Disclosure of Invention
The invention provides a method and a device for rapid abstract modeling of business application, which mainly aim to improve the intelligent degree and flexibility of business processing and improve the efficiency of business processing.
In order to achieve the above object, the present invention provides a method for fast abstract modeling of service applications, including:
Receiving an application modeling instruction, and determining a service to be modeled and original service data based on the application modeling instruction;
Determining the service application field of the service to be modeled, and acquiring a service abstract application library based on the service application field, wherein the service abstract application library comprises a plurality of service abstract applications, and each service abstract application is provided with a parameter interface;
Carrying out service division on the service to be modeled to obtain a unit service group, wherein the unit service group comprises a plurality of unit services, and each unit service is provided with a parameter interface;
Constructing an effective service application group based on a service abstract application library and a unit service group, wherein the effective service application group comprises a plurality of effective service applications, the effective service applications comprise public service applications and proprietary service applications, and the public service applications come from the service abstract application library;
determining the application scene of the service to be modeled, and executing dedicated parameter setting on the effective service application group based on the application scene to obtain a target service application group;
arranging a service scene application model based on the target service application group;
According to the original service data and the service scene application model, carrying out service automation processing to obtain a service execution result;
performing intelligent analysis on the service execution result according to a pre-constructed case library to obtain executable confidence, wherein the intelligent analysis refers to analysis by using a large language model;
If the executable confidence coefficient is not greater than the confidence coefficient threshold value, manually correcting the service execution result to obtain a service correction result, and marking the service correction result as an executable result;
if the executable confidence is greater than a confidence threshold, marking the service execution result as an executable result;
And carrying out case data combination based on the original service data and the executable result to obtain a current case, supplementing the current case to the case library, and completing the quick abstract modeling of the service application.
Optionally, the acquiring a service abstract application library based on the service application field includes:
constructing a business application process in the business application field, wherein the business application process comprises a plurality of business process nodes;
Sequentially extracting business process nodes from a plurality of business process nodes of a business application process, and defining node interfaces of the business process nodes to obtain programmable business nodes, wherein the node interfaces comprise an input parameter interface and an output parameter interface;
Performing application development on the programmable service node to obtain a service abstract application, wherein the application development comprises language programming;
and summarizing the business abstract application to obtain a business abstract application library.
Optionally, the constructing an effective service application group based on the service abstract application library and the unit service group includes:
sequentially extracting unit services from the unit service group, and determining flow data of the unit services, wherein the flow data comprises input data and output data;
judging whether a business abstract application matched with the unit business exists in a business abstract application library or not based on the streaming data;
if the business abstract application library contains the business abstract application matched with the unit business, extracting public business application from the business abstract application library;
if the service abstract application matched with the unit service does not exist in the service abstract application library, constructing a special service application of the unit service;
summarizing the public service application or the special service application to obtain an effective service application group.
Optionally, the determining whether the service abstraction application matched with the unit service exists in the service abstraction application library includes:
Sequentially extracting service abstract applications from a service abstract application library, confirming service node interfaces of the service abstract applications, and judging whether the unit service is matched with the service abstract applications or not based on streaming data and the service node interfaces;
if the unit service is confirmed to be not matched with the service abstract application, the service abstract application is marked as a non-matched application;
Summarizing the non-matching application to obtain a non-matching application set;
if the number of the non-matching applications in the non-matching application set is equal to the number of the business abstraction applications in the business abstraction application library, no business abstraction application matched with the unit business exists in the business abstraction application library;
Otherwise, there is a business abstraction application in the business abstraction application library that matches the unit business.
Optionally, based on the application scenario, dedicated parameter setting is performed on the effective service application group to obtain a target service application group, including:
Sequentially extracting effective service applications from the effective service application group, and confirming configuration options in the effective service applications;
Acquiring exclusive configuration parameters of an application scene, and updating configuration options in the effective service application by utilizing the exclusive configuration parameters to obtain a target service application;
and summarizing the target business application to obtain a target business application group.
Optionally, the arranging the service scene application model based on the target service application group includes:
determining a target service flow of the service to be modeled, wherein the target service flow comprises a plurality of target service nodes, and each target service node corresponds to one target service application in a target service application group;
Determining an application connection relation of each target service application in the target service application group based on a target service flow to obtain an application connection relation group, wherein the application connection relation group comprises a series relation and a parallel relation;
and integrating each target service application in the target service application group according to the application connection relation group to obtain a service scene application model, wherein the integration mode comprises API interface calling.
Optionally, the performing intelligent analysis on the service execution result according to the pre-constructed case library to obtain an executable confidence coefficient includes:
Searching a case set in the same field in a case base based on the service application field, wherein the case set in the same field comprises a plurality of cases in the same field, and the cases in the same field are cases in the service application field;
sequentially extracting cases in the same field in the case set in the same field, determining a case service application group of the cases in the same field, and acquiring case configuration parameters corresponding to each case service application in the case service application group to obtain a case configuration parameter group;
Determining case input data of the same-field case;
Based on case input data and case configuration parameter sets, carrying out similarity analysis on cases in the same field and the business to be modeled to obtain business similarity;
if the service similarity is greater than a preset similarity threshold, marking the service similarity as effective service similarity;
summarizing the effective service similarity to obtain an effective service similarity group;
If the effective service similarity group is a preset empty set, marking the executable confidence coefficient as 0;
If the effective service similarity group is not an empty set, extracting a similar case group corresponding to the effective service similarity group from the case set in the same field;
And evaluating the service execution result by using the similar case group to obtain the executable confidence coefficient.
Optionally, the evaluating the service execution result by using the similar case group to obtain the executable confidence coefficient includes:
sequentially extracting similar cases from the similar case group, and obtaining case execution results of the similar cases;
performing similarity evaluation on the case execution result and the service execution result by using the large language model to obtain action similarity;
Summarizing the action similarity to obtain an action similarity group, and calculating executable confidence based on the action similarity group and the effective service similarity group, wherein the executable confidence is expressed as:
Wherein, theIndicating the degree of confidence in the executability,Represents the number of action similarities in the action similarity group or the number of effective service similarities in the effective service similarity group,AndRespectively represent the first of the valid service similarity groupsIndividual effective business similarity and the firstA degree of similarity of the effective traffic is determined,Representing the first of the group of action similaritiesAnd each action similarity.
Optionally, the manually correcting the service execution result to obtain a service correction result includes:
confirming an abnormal similar case group of the executable confidence in a case library, and obtaining an abnormal execution result group corresponding to the abnormal similar case group;
inputting the abnormal execution result set and the business execution result into a large language model to obtain a risk point set, wherein the risk point set comprises a plurality of risk points;
and generating a manual task list based on the risk point group, and manually correcting by using the manual task list to obtain a service correction result.
In order to achieve the above object, the present invention further provides a device for rapid abstract modeling of business applications, including:
The business application construction module is used for receiving an application modeling instruction, determining a business to be modeled and original business data based on the application modeling instruction, determining a business application field of the business to be modeled, and acquiring a business abstract application library based on the business application field, wherein the business abstract application library comprises a plurality of business abstract applications, and each business abstract application is provided with a parameter interface;
The system comprises a dedicated parameter setting module, a service abstraction application library and a dedicated parameter setting module, wherein the dedicated parameter setting module is used for carrying out service division on the service to be modeled to obtain a unit service group, the unit service group comprises a plurality of unit services, each unit service is provided with a parameter interface, an effective service application group is constructed based on the service abstraction application library and the unit service group, the effective service application group comprises a plurality of effective service applications, the effective service applications comprise public service applications and dedicated service applications, the public service applications come from the service abstraction application library, the application scene of the service to be modeled is determined, and dedicated parameter setting is executed on the effective service application group based on the application scene to obtain a target service application group;
the language model analysis module is used for arranging a service scene application model based on a target service application group, carrying out service automation processing according to original service data and the service scene application model to obtain a service execution result, and carrying out intelligent analysis on the service execution result according to a pre-constructed case library to obtain an executable confidence coefficient, wherein the intelligent analysis means that a large language model is utilized for analysis;
and the execution result correction module is used for manually correcting the service execution result to obtain a service correction result and marking the service correction result as an executable result if the executable confidence is not greater than a confidence threshold, marking the service execution result as an executable result if the executable confidence is greater than the confidence threshold, merging case data based on the original service data and the executable result to obtain a current case, and supplementing the current case to the case library.
In order to solve the above-mentioned problems, the present invention also provides an electronic apparatus including:
A memory storing at least one instruction, and
And the processor executes the instructions stored in the memory to realize the business application rapid abstract modeling method.
In order to solve the above-mentioned problems, the present invention also provides a computer-readable storage medium having stored therein at least one instruction that is executed by a processor in an electronic device to implement the business application rapid abstraction modeling method described above.
The invention solves the problems in the background art by determining the service application field and acquiring corresponding service abstract application libraries, wherein the service abstract application libraries are provided with standardized parameter interfaces, and can be conveniently combined and invoked, thus greatly improving the modeling efficiency and flexibility, dividing complex service to be modeled into a plurality of relatively independent unit services, enabling each unit service to be clearer, easier to understand and process, setting the parameter interfaces for each unit service, ensuring smoother data interaction and collaborative work between each unit service, then constructing an effective service application group, realizing individuation and scenerization application of the service abstract application libraries, extracting proper public service application from the libraries according to the concrete requirements of the service to be modeled, developing special service application aiming at the special requirements, thus not only fully utilizing the prior service abstract application resources, avoiding repeated development, but also meeting the unique requirements of the service to be modeled, improving the service efficiency and flexibility, ensuring the integrity and accuracy of the service flow, enabling the specific service application to be more smoothly attached to each service application by setting the special parameter of the effective service application group, simultaneously optimizing the service application model and automatically executing the service to the situation under the situation of the service, automatically optimizing the service abstract application model and the situation, and automatically meeting the requirements of the service to realize the situation, and automatically optimizing the service application model by the requirements, the method and the system can rapidly and efficiently complete complex business processes, greatly improve the speed and efficiency of business processing, further, the scheme also introduces a large language model and a case library to conduct intelligent analysis to obtain executable confidence coefficient, which is helpful for judging the credibility and rationality of business execution results, finding potential problems and risks in time, and carrying out manual correction when the executable confidence coefficient is lower, which can ensure that the final business results meet actual requirements and business specifications, and finally, supplementing current cases to the case library, so that the content and diversity of the case library can be continuously enriched, various business scenes and processing methods can be reflected more comprehensively and accurately by the case library, and the accuracy and reliability of intelligent analysis are further improved. Therefore, the invention can improve the intelligent degree and flexibility of the business processing and improve the efficiency of the business processing.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
The embodiment of the application provides a rapid abstract modeling method for business application. The execution subject of the business application rapid abstract modeling method comprises, but is not limited to, at least one of a server, a terminal and the like which can be configured to execute the method provided by the embodiment of the application. In other words, the business application fast abstract modeling method may be performed by software or hardware installed in a terminal device or a server device, and the software may be a blockchain platform. The server side comprises, but is not limited to, a single server, a server cluster, a cloud server or a cloud server cluster and the like.
Referring to fig. 1, a flow chart of a method for fast abstract modeling of a business application according to an embodiment of the invention is shown. In this embodiment, the service application rapid abstract modeling method includes:
s1, receiving an application modeling instruction, and determining a service to be modeled and original service data based on the application modeling instruction.
It can be understood that the application modeling instruction refers to an instruction for modeling a specific service initiated by a person, and the service to be modeled refers to the specific service indicated in the application modeling instruction. The original business data refers to an initial data set which is provided by a user and related to business to be modeled, and comprises business operation records, transaction stream, file content or original information returned by an external system interface.
Illustratively, the raw business data is an account transaction statement (CSV format) and associated customer information database table of section B of third quarter of bank 2023.
S2, determining the service application field of the service to be modeled, and acquiring a service abstract application library based on the service application field, wherein the service abstract application library comprises a plurality of service abstract applications, and each service abstract application is provided with a parameter interface.
It is obvious that the service application field refers to an application field related to a service to be modeled, for example, if the service to be modeled is "accounting check of a company in a third quarter B of 2023", the service application field is "accounting check". The service abstract application library refers to an artificially constructed application database capable of executing all possible services in the service application field, wherein the service abstract application refers to a program or a plug-in capable of executing a certain service in the service application field. The parameter interface refers to a standardized protocol for data interaction between a service abstract application and an external system or a flow node, and comprises a parameter input interface and a parameter output interface, wherein the parameter input interface refers to an interface corresponding to data to be input before the service abstract application is executed, for example, a character string parameter of a file path is received, and the parameter output interface refers to an interface corresponding to data output after the service abstract application is executed.
In detail, the obtaining the service abstract application library based on the service application field includes:
constructing a business application process in the business application field, wherein the business application process comprises a plurality of business process nodes;
Sequentially extracting business process nodes from a plurality of business process nodes of a business application process, and defining node interfaces of the business process nodes to obtain programmable business nodes, wherein the node interfaces comprise an input parameter interface and an output parameter interface;
Performing application development on the programmable service node to obtain a service abstract application, wherein the application development comprises language programming;
and summarizing the business abstract application to obtain a business abstract application library.
It should be explained that the service application flow refers to a step flow for completing a task specified in a service application field, and the service application flow is formulated by related research personnel, and includes a plurality of service flow nodes, where a service flow node refers to a certain action required to be completed in the service application flow.
The business application process is exemplified by 'accounting check', wherein the business process node comprises a first node for uploading accounting files, an input interface of the first node is a file path, an output interface is analyzed accounting data, a second node for checking transaction compliance, the input interface of the second node is accounting data, the output interface is a compliance mark, a third node for generating check reports, the input interface of the third node is a compliance mark, and the output interface is a PDF report path.
Further, the node interface refers to a data transmission specification between the business process node and the upstream and downstream nodes or systems, and comprises an input parameter interface and an output parameter interface, wherein the input parameter interface refers to an inlet for receiving a processing result of a preceding node or external input data, and the output parameter interface refers to an outlet for transmitting the processing result of the current node to a subsequent node. The programmable service node refers to a service flow node obtained after definition of a node interface. The detailed process of defining the node interface of the business process node in the above steps is to define the format of input data (such as JSON Schema) and the structure of output data (such as database table fields) through an API. The aim of defining the node interface of the business process node is to realize standardization and modularization of the business process node, ensure seamless connection of data among different nodes and reduce development complexity.
It can be understood that the service abstract application refers to a programmable service node after application development, wherein application development refers to implementing data processing logic defined in a node interface through a programming language, and an implementation path of the application development is that writing scripts or services converts data of an input parameter interface into an expected result of an output parameter interface. The purpose of application development of the programmable service node is that the programmable service node defines a node interface, but an implementation path between an input parameter interface and an output parameter interface is not determined, that is, a specific action corresponding to the programmable service node is not set, so that the application development mode can be implemented by converting abstract interface definition into a specific service logic execution unit, for example, developing a "file analysis" plug-in unit, and realizing the function of extracting accounting data from a CSV file.
Importantly, the business abstraction application provided by the scheme can be independently developed and deployed on line.
S3, carrying out service division on the service to be modeled to obtain a unit service group, wherein the unit service group comprises a plurality of unit services, and each unit service is provided with a parameter interface.
It can be understood that the unit service refers to the divided service to be modeled, wherein the purpose of performing service division on the service to be modeled is to make the completed service into individual actions that can be performed independently because the service to be modeled is a complete service flow, and the flow includes a plurality of actions that need to be performed. The realization mode of the independent actions can be completed through the service abstract application established in the service abstract application library, thereby avoiding repeated establishment of service applications of different services, when the service application corresponding to a certain action is required to be called, the corresponding service application is only required to be extracted from the service abstract application library, and the complete service to be modeled can be modeled by mutually inserting the service application from which each action is extracted.
Further, the method for dividing the unit service group comprises the steps of firstly determining the realization step flow of the service to be modeled, and then extracting each individual action in the realization step flow, wherein the individual actions form the unit service group.
S4, constructing an effective service application group based on the service abstract application library and the unit service group, wherein the effective service application group comprises a plurality of effective service applications, the effective service applications comprise public service applications and proprietary service applications, and the public service applications come from the service abstract application library.
It should be explained that the effective service application group refers to a combination of service applications capable of executing each unit service in the unit service group, where the effective service applications include public service applications and private service applications, where the public service applications are service abstract applications that can be extracted from a service abstract application library, where the applications in the service abstract application library are constructed according to the flow of the corresponding field, but not all the implementation flows of the services in the field are identical to the flow of constructing the service abstract application library, and there may be a difference in partial actions between them, and the private service applications are applications that are artificially constructed and are differential actions between the unit service group and the service abstract application library. The business to be modeled is checking accounts of a company in March and marking accounts for reimbursement, wherein the business to be modeled comprises the accounts checking and marking specific accounts, and the application for realizing the marking of the specific accounts is special business application.
In detail, the constructing an effective service application group based on the service abstract application library and the unit service group includes:
sequentially extracting unit services from the unit service group, and determining flow data of the unit services, wherein the flow data comprises input data and output data;
judging whether a business abstract application matched with the unit business exists in a business abstract application library or not based on the streaming data;
if the business abstract application library contains the business abstract application matched with the unit business, extracting public business application from the business abstract application library;
if the service abstract application matched with the unit service does not exist in the service abstract application library, constructing a special service application of the unit service;
summarizing the public service application or the special service application to obtain an effective service application group.
It is understood that the streaming data refers to data involved in executing a unit service, and the streaming data includes input data and output data, where the input data refers to data that needs to be input before executing the unit service, and the output data refers to data obtained after executing the unit service, and the streaming data is matched with a parameter interface in a service abstraction application (i.e., the input data is the same as the parameter input interface, and the output data is the same as the parameter output interface).
The streaming data are exemplified by a "customer identification number" and a "transaction amount", and are matched if a parameter input interface of a certain business abstraction application requires a "character string type customer ID" and a "floating point transaction amount", and are not matched if the parameter interface requires an "integer type customer number".
Further, the public service application is a service abstract application matched with the unit service in the service abstract application library, the private service application is a service application for building a unit service without the matched service abstract application, and the method for building the private service application is the same as the method for building the service abstract application in the service abstract application library, and can refer to the content in the embodiment S2.
In detail, the determining whether the service abstraction application matched with the unit service exists in the service abstraction application library includes:
Sequentially extracting service abstract applications from a service abstract application library, confirming service node interfaces of the service abstract applications, and judging whether the unit service is matched with the service abstract applications or not based on streaming data and the service node interfaces;
if the unit service is confirmed to be not matched with the service abstract application, the service abstract application is marked as a non-matched application;
Summarizing the non-matching application to obtain a non-matching application set;
if the number of the non-matching applications in the non-matching application set is equal to the number of the business abstraction applications in the business abstraction application library, no business abstraction application matched with the unit business exists in the business abstraction application library;
Otherwise, there is a business abstraction application in the business abstraction application library that matches the unit business.
It is understood that the service node interface refers to a parameter interface of a service abstraction application. If the parameter input interface and the parameter output interface corresponding to the service node interface are identical to the input data and the output data in the streaming data respectively, the unit service can be considered to be matched with the service abstract application, and if the parameter input interface and the parameter output interface corresponding to the service node interface are not identical to the input data and the output data in the streaming data respectively, the unit service can be considered to be not matched with the service abstract application.
Further, when the number of non-matching applications in the non-matching application set is equal to the number of business abstraction applications in the business abstraction application library, it is indicated that the unit business is not matched with all business abstraction applications in the business abstraction application library, i.e. no business abstraction application matched with the unit business exists in the business abstraction application library, otherwise, business abstraction application matched with the unit business exists in the business abstraction application library.
S5, determining the application scene of the service to be modeled, and executing dedicated parameter setting on the effective service application group based on the application scene to obtain a target service application group.
The application scene is to be explained, and is a use scene of a business to be modeled, which is determined in real time by an operator, for example, a business to be modeled is an account check, and the account check is to check accounts of a company A in a department B in March, and the company A, the March and the department B are application scenes. The target service application group refers to an effective service application group after proprietary parameter setting.
Further, the execution-specific parameter setting refers to that different configuration parameters need to be set for the valid service application group for different application scenarios, wherein the configuration parameters refer to dynamic variables related to the application scenario, such as a time range (e.g. "2023 years Q3"), department codes (e.g. "B department"), and accounts types (e.g. "accounts receivable").
In detail, based on the application scenario, the method for performing dedicated parameter setting on the effective service application group to obtain the target service application group includes:
Sequentially extracting effective service applications from the effective service application group, and confirming configuration options in the effective service applications;
Acquiring exclusive configuration parameters of an application scene, and updating configuration options in the effective service application by utilizing the exclusive configuration parameters to obtain a target service application;
and summarizing the target business application to obtain a target business application group.
It can be appreciated that the configuration options refer to adjustable variables or conditions in the business abstraction application, such as an accounting check range, and the dedicated configuration parameters refer to numerical parameters involved in an application scenario.
For example, if a certain effective business application is an application related to accounting check, there will be a check range in the effective business application, where the check range is a configuration option, and an application scenario is "accounting check of B department of a company in march", where the check range should be march, that is, the check range in the effective business application needs to be set as march.
S6, arranging a service scene application model based on the target service application group.
The service scene application model refers to a model obtained by combining and arranging all target service applications in the target service application group. Because the target service application contained in the target service application group only represents a single action, the target service application group is arranged, namely actions corresponding to the applications are combined, so that the combined obtained model can complete the service to be modeled.
In detail, the arranging the service scene application model based on the target service application group includes:
determining a target service flow of the service to be modeled, wherein the target service flow comprises a plurality of target service nodes, and each target service node corresponds to one target service application in a target service application group;
Determining an application connection relation of each target service application in the target service application group based on a target service flow to obtain an application connection relation group, wherein the application connection relation group comprises a series relation and a parallel relation;
and integrating each target service application in the target service application group according to the application connection relation group to obtain a service scene application model, wherein the integration mode comprises API interface calling.
It can be understood that the target service flow refers to an action flow for executing the service to be modeled, and the target service node refers to a certain action in the action flow. The application connection relationship refers to a connection relationship of the target service application in a service scene application model, wherein the series connection relationship refers to a time sequence of the two target service applications when the two target service applications are executed, and the parallel connection relationship refers to the same execution sequence of the two target service applications when the two target service applications are executed.
The method comprises the steps of uploading a file, checking a transaction and adjusting accounts, wherein the order of executing the operations is that the file, the checking the transaction and the account are all in a series connection, and each target business application in a target business application group is integrated into a series connection according to an application connection group, namely the target business application corresponding to the uploading file, the checking the transaction and the account is connected in series, and a business scene application model is obtained after the series connection is completed, wherein the series connection mode comprises API interface calling.
And S7, carrying out service automation processing according to the original service data and the service scene application model to obtain a service execution result.
It can be understood that the service automation processing refers to inputting original service data into a service scene application model, and sequentially processing the original service data by a target service application group connected in the service scene application model, and obtaining a service execution result after processing is completed.
And S8, performing intelligent analysis on the service execution result according to a pre-constructed case library to obtain executable confidence, wherein the intelligent analysis refers to analysis by using a large language model.
It is understood that the case library refers to a collection of a large number of cases obtained during a past period of time that were artificially constructed. The executable confidence coefficient refers to a numerical value of the credibility of the service execution result, the larger the executable confidence coefficient is, the higher the credibility of the service execution result is, when the executable confidence coefficient is larger than a preset confidence coefficient threshold value, the service processing is carried out according to the service execution result, otherwise, the intervention is needed manually, and the service execution result is modified.
In detail, the intelligent analysis is performed on the service execution result according to the pre-constructed case library to obtain the executable confidence coefficient, which comprises the following steps:
Searching a case set in the same field in a case base based on the service application field, wherein the case set in the same field comprises a plurality of cases in the same field, and the cases in the same field are cases in the service application field;
sequentially extracting cases in the same field in the case set in the same field, determining a case service application group of the cases in the same field, and acquiring case configuration parameters corresponding to each case service application in the case service application group to obtain a case configuration parameter group;
Determining case input data of the same-field case;
Based on case input data and case configuration parameter sets, carrying out similarity analysis on cases in the same field and the business to be modeled to obtain business similarity;
if the service similarity is greater than a preset similarity threshold, marking the service similarity as effective service similarity;
summarizing the effective service similarity to obtain an effective service similarity group;
If the effective service similarity group is a preset empty set, marking the executable confidence coefficient as 0;
If the effective service similarity group is not an empty set, extracting a similar case group corresponding to the effective service similarity group from the case set in the same field;
And evaluating the service execution result by using the similar case group to obtain the executable confidence coefficient.
It can be appreciated that the same domain case set refers to a set of cases in the case library that are the same as the business domain and the business application domain. The case service application group refers to a target service application group in the case of the same field. The case configuration parameter set refers to a combination of dedicated configuration parameters of each case service application in the case service application set. The case input data refers to data input when the case in the same field is executed in the past, and the data corresponds to the original business data. The business similarity refers to a numerical value of similarity degree between a case in the same field and a business to be modeled, and the greater the business similarity is, the greater the similarity degree between the case in the same field and the business to be modeled is, wherein similarity analysis on the case in the same field and the business to be modeled refers to calculating Euclidean geometric distances between case input data and case configuration parameter sets and business input data and proprietary configuration parameter sets, and taking the Euclidean geometric distances as the business similarity. The similarity threshold refers to a constant set by a person. The similar case group refers to a combination of cases in the same domain corresponding to the effective service similarity group.
In detail, the evaluating the service execution result by using the similar case group to obtain the executable confidence comprises:
sequentially extracting similar cases from the similar case group, and obtaining case execution results of the similar cases;
performing similarity evaluation on the case execution result and the service execution result by using the large language model to obtain action similarity;
Summarizing the action similarity to obtain an action similarity group, and calculating executable confidence based on the action similarity group and the effective service similarity group, wherein the executable confidence is expressed as:
Wherein, theIndicating the degree of confidence in the executability,Represents the number of action similarities in the action similarity group or the number of effective service similarities in the effective service similarity group,AndRespectively represent the first of the valid service similarity groupsIndividual effective business similarity and the firstA degree of similarity of the effective traffic is determined,Representing the first of the group of action similaritiesAnd each action similarity.
It should be explained that the case execution result refers to the service execution result of the similar case. The large language model refers to a pre-trained model (such as GPT-4) based on a transducer architecture. The action similarity refers to a numerical value of the similarity degree between the case execution result and the service execution result, and the greater the action similarity is, the higher the similarity degree between the case execution result and the service execution result is. The step of carrying out similarity evaluation on the case execution result and the service execution result by utilizing the pre-acquired large language model is to convert the case execution result and the current service execution result into natural language description, then extract semantic vectors of the case execution result and the current service execution result by utilizing the large language model, and finally calculate cosine similarity of the semantic vectors between the case execution result and the current service execution result as action similarity.
And S9, if the executable confidence coefficient is not greater than a confidence coefficient threshold value, manually correcting the service execution result to obtain a service correction result, and marking the service correction result as an executable result.
It can be understood that the service correction result refers to the service execution result after manual correction.
In detail, the manually correcting the service execution result to obtain a service correction result includes:
confirming an abnormal similar case group of the executable confidence in a case library, and obtaining an abnormal execution result group corresponding to the abnormal similar case group;
inputting the abnormal execution result set and the business execution result into a large language model to obtain a risk point set, wherein the risk point set comprises a plurality of risk points;
and generating a manual task list based on the risk point group, and manually correcting by using the manual task list to obtain a service correction result.
It can be appreciated that the abnormal similar case group refers to a similar case group when the executable confidence is not greater than the confidence threshold. The abnormal execution result group refers to the combination of the business execution results corresponding to the abnormal similar case group. The risk point refers to a potential problem that errors may be caused in a service execution result, for example, accounting adjustment amount exceeds 20% of a historical average value. The large language model can identify an abnormal mode and generate a risk prompt of natural language description, for example, the condition that the adjustment amount is abnormal is detected, so that the difference between an abnormal execution result set and a business execution result can be identified through the large language model, and a risk point set is obtained. The manual task list comprises service execution results of the risk point group. The manual correction refers to the modification of the risk point group in the manual task list by a professional.
And S10, if the executable confidence coefficient is larger than a confidence coefficient threshold value, marking the service execution result as an executable result.
It can be appreciated that if the executable confidence is greater than the confidence threshold, the business execution result can be directly executed without human intervention.
And S11, carrying out case data combination based on the original service data and the executable result to obtain a current case, supplementing the current case to the case library, and completing the quick abstract modeling of the service application.
It should be explained that the current case represents the service to be modeled after the current processing, and the rotational speed configuration parameter set composed of the original service data, the executable result and the dedicated configuration parameter can be used as the case. The current cases are supplemented to the case base, so that the case base can be further enriched, and the accuracy of subsequent business processing is improved.
Furthermore, the laboratory environment is provided for the user by using the totally independent sandbox space, wherein the laboratory environment allows the user to carry out temporary experiments on the arranged application model so as to verify whether the arrangement effect and parameters are correct, and the sandbox data space is totally isolated from the production line and can be cleaned at any time. Meanwhile, the scheme also comprises an independent log tracking module, and each step of execution log can be recorded in the process of automatically executing the business scene application model and is used for users to inquire.
The invention solves the problems in the background art by determining the service application field and acquiring corresponding service abstract application libraries, wherein the service abstract application libraries are provided with standardized parameter interfaces, and can be conveniently combined and invoked, thus greatly improving the modeling efficiency and flexibility, dividing complex service to be modeled into a plurality of relatively independent unit services, enabling each unit service to be clearer, easier to understand and process, setting the parameter interfaces for each unit service, ensuring smoother data interaction and collaborative work between each unit service, then constructing an effective service application group, realizing individuation and scenerization application of the service abstract application libraries, extracting proper public service application from the libraries according to the concrete requirements of the service to be modeled, developing special service application aiming at the special requirements, thus not only fully utilizing the prior service abstract application resources, avoiding repeated development, but also meeting the unique requirements of the service to be modeled, improving the service efficiency and flexibility, ensuring the integrity and accuracy of the service flow, enabling the specific service application to be more smoothly attached to each service application by setting the special parameter of the effective service application group, simultaneously optimizing the service application model and automatically executing the service to the situation under the situation of the service, automatically optimizing the service abstract application model and the situation, and automatically meeting the requirements of the service to realize the situation, and automatically optimizing the service application model by the requirements, the method and the system can rapidly and efficiently complete complex business processes, greatly improve the speed and efficiency of business processing, further, the scheme also introduces a large language model and a case library to conduct intelligent analysis to obtain executable confidence coefficient, which is helpful for judging the credibility and rationality of business execution results, finding potential problems and risks in time, and carrying out manual correction when the executable confidence coefficient is lower, which can ensure that the final business results meet actual requirements and business specifications, and finally, supplementing current cases to the case library, so that the content and diversity of the case library can be continuously enriched, various business scenes and processing methods can be reflected more comprehensively and accurately by the case library, and the accuracy and reliability of intelligent analysis are further improved. Therefore, the invention can improve the intelligent degree and flexibility of the business processing and improve the efficiency of the business processing.
Fig. 2 is a functional block diagram of a service application rapid abstract modeling apparatus according to an embodiment of the invention.
The business application rapid abstract modeling apparatus 100 of the present invention may be installed in an electronic device. The business application rapid abstract modeling apparatus 100 may include a business application construction module 101, a proprietary parameter setting module 102, a language model analysis module 103, and an execution result modification module 104 according to the implemented functions. The module of the invention, which may also be referred to as a unit, refers to a series of computer program segments, which are stored in the memory of the electronic device, capable of being executed by the processor of the electronic device and of performing a fixed function.
The service application construction module 101 is configured to receive an application modeling instruction, determine a service to be modeled and original service data based on the application modeling instruction, determine a service application field of the service to be modeled, and obtain a service abstract application library based on the service application field, where the service abstract application library includes a plurality of service abstract applications, and each service abstract application is provided with a parameter interface;
The dedicated parameter setting module 102 is configured to divide the service to be modeled to obtain a unit service group, where the unit service group includes a plurality of unit services, each unit service has a parameter interface, and an effective service application group is constructed based on a service abstract application library and the unit service group, where the effective service application group includes a plurality of effective service applications, and the effective service applications include public service applications and proprietary service applications, and the public service applications come from the service abstract application library, determine an application scenario of the service to be modeled, and execute dedicated parameter setting on the effective service application group based on the application scenario, to obtain a target service application group;
the language model analysis module 103 is configured to arrange a service scene application model based on a target service application group, perform service automation processing according to original service data and the service scene application model to obtain a service execution result, and perform intelligent analysis on the service execution result according to a pre-constructed case library to obtain an executable confidence coefficient, where the intelligent analysis refers to analysis by using a large language model;
The execution result correction module 104 is configured to, if the executable confidence coefficient is not greater than a confidence coefficient threshold, manually correct the service execution result to obtain a service correction result, and record the service correction result as an executable result, if the executable confidence coefficient is greater than the confidence coefficient threshold, record the service execution result as an executable result, and perform case data merging based on the original service data and the executable result to obtain a current case, and supplement the current case to the case library.
In detail, the modules in the service application rapid abstract modeling apparatus 100 in the embodiment of the present invention use the same technical means as the service application rapid abstract modeling method described in fig. 1, and can generate the same technical effects, which are not described herein.
Fig. 3 is a schematic structural diagram of an electronic device for implementing a business application rapid abstract modeling method according to an embodiment of the present invention.
The electronic device 1 may comprise a processor 10, a memory 11 and a bus 12, and may further comprise a computer program stored in the memory 11 and executable on the processor 10, such as a business application fast abstract modeling method program.
The memory 11 includes at least one type of readable storage medium, including flash memory, a mobile hard disk, a multimedia card, a card memory (e.g., SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. The memory 11 may in some embodiments be an internal storage unit of the electronic device 1, such as a removable hard disk of the electronic device 1. The memory 11 may in other embodiments also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a smart memory card (SMART MEDIA CARD, SMC), a Secure Digital (SD) card, a flash memory card (FLASH CARD) or the like, which are provided on the electronic device 1. Further, the memory 11 further comprises an internal storage unit of the electronic device 1, and also comprises an external storage device. The memory 11 may be used not only for storing application software installed in the electronic device 1 and various types of data, such as codes of a business application rapid abstraction modeling method program, but also for temporarily storing data that has been output or is to be output.
The processor 10 may be comprised of integrated circuits in some embodiments, for example, a single packaged integrated circuit, or may be comprised of multiple integrated circuits packaged with the same or different functions, including one or more central processing units (Central Processing unit, CPU), microprocessors, digital processing chips, graphics processors, combinations of various control chips, and the like. The processor 10 is a Control Unit (Control Unit) of the electronic device, connects various components of the entire electronic device using various interfaces and lines, executes or executes programs or modules (e.g., business application rapid abstraction modeling method programs, etc.) stored in the memory 11, and invokes data stored in the memory 11 to perform various functions of the electronic device 1 and process data.
The bus 12 may be a peripheral component interconnect standard (PERIPHERAL COMPONENT INTERCONNECT, PCI) bus, or an extended industry standard architecture (extended industry standard architecture, EISA) bus, among others. The bus 12 may be divided into an address bus, a data bus, a control bus, etc. The bus 12 is arranged to enable a connection communication between the memory 11 and at least one processor 10 etc.
Fig. 3 shows only an electronic device with components, it being understood by a person skilled in the art that the structure shown in fig. 3 does not constitute a limitation of the electronic device 1, and may comprise fewer or more components than shown, or may combine certain components, or may be arranged in different components.
For example, although not shown, the electronic device 1 may further include a power source (such as a battery) for supplying power to each component, and preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that functions of charge management, discharge management, power consumption management, and the like are implemented through the power management device. The power supply may also include one or more of any of a direct current or alternating current power supply, recharging device, power failure detection circuit, power converter or inverter, power status indicator, etc. The electronic device 1 may further include various sensors, bluetooth modules, wi-Fi modules, etc., which will not be described herein.
Further, the electronic device 1 may also comprise a network interface, optionally the network interface may comprise a wired interface and/or a wireless interface (e.g. WI-FI interface, bluetooth interface, etc.), typically used for establishing a communication connection between the electronic device 1 and other electronic devices.
The electronic device 1 may optionally further comprise a user interface, which may be a Display, an input unit, such as a Keyboard (Keyboard), or a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like. The display may also be referred to as a display screen or display unit, as appropriate, for displaying information processed in the electronic device 1 and for displaying a visual user interface.
The business application fast abstract modeling method program stored in the memory 11 in the electronic device 1 is a combination of a plurality of instructions, which when run in the processor 10, can realize:
Receiving an application modeling instruction, and determining a service to be modeled and original service data based on the application modeling instruction;
Determining the service application field of the service to be modeled, and acquiring a service abstract application library based on the service application field, wherein the service abstract application library comprises a plurality of service abstract applications, and each service abstract application is provided with a parameter interface;
Carrying out service division on the service to be modeled to obtain a unit service group, wherein the unit service group comprises a plurality of unit services, and each unit service is provided with a parameter interface;
Constructing an effective service application group based on a service abstract application library and a unit service group, wherein the effective service application group comprises a plurality of effective service applications, the effective service applications comprise public service applications and proprietary service applications, and the public service applications come from the service abstract application library;
determining the application scene of the service to be modeled, and executing dedicated parameter setting on the effective service application group based on the application scene to obtain a target service application group;
arranging a service scene application model based on the target service application group;
According to the original service data and the service scene application model, carrying out service automation processing to obtain a service execution result;
performing intelligent analysis on the service execution result according to a pre-constructed case library to obtain executable confidence, wherein the intelligent analysis refers to analysis by using a large language model;
If the executable confidence coefficient is not greater than the confidence coefficient threshold value, manually correcting the service execution result to obtain a service correction result, and marking the service correction result as an executable result;
if the executable confidence is greater than a confidence threshold, marking the service execution result as an executable result;
And carrying out case data combination based on the original service data and the executable result to obtain a current case, supplementing the current case to the case library, and completing the quick abstract modeling of the service application.
Specifically, the specific implementation method of the above instructions by the processor 10 may refer to descriptions of related steps in the corresponding embodiments of fig. 1 to 3, which are not repeated herein.
Further, the modules/units integrated in the electronic device 1 may be stored in a computer readable storage medium if implemented in the form of software functional units and sold or used as separate products. The computer readable storage medium may be volatile or nonvolatile. For example, the computer readable medium may include any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM).
The present invention also provides a computer readable storage medium storing a computer program which, when executed by a processor of an electronic device, can implement:
Receiving an application modeling instruction, and determining a service to be modeled and original service data based on the application modeling instruction;
Determining the service application field of the service to be modeled, and acquiring a service abstract application library based on the service application field, wherein the service abstract application library comprises a plurality of service abstract applications, and each service abstract application is provided with a parameter interface;
Carrying out service division on the service to be modeled to obtain a unit service group, wherein the unit service group comprises a plurality of unit services, and each unit service is provided with a parameter interface;
Constructing an effective service application group based on a service abstract application library and a unit service group, wherein the effective service application group comprises a plurality of effective service applications, the effective service applications comprise public service applications and proprietary service applications, and the public service applications come from the service abstract application library;
determining the application scene of the service to be modeled, and executing dedicated parameter setting on the effective service application group based on the application scene to obtain a target service application group;
arranging a service scene application model based on the target service application group;
According to the original service data and the service scene application model, carrying out service automation processing to obtain a service execution result;
performing intelligent analysis on the service execution result according to a pre-constructed case library to obtain executable confidence, wherein the intelligent analysis refers to analysis by using a large language model;
If the executable confidence coefficient is not greater than the confidence coefficient threshold value, manually correcting the service execution result to obtain a service correction result, and marking the service correction result as an executable result;
if the executable confidence is greater than a confidence threshold, marking the service execution result as an executable result;
And carrying out case data combination based on the original service data and the executable result to obtain a current case, supplementing the current case to the case library, and completing the quick abstract modeling of the service application.
In the several embodiments provided in the present invention, it should be understood that the disclosed apparatus, device and method may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, and there may be additional divisions of a practical implementation.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units can be realized in a form of hardware or a form of hardware and a form of software functional modules.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof.
Finally, it should be noted that the above-mentioned embodiments are merely for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications and equivalents may be made to the technical solution of the present invention without departing from the spirit and scope of the technical solution of the present invention.