A business process automation processing system based on browser scripts

Abstract

The present invention discloses a business process automation processing system based on browser scripts, which belongs to the technical field of business process management and includes a platform end and a user end; the platform end includes a script module and a script library; the user end includes a business analysis module, an automatic processing module, and a calibration module; the script module is used to perform script demand analysis, obtain corresponding script demands, set corresponding browser scripts according to the script demands, and store the browser scripts in the script library after aggregating them; the script library is used to store browser scripts; the business analysis module is used to match browser scripts that meet user demands from the script library of the platform end, verify and adjust the browser scripts, obtain target scripts, and configure the target scripts accordingly; the automatic processing module is used to perform business process automation processing according to the target script; the calibration module is used to calibrate the business process automation processing process, obtain calibration results, and perform corresponding processing according to the calibration results.

Description

Technical Field

[0001] The present invention belongs to the technical field of business process management, and in particular is a business process automation processing system based on browser scripts. Background Art

[0002] With the booming internet economy, small and medium-sized enterprises (SMEs) and individual users (such as e-commerce sellers, freelancers, and small service organizations) are facing an urgent need for digital transformation. For example, in e-commerce operations, users must frequently capture competitor prices, list products in batches, and automatically process orders to maintain competitiveness. In financial services, cross-platform financial product yield monitoring and automated report generation are key to improving efficiency. However, these users generally face significant pain points: weak technical skills and limited development resources. For one thing, SMEs often lack dedicated IT teams. For another, existing automation solutions (such as Selenium-based or RPA-based frameworks) generally require programming skills or complex configuration capabilities, making it difficult for non-technical personnel to independently write and maintain scripts. Furthermore, dynamic updates to web page structures (such as element ID changes and the loading of multiple nested iframes) often lead to interruptions in automated processes, which users, lacking the technical background to independently fix, further exacerbates maintenance costs.

[0003] Based on this, in order to solve the above problems, the present invention provides a business process automation processing system based on browser scripts. Summary of the Invention

[0004] In order to solve the problems existing in the above solutions, the present invention provides a business process automation processing system based on browser scripts.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A business process automation processing system based on browser scripts, including a platform side and a user side;

[0007] The platform end includes a script module and a script library;

[0008] The script module is used to perform script requirement analysis, obtain corresponding script requirements, set corresponding browser scripts according to the script requirements, and aggregate the browser scripts and store them in a script library.

[0009] The script library is used to store browser scripts.

[0010] Furthermore, script requirements analysis is performed, including:

[0011] Obtain various business processing requirements for business process automation; identify browser information related to the business processing requirements;

[0012] Decomposing the business processing requirements into steps to obtain atomic operation data of the business processing requirements; combining the atomic operation data with corresponding browser information to obtain several groups of analysis combinations;

[0013] Classify each analysis combination to obtain the corresponding requirement classification, and set the script requirements according to the requirement classification.

[0014] Furthermore, each analysis combination is classified into:

[0015] Step SA1: Establish a combined evaluation model;

[0016] Step SA2: arbitrarily select two analysis combinations to form input data and input them into the combined evaluation model for analysis to obtain corresponding combined evaluation results; the combined evaluation results include whether they meet the classification requirements or do not meet the classification requirements;

[0017] Step SA3: When the combined evaluation result meets the classification requirements, the corresponding two analysis combinations are classified into one category, marked as the intermediate category, and proceed to step SA4;

[0018] When the combined evaluation result does not meet the classification requirements, the two analysis combinations are marked as corresponding different class labels, and the process returns to step SA2;

[0019] Step SA4: Select one of the remaining analysis combinations to form the input data corresponding to the intermediate class and input it into the combined evaluation model for analysis to obtain the corresponding combined evaluation result;

[0020] When the combined evaluation result is in compliance with the classification requirements, the corresponding analysis combination is merged into the intermediate class to form a new intermediate class;

[0021] When the combined evaluation result does not meet the classification requirements, the corresponding different class markings are performed;

[0022] Step SA5: looping step SA4 until there is no combined evaluation result that meets the classification requirements, marking the intermediate class as a demand classification;

[0023] Step SA6: Loop steps SA2 to SA5 until all analysis combinations are classified.

[0024] Furthermore, the expression of the combined evaluation model is: ;

[0025] Where: s is the number of analysis combinations that need to be evaluated; the output data is the combined evaluation value HP(s), which is 1 or 0;

[0026] When the combined evaluation value is 1, the combined evaluation result meets the classification requirements;

[0027] When the combined evaluation value is 0, the combined evaluation result does not meet the classification requirements.

[0028] The user terminal includes a business analysis module and an automatic processing module;

[0029] The business analysis module is used to match browser scripts that meet user needs from the script library on the platform side, verify and adjust the browser scripts, obtain the target script, and configure the target script accordingly.

[0030] Furthermore, the browser scripts that meet the user's needs are matched from the script library on the platform side, including:

[0031] Obtain user needs, extract features of user needs, obtain corresponding matching feature groups, input the matching feature groups into the script library on the platform side for matching, and obtain browser scripts that meet user needs.

[0032] Furthermore, the browser scripts are verified and adjusted, including:

[0033] Step SC1: Generate corresponding interactive simulation data according to the browser script and user needs;

[0034] Step SC2: displaying the interactive simulation data to the user;

[0035] When the user is satisfied with the interactive simulation data, the adjustment is completed and the analysis ends;

[0036] When the user is not satisfied with the interaction simulation data, the interaction simulation data is marked accordingly to obtain interaction marking data; and new interaction simulation data is generated according to the interaction marking data;

[0037] Step SC3: Loop step SC2.

[0038] Furthermore, when the number of cycles of step SA2 reaches the threshold value X1, a prompt is given to the corresponding staff of the platform, who then assists the user in adjusting the browser script.

[0039] The automatic processing module is used to perform business process automation processing according to the target script.

[0040] Furthermore, the user terminal also includes a calibration module, which is used to calibrate the business process automation processing process, obtain corresponding calibration results, and perform corresponding processing according to the calibration results.

[0041] Furthermore, the calibration of the business process automation process is real-time calibration and / or pre-calibration.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] Through the mutual cooperation between various modules, the present invention systematically solves the three core pain points faced by small and medium-sized enterprises and individual users in digital transformation: high technical barriers, high maintenance costs, and poor cross-scenario adaptability; through a unified task description language and dynamic adapter layer, operations on different platforms are converted into standardized instruction sets, realizing the reuse of single scripts on multiple platforms. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0045] Figure 1 This is a principle block diagram of the present invention. DETAILED DESCRIPTION

[0046] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0047] like Figure 1 As shown, a business process automation processing system based on browser scripts includes a platform end and a user end;

[0048] The platform end is connected to the user end of each user in a communication manner.

[0049] The platform end includes a script module and a script library;

[0050] The script module is used to perform script requirement analysis and obtain corresponding script requirements. Platform staff sets corresponding browser scripts according to the script requirements and aggregates and stores the browser scripts in a script library.

[0051] The script library is used to store browser scripts, including browser scripts obtained through other means.

[0052] In one embodiment, performing script requirement analysis includes:

[0053] Determine the various business processing requirements for automated business processes based on existing methods such as big data, or analyze them based on relevant historical data; identify browser information related to the corresponding business processing requirements, that is, which browsers the business processing requirements will be encountered on, and determine this information based on various existing browsers; browser information includes browser type, version, and other relevant information that affects script settings;

[0054] Various business processing requirements are broken down into steps to obtain corresponding atomic operation data; the atomic operation data is combined with the corresponding browser information to obtain several groups of analysis combinations, that is, the atomic operation data is combined with each corresponding browser information one by one. For example, if the atomic operation data A and the browser information corresponding to the business processing requirement is 1 and 2, then A1 and A2 form different analysis combinations.

[0055] Classify each analysis combination to form a corresponding demand classification, and set the script requirements according to the demand classification, that is, one demand classification corresponds to one script requirement. Subsequently, set the corresponding browser script for the script requirement. Generally, one browser script corresponds to one script requirement, which is applicable to various analysis combinations corresponding to the script requirement.

[0056] Decompose various business processing requirements into steps based on existing step decomposition methods. For example, decompose a business processing requirement into opening a webpage → entering an account → clicking login → verifying the jump, and record the input / output data and dependent conditions of each step, such as network status and the existence of page elements; and integrate them into atomic operation data. Other examples include:

[0057] Linear process decomposition (applicable to simple businesses without branches):

[0058] step:

[0059] Determine the starting and ending points: clarify the business trigger conditions (such as scheduled / manual) and expected results (such as generating a report).

[0060] Segment by operation sequence: break down the process into linear steps such as "open web page → enter data → click button → verify result".

[0061] Label dependencies: record the input source (such as the output of the previous step, external files) and output target of each step.

[0062] Conditional branch disassembly (applicable to businesses involving logical judgment):

[0063] step:

[0064] Identify decision points: Find the conditions that need to be judged in the process (such as "price < 100 yuan" and "inventory > 0").

[0065] Draw a branch tree: Use decision points as nodes to construct an "if-then-else" branch structure.

[0066] Merge common paths: Merge the same operations after branching (such as submitting an order) into a common module.

[0067] Iterative decomposition (applicable to repetitive business):

[0068] step:

[0069] Locate loop bodies: Identify operations that need to be repeated (such as fetching multiple pages of data or uploading files in batches).

[0070] Determine the termination condition: set the loop end rules (such as "crawl to page 10" and "file list is empty").

[0071] Abstract variables: Use variables to replace dynamic values ​​in loops.

[0072] In one embodiment, each analysis combination is classified. The classification standard is that a browser script can be set to apply to the processing of various analysis combinations within the category. It can also include subsequent loading of components as needed. The specific standard is set according to the needs of the platform to control the number of pre-set browser scripts; according to the classification standard, the analysis combinations are classified based on various current classification methods, clustering algorithms, etc.

[0073] In one embodiment, each analysis combination is classified, including:

[0074] Step SA1: Establish a combined evaluation model. The combined evaluation model is used to evaluate whether the same browser script can be applied to the corresponding analysis combinations. The evaluation can be made based on the historical data of various existing browser scripts.

[0075] Step SA2: arbitrarily select two analysis combinations to form input data and input them into the combined evaluation model for analysis to obtain corresponding combined evaluation results; the combined evaluation results include whether they meet the classification requirements or do not meet the classification requirements;

[0076] Step SA3: When the combined evaluation result meets the classification requirements, the corresponding two analysis combinations are classified into one category, marked as the intermediate category, and proceed to step SA4;

[0077] When the combined evaluation result does not meet the classification requirements, the two analysis combinations are marked as corresponding different class labels, and no subsequent similar analysis is performed; return to step SA2;

[0078] Step SA4: Select any one of the remaining analysis combinations that forms the input data corresponding to the intermediate class and input it into the combined evaluation model for analysis, excluding analysis combinations that have different class labels from the intermediate class, and obtain the corresponding combined evaluation result; this means that the analysis combination cannot meet the classification criteria with all analysis combinations in the intermediate class;

[0079] When the combined evaluation result is in compliance with the classification requirements, the corresponding analysis combination is merged into the intermediate class to form a new intermediate class;

[0080] When the combined evaluation result does not meet the classification requirements, the corresponding different class markings are performed;

[0081] Step SA5: Loop step SA4 until there is no combined evaluation result that meets the classification requirements, and mark the current intermediate class as the demand classification;

[0082] Step SA6: loop through steps SA2 to SA5 until all analysis combinations are classified, and the remaining analysis combinations with different class labels are treated as independent demand classifications.

[0083] In one embodiment, the combined evaluation model is established based on existing methods, such as machine learning, deep learning algorithms, etc., and a training set is established using corresponding historical data for training. The training set includes input data and output data. The input data is multiple analysis combinations for evaluation, and the output data is whether each analysis combination meets the requirements with each other, that is, the combined evaluation result; analysis is performed through the combined evaluation model after successful training.

[0084] In one embodiment, the expression of the combined evaluation model is: ;

[0085] Where: s is the number of analysis combinations that need to be evaluated; the output data is the combined evaluation value HP(s), which is 1 or 0; training is performed using the set training set;

[0086] When the combined evaluation value is 1, the combined evaluation result meets the classification requirements;

[0087] When the combined evaluation value is 0, the combined evaluation result does not meet the classification requirements.

[0088] The user terminal includes a business analysis module and an automatic processing module;

[0089] The business analysis module is used to match a browser script that meets user needs from the script library on the platform side, verify and adjust the browser script, obtain the target script, and configure the target script accordingly.

[0090] In one embodiment, browser scripts that meet user needs are matched from the script library on the platform side. The corresponding matching can be performed based on existing methods. For example, the platform fully communicates with the user to clarify their business scenarios, functional requirements (such as automated task types, data processing logic), performance requirements (such as response time, concurrency) and security compliance requirements (such as data encryption, permission control), and integrate them into user needs. Candidate scripts are quickly located in the script library based on the demand keywords (such as "user login automation" and "data cleaning"), label classification (such as industry, technology stack) or metadata (such as version, dependency library) in the user needs; static analysis (such as code structure, function call) or dynamic testing (such as simulation operation) is used to verify whether the script covers the core requirements, and scripts with compatibility issues (such as dependency conflicts) are excluded.

[0091] In one embodiment, matching a browser script that meets user needs from a script library on the platform side includes:

[0092] Obtain user needs, extract features of user needs, and obtain corresponding matching feature groups, that is, extract features according to the analysis combination, input the matching feature groups into the script library on the platform side for matching, and obtain corresponding browser scripts.

[0093] In one embodiment, when there is no browser script that meets the user's needs in the script library, an early warning will be issued to the platform staff, and manual processing will be used to provide services to the user, such as informing that the service scope is out of reach and manually creating a new browser script that meets the user's requirements. At this time, the browser script does not need to undergo subsequent verification and adjustment, and can be directly marked as the target script and stored in the script library.

[0094] In one embodiment, the verification adjustment of the browser script may be performed based on an existing method.

[0095] In one embodiment, performing validation adjustments on the browser script includes:

[0096] Step SC1: Generate corresponding interactive simulation data based on the browser script and user requirements. That is, according to the user requirements such as application scenarios and the browser script, generate a simulated browser script operation process data. The user can mark the operation process data, marking the corresponding parts that do not meet the requirements and the operations that are expected to be achieved;

[0097] Step SC2: displaying the interactive simulation data to the user;

[0098] When the user is satisfied with the interactive simulation data, the adjustment is completed and the analysis ends;

[0099] When the user is dissatisfied with the interaction simulation data, the interaction simulation data is marked accordingly to obtain interaction marking data; and new interaction simulation data that overcomes the corresponding marking is generated based on the interaction marking data;

[0100] Step SC3: Loop step SC2.

[0101] In one embodiment, when the number of cycles of step SA2 reaches the threshold value X1, the platform will manually prompt the corresponding work, and then quickly assist the user to complete the adjustment; the threshold value X1 is set by the platform according to actual conditions, such as 3, 5, etc.

[0102] In one embodiment, corresponding interactive simulation data is generated according to the browser script and user needs. It is generated based on existing artificial intelligence technology, and can also be generated using other existing technologies. For example, the browser script is directly loaded and applied on the platform, and an operation video of the corresponding script is recorded. The operation video is markable to form interactive simulation data; users can mark problematic operations (such as wrong clicks, data errors) and correct operations (such as expected behaviors) on the video timeline; and can add text annotations, screenshot highlights, classification labels (such as "UI errors", "logical defects"), etc.

[0103] In one implementation, new interaction simulation data that overcomes the corresponding marks is generated based on the interaction mark data. The generation is based on existing technologies and mainly has two directions. One is to "simulate" the correction operation directly on the video (such as erasing the wrong click and inserting the correct operation) to generate a corrected version of the video and obtain the interaction simulation data; the other is to extract the problem operations (such as wrong clicks, data errors) and correct operations (such as expected behaviors) in the video, automatically modify the original browser script according to the annotations, avoid the problem operations or replace them with correct operations, run the corrected script, and obtain the interaction simulation data.

[0104] For example, based on the marked timestamps, frames of problematic operations are extracted from the video (such as the moment when the wrong button is clicked); image processing techniques (such as OpenCV's inpaint) are used to remove erroneous elements (such as highlighting the wrong button); and animations of correct operations are superimposed at the same time point (such as simulating clicking an alternative button).

[0105] Record a short video of the correct operation (such as clicking an alternative button). Use a video editing tool (such as FFmpeg) to insert the short video into the original video at the problematic time.

[0106] In one embodiment, the target script is configured accordingly. This is based on existing methods, such as transferring the script file directly to the user's server. This is suitable for scenarios sensitive to data privacy. The script is deployed via a SaaS platform or container service (such as Kubernetes) to achieve elastic scaling and remote management. Deploying the script to IoT devices or edge nodes reduces latency and cloud load. Configuration is directly assisted by platform staff.

[0107] The automatic processing module is used to perform corresponding business process automation processing according to the target script.

[0108] In one embodiment, the user terminal further includes a calibration module, which is used to calibrate the business process automation process, obtain corresponding calibration results, and perform corresponding processing according to the calibration results.

[0109] The calibration result is normal or abnormal calibration as well as the cause and phenomenon of the abnormality.

[0110] In one embodiment, the calibration of the business process automation process is real-time calibration, that is, real-time calibration is performed during the work of the target script, and the execution of the target script is not interfered with, such as the target script is executed after determining the corresponding operation steps; that is, the current common real-time calibration.

[0111] In one embodiment, the calibration of the business process automation process is pre-calibration, which differs from real-time calibration in that: after the operation steps are determined, they are not executed directly, but calibrated for the operation steps. Execution is performed only after the calibration passes. Otherwise, a corresponding warning is issued.

[0112] In one embodiment, the calibration of the business process automation processing is a combination of pre-calibration and real-time calibration, that is, pre-calibration is performed regularly or at an intelligently determined time during the real-time calibration process, such as combining existing anomaly prediction technology to predict the anomaly probability. When the anomaly probability is greater than a preset value, a pre-calibration period is determined, and pre-calibration is performed during the pre-calibration period, and real-time calibration is performed at other times; other methods can also be used to combine real-time calibration and pre-calibration.

[0113] In one embodiment, calibrating a business process automation process includes:

[0114] Preset standard data, i.e., the standard data of the target script under normal conditions when performing business process automation. Each calibration item and calibration standard is determined based on the standard data, which is generally determined and set by the platform.

[0115] The operation processing data corresponding to the target script is obtained in real time, the operation processing data is calibrated according to the calibration items and calibration standards, whether there are abnormal conditions, and the calibration result is determined.

[0116] In one embodiment, the business process automation process is calibrated, and calibration can also be performed based on various existing calibration methods, such as establishing an intelligent model based on machine learning, deep learning algorithms, etc. for intelligent calibration.

[0117] The above formulas are all calculated by removing dimensions and taking their numerical values. The formula is a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and preset thresholds in the formula are set by technicians in this field according to actual conditions or obtained by simulating a large amount of data.

[0118] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A business process automation processing system based on browser scripts, characterized in that: Including platform side and user side; The platform end includes a script module and a script library; the user end includes a business analysis module, an automatic processing module, and a calibration module; The script module is used to perform script requirement analysis, obtain corresponding script requirements, set corresponding browser scripts according to the script requirements, and aggregate the browser scripts and store them in a script library; The script library is used to store browser scripts; The business analysis module is used to match browser scripts that meet user needs from the script library on the platform side, verify and adjust the browser scripts, obtain the target script, and configure the target script accordingly; The automatic processing module is used to perform business process automation processing according to the target script; The calibration module is used to calibrate the business process automation process, obtain corresponding calibration results, and perform corresponding processing according to the calibration results; Conduct script requirements analysis, including: Obtain various business processing requirements for business process automation; identify browser information related to the business processing requirements; Decomposing the business processing requirements into steps to obtain atomic operation data of the business processing requirements; combining the atomic operation data with corresponding browser information to obtain several groups of analysis combinations; Each analysis combination is classified through a preset merged evaluation model to obtain a corresponding requirement classification. The merged evaluation model is used to evaluate whether the same browser script can be applied between the corresponding analysis combinations. The script requirements are set according to the requirement classification, and one requirement classification corresponds to one script requirement.

2. A browser script-based business process automation processing system according to claim 1, characterized in that: Categorize each analysis combination, including: Step SA1: calling the preset merge evaluation model; Step SA2: arbitrarily select two analysis combinations to form input data and input them into the combined evaluation model for analysis to obtain corresponding combined evaluation results; the combined evaluation results include whether they meet the classification requirements or do not meet the classification requirements; Step SA3: When the combined evaluation result meets the classification requirements, the corresponding two analysis combinations are classified into one category, marked as the intermediate category, and proceed to step SA4; When the combined evaluation result does not meet the classification requirements, the two analysis combinations are marked as corresponding different class labels, and the process returns to step SA2; Step SA4: Select one of the remaining analysis combinations to form the input data corresponding to the intermediate class and input it into the combined evaluation model for analysis to obtain the corresponding combined evaluation result; When the combined evaluation result is in compliance with the classification requirements, the corresponding analysis combination is merged into the intermediate class to form a new intermediate class; When the combined evaluation result does not meet the classification requirements, the corresponding different class markings are performed; Step SA5: looping step SA4 until there is no combined evaluation result that meets the classification requirements, marking the intermediate class as a demand classification; Step SA6: Loop steps SA2 to SA5 until all analysis combinations are classified.

3. A browser script-based business process automation processing system according to claim 2, characterized in that: The expression of the combined evaluation model is: ; Where: s is the number of analysis combinations that need to be evaluated; the output data is the combined evaluation value HP(s), which is 1 or 0; When the combined evaluation value is 1, the combined evaluation result meets the classification requirements; When the combined evaluation value is 0, the combined evaluation result does not meet the classification requirements.

4. The browser script-based business process automation processing system according to claim 1, characterized in that: Match browser scripts that meet user needs from the platform's script library, including: Obtain user needs, extract features of user needs, obtain corresponding matching feature groups, input the matching feature groups into the script library on the platform side for matching, and obtain browser scripts that meet user needs.

5. The browser script-based business process automation processing system according to claim 1, characterized in that: Make validation adjustments to browser scripts, including: Step SC1: Generate corresponding interactive simulation data according to the browser script and user needs; Step SC2: displaying the interactive simulation data to the user; When the user is satisfied with the interactive simulation data, the adjustment is completed and the analysis ends; When the user is not satisfied with the interaction simulation data, the interaction simulation data is marked accordingly to obtain interaction marking data; and new interaction simulation data is generated according to the interaction marking data; Step SC3: Loop step SC2.

6. The browser script-based business process automation processing system according to claim 5, characterized in that: When the number of cycles of step SA2 reaches the threshold value X1, a prompt is given to the corresponding staff of the platform, who then assists the user in adjusting the browser script.

7. The browser script-based business process automation processing system according to claim 1, characterized in that: The calibration of the business process automation process is real-time calibration and / or pre-calibration.

Images