores the audit task number into a memory cache, mounts the audit task to a queue to be audited, changes the task state into a state to be audited, shifts to an audit engine for service audit after the fact, returns the audit task number to an audit requester (the audit requester can conveniently check and manage the audit task process), finishes the thread, returns the thread to a thread pool, and enters the step S205; otherwise, changing the task state into failure of auditing data analysis, returning error information to the requester, ending the thread and returning the thread to the thread pool.

S205: the business data auditing component, the auditing result assembling component, the result returning component and the auditing task persistence component of the post-auditing engine respectively process the corresponding queues:

(1) and the business data auditing component monitors whether the queue to be audited is to be audited, when the queue to be audited is monitored to be audited to be awaiting audited, the task to be audited is taken out from the queue head, an idle thread is taken out from the thread pool after the task to be audited is taken out to execute the task to be audited, the second task management changes the state of the auditing task to be auditing, and an auditing engine extracts the business rule packet related to the auditing from the compiled business rule base after the auditing according to the auditing requirement of the auditing task. And establishing an audit session based on the extracted service rule packet, extracting the service JavaBean object from the memory cache according to the task number, loading the service JavaBean object into the audit session, and executing the audit session. And after the audit session is finished, the audit result binds a task number, stores the task number into a memory cache, changes the state of the audit task into a to-be-assembled state, mounts the audit task into a to-be-assembled queue, finishes the thread and returns the thread to the thread pool.

(2) The auditing result assembling component monitors whether the queue to be assembled is a task to be assembled or not, when the task to be assembled is monitored, the task to be assembled is taken out from the head of the queue, an idle thread is taken out from the thread pool after the task to be assembled is taken out, the state of the auditing task is changed into the assembly, the auditing result is extracted from the memory cache according to the task number, the auditing result is assembled according to a return result standard format, after the assembly is completed, the task number is bound, the assembly result is stored in the memory cache, the state of the auditing task is changed to be returned, the auditing task is mounted to the queue to be returned, the thread is finished, and the thread is returned to the thread pool.

(3) The result is returned to the component and is monitored whether the queue to be returned is to be returned or not, when the queue to be returned is monitored to be returned is to be returned, the task to be returned is taken out from the queue head, an idle thread is taken out from the thread pool after the task to be returned is taken out, the state of the audit task is changed into the return, the assembly result is extracted from the memory cache according to the task number, the data interface module encrypts and compresses the assembly result according to the address of a requester bound by the audit task, the assembly result is sent to a requester, after the successful sending, the state of the audit task is changed into the backup, the audit task is mounted to the queue to be backed up, the thread is finished, and the thread is returned.

(4) The auditing task persistence component monitors whether a queue to be backed up has a task to be backed up, when the queue to be backed up has the task to be backed up, the task to be backed up is taken out from the head of the queue, an idle thread is taken out from the thread pool after the task to be backed up is taken out to execute the task to be backed up, the state of the auditing task is changed into backup, all data related to the auditing task in the memory cache are persisted and stored in a database, after the data are stored, all data of the auditing task in the memory cache are cleared, the pressure of the memory cache is reduced, the state of the auditing task is changed into successful auditing, the thread is finished and returned to the thread pool, and the auditing is finished after.

The medical insurance reimbursement auditing, such as the reimbursement auditing of all expenses generated by one-time hospitalization of a patient, is characterized in that the auditing data volume is large, the second-level auditing time is allowed, the requirement on real-time performance is not high, and the auditing can be executed in a post-auditing mode. For example, when a service application is handled on a counter, a service applicant is required to be checked whether to qualify for handling the service, and then the service applicant can enter a service handling process.

Claims

1. The utility model provides a platform is examined and examined to universalization intelligence which characterized in that: the system comprises a project construction module and an intelligent auditing module;

the intelligent auditing module is used for realizing the auditing of the service data of the auditing platform;

2. The generalized intelligent audit platform according to claim 1, wherein: the in-service auditing engine is an auditing engine of a linear execution non-pipeline mode; the business audit engine comprises a business data audit component, an audit result assembly component and an audit task persistence component, wherein the business data audit component is used for auditing business data, the audit result assembly component is used for assembling audit results according to a return result standard format, and the audit task persistence component is used for persisting audit task information, business data and audit results to a physical database.

3. The generalized intelligent audit platform according to claim 1, wherein: the post-event auditing engine is an auditing engine of a parallel execution pipeline mode; the in-service audit engine comprises a business data audit component, an audit result assembly component, a result return component and an audit task persistence component, wherein the business data audit component is used for auditing business data, the audit result assembly component is used for assembling audit results according to a returned result standard format, the result return component returns the assembled audit results to an audit requester based on a second audit data interface subunit, and the audit task persistence component is used for persisting audit task information, business data and audit results to a physical database.

4. A universal intelligent auditing method is characterized in that: the method comprises a project construction step and an intelligent auditing step; the project building step comprises adding instantiated projects, and the adding of the instantiated projects comprises the following sub-steps:

s1312: executing the structured SQL statement in the instantiated project business database, completing the creation of a business data table, and updating a generation mark of data table information of the business data structure to indicate that the business data table is generated;

the construction of the service JavaBean set comprises the following sub-steps:

s1322: compiling the java file to generate a class file by using a java language programming compiler java, and updating a generation mark of data table information of the service data structure to indicate that the service JavaBean is generated;

the construction of the business rule base comprises the following sub-steps:

5. The generalized intelligent auditing method of claim 4, characterized in that: the basic information of the instantiated project comprises a project name, a project number, a project introduction, a service database account and a password; after the adding application is submitted, a corresponding database account is established in the database, the established database account is used as a business database of an instantiated project, and business data of the instantiated project is stored;

6. The generalized intelligent auditing method of claim 4, characterized in that: the project building step also comprises modification, viewing and deletion of the instantiated project.

7. The generalized intelligent auditing method of claim 4, characterized in that: the intelligent auditing step comprises an in-affair auditing sub-step and an after-affair auditing sub-step; the in-process auditing sub-step is in a non-pipeline mode of linear execution, and the after-process auditing sub-step is in a pipeline mode of parallel execution.

8. The generalized intelligent auditing method of claim 7, characterized in that: the in-process auditing substep comprises the following substeps:

9. The generalized intelligent auditing method of claim 7, characterized in that: the post audit substep comprises the following substeps:

(4) the backup component monitors whether a queue to be backed up has a task to be backed up, when the queue to be backed up has the task to be backed up, the task to be backed up is taken out from the queue head, an idle thread is taken out from the thread pool after the task to be backed up is taken out, the state of the audit task is changed into backup, all data which are stored in the memory cache and are related to the audit task are durably stored in a database, after the data are stored, all data of the audit task in the memory cache are cleared, the pressure of the memory cache is reduced, the state of the audit task is changed into audit success, the thread is ended, and the thread is returned to the thread pool.