An automatic construction method, system, device and medium for a full-link component mapping process based on optimal path search

Abstract

The application relates to the technical field of component mapping and flow construction, and discloses a full-link component mapping flow automatic construction method, system, device and medium based on optimal path searching, which comprises the following steps: generating a business flow sketch according to external observable characteristics of a business instance and determining necessary flow stages; identifying key component anchor points for each necessary flow stage based on component events, and forming a candidate anchor point sequence according to the stage sequence; constructing a component candidate relationship graph based on the relationship among components, calculating candidate edge costs and marking edge types; searching for an optimal path under anchor point covering constraints and sequence constraints, obtaining a main flow path and remaining candidate path segments; dividing the remaining segments according to edge types and splicing to generate a full-link component mapping flow; and performing flow integrity checking. The application avoids bypass calling and key stage omission through the flow sketch and anchor point constraints, restores nonlinear segments, and improves the integrity and interpretability of full-link flow construction.

Description

Technical Field

[0001] This invention relates to the field of component mapping and process construction technology, and in particular to an automatic construction method, system, device and medium for full-link component mapping process based on optimal path search. Background Technology

[0002] Existing technologies for component dependency identification, call chain tracing, and automatic system topology mapping still suffer from limitations such as a narrow perspective on process construction, lack of constraints on key components, reliance on a single path selection criterion, and difficulty in automatically splicing together multiple types of process fragments. These limitations make it difficult to meet the actual needs of complex business systems for automating, completing, and interpretably constructing end-to-end component mapping processes. Traditional methods mostly rely on explicit call chains, service dependencies, log events, or static topology configurations for analysis, neglecting the process sequence constraints and key component coverage requirements that real business processes should follow in stages such as entry point access, identity verification, business processing, data persistence, asynchronous triggering, and result return. This "call-side level, topology level" analysis approach fails to reflect the true organizational logic of component relationships at the business process level. Consequently, when faced with side-channel calls, asynchronous chain breaks, implicit data dependencies, missing process stages, or multiple solutions to component paths, the system struggles to stably construct end-to-end component mapping processes that conform to the actual business execution process. Summary of the Invention

[0003] In view of the aforementioned existing problems, the present invention provides a method, system, device and medium for automatically constructing a full-link component mapping process based on optimal path search.

[0004] Therefore, the technical problem solved by this invention is: To address the aforementioned technical problems, this invention provides the following technical solution: an automatic construction method for end-to-end component mapping process based on optimal path search, comprising: generating a business process sketch consisting of process stages based on the externally observable feature vectors of a business instance, and determining the necessary set of process stages; Based on component events, at least one key component anchor point is identified for each necessary process stage, and the candidate anchor points are arranged in the order of the process stages to form a sequence. Based on the relationships between components, candidate edges for components are established, a candidate relationship graph for components is constructed, the edge cost of the candidate edges is calculated, and the edge type is marked. Under anchor point coverage constraints and anchor point order constraints, optimal path search is performed based on the candidate anchor point sequence and component candidate relationship graph to obtain the set of remaining candidate path segments. The remaining candidate path segments are divided into path types and spliced ​​together to generate a full-link component mapping process. The integrity of the full-link component mapping process is verified, and the process is output when the verification passes.

[0005] As a preferred embodiment of the automatic construction method for end-to-end component mapping process based on optimal path search described in this invention, the step of generating a business process sketch composed of process stages based on the externally observable feature vectors of business instances and determining the necessary set of process stages includes: collecting business instances within a time window, extracting the externally observable features of each business instance, and forming a sequence of stage events; the externally observable features include business entry point, business object, data endpoint, message topic, and process end status; Sequence pattern mining is performed on the phase event sequence of the business instance to filter out the stable process phase sequence, and the confidence of the phase transition is calculated to generate the business process sketch. The business process sketch includes at least the entry access stage, identity verification stage, business processing stage, data persistence stage, asynchronous triggering stage, and result return stage; Based on the stability of each process stage in business instances, process stages with stability exceeding a threshold are identified as necessary process stages.

[0006] As a preferred embodiment of the automatic construction method for full-link component mapping process based on optimal path search described in this invention, the step of identifying at least one key component anchor point for each necessary process stage based on component events and arranging them into a candidate anchor point sequence according to the process stage order includes: for each necessary process stage, calculating the interface behavior matching strength, data access matching strength, message event matching strength, and state change matching strength between the component and the process stage according to the various component events generated by the component, and comprehensively obtaining the anchoring strength of the component to the process stage; The component events include component runtime events, API call events, data access events, message events, and state change events; The component with the highest anchoring strength in each necessary process stage is identified as the main anchor point of the current stage, and components with anchoring strength exceeding a preset threshold are included in the candidate anchor point set of the current stage. The candidate anchor point sequence is obtained by arranging the candidate anchor point sets of each stage according to the process stage order in the business process sketch.

[0007] As a preferred embodiment of the automatic construction method for full-link component mapping process based on optimal path search described in this invention, the relationships between components include link tracing relationships, data object sharing relationships, message production and consumption relationships, and state transition relationships. For any two components in the system, if at least one of the relationships between the components exists, then a candidate edge is established between them. Based on the main relationship type on which the candidate edges are based, the edge type is marked as a call chain edge, a data dependency edge, a message trigger edge, a state transition edge, or a mixed edge of multiple types. The overall credibility of the candidate edges is calculated by combining the evidence strength of the candidate edges in link tracing, data sharing, message consumption and state transition, and the edge cost of the candidate edges is determined based on the overall credibility. The strength of evidence includes time support strength, data support strength, message support strength, and state support strength.

[0008] As a preferred embodiment of the automatic construction method for full-link component mapping process based on optimal path search described in this invention, wherein: under the constraints of anchor point coverage and anchor point order, the optimal path search is performed according to the candidate anchor point sequence and the candidate component relationship graph to obtain the set of remaining candidate path segments, which includes: taking the set of adjacent candidate anchor points in the candidate anchor point sequence as intervals, and searching for the local optimal path segment that satisfies the anchor point coverage constraint and has the minimum path cost in each interval; The anchor point coverage constraint requires the path to pass through at least one component in the current interval start and end candidate anchor point set, and the anchor point order constraint requires the path to pass through the candidate anchor point set in the same order as the process stage order. The local optimal path segments in each interval are spliced ​​together in the order of the process stages to form the main process path; the path cost is determined by the edge cost of the candidate edges, the anchor point coverage, the degree of conformity of the anchor point order, and the path length. Candidate path segments that were not selected for the main process path but are associated with the main process path through shared components, shared data objects, or messages are retained as the set of remaining candidate path segments.

[0009] As a preferred embodiment of the automatic construction method for full-link component mapping process based on optimal path search described in this invention, the step of dividing the path types of the remaining candidate path segments and splicing them together to generate the full-link component mapping process includes: for each remaining candidate path segment, if the starting point is located on the main process path and the ending point re-enters the main process path or reaches the same result state as the main process, then it is divided into a branch path. If the main edge type among the remaining candidate path segments is a message-triggered edge, it is classified as an asynchronous path; if the main edge type is a data-dependent edge, it is classified as a data-dependent path; if the remaining candidate path segments are associated with abnormal state events, timeout events, or rollback events, they are classified as compensation paths. The obtained branch paths, asynchronous paths, data dependency paths, and compensation paths are combined with the main process path to generate a full-link component mapping process that includes the main process, branch processes, asynchronous processes, data dependency processes, and compensation processes.

[0010] As a preferred embodiment of the automatic construction method for full-link component mapping process based on optimal path search described in this invention, the step of performing integrity verification on the full-link component mapping process and outputting the process when the verification passes includes: determining the process integrity score based on the number of covered stages in the necessary process stage set, the number of path break segments in the full-link component mapping process, the number of reverse order directions of component connection direction opposite to the process stage order, and the number of isolated components not connected to the main process. When the process integrity score is greater than or equal to the preset integrity threshold, the full-link component mapping process is output; When the process integrity score is less than the threshold, the process returns to adjust the identification of the key component anchor points or the construction of the component candidate relationship graph, and regenerates the full-link component mapping process until the verification passes.

[0011] This invention provides an automatic construction system for the entire component mapping process based on optimal path search.

[0012] As a preferred embodiment of the automatic construction system for full-link component mapping process based on optimal path search described in this invention, it includes: a process sketch generation module, an anchor point identification module, a candidate relationship graph construction module, an optimal path search module, and a process segment splicing and verification module. The process sketch generation module is used to generate a business process sketch based on the externally observable characteristics of the business instance and to determine the necessary set of process stages. The anchor point identification module is used to identify at least one key component anchor point for each necessary process stage based on component events, and arrange them in the order of process stages to form a candidate anchor point sequence. The candidate relationship graph construction module is used to establish candidate edges for components in the system based on the relationships between components, construct a candidate relationship graph of components, calculate the edge cost of candidate edges, and mark the edge type; The optimal path search module is used to perform optimal path search based on the candidate anchor sequence and component candidate relationship graph under anchor coverage constraints and anchor order constraints, to obtain the main process path and the set of remaining candidate path segments associated with the main process. The process segment splicing and verification module is used to divide the remaining candidate path segments into branch paths, asynchronous paths, data dependency paths and compensation paths according to the main process path, the set of remaining candidate path segments and the edge type, and splice them to generate a full-link component mapping process; perform integrity verification on the generated full-link component mapping process, and output the process when the verification passes.

[0013] The present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of an automatic construction method for a full-link component mapping process based on optimal path search.

[0014] The present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of an automatic construction method for a full-link component mapping process based on optimal path search are implemented.

[0015] The beneficial effects of this invention are as follows: This invention constructs a business process sketch using externally observable features, avoiding accidental entry into the main process via bypass calls; it identifies key anchor points based on component events and applies coverage and sequence constraints to prevent the omission of key stages; it calculates edge costs using multi-source evidence to ensure the selection of the main process path that conforms to the business execution logic; and it categorizes and splices remaining fragments according to edge type to restore asynchronous, data-dependent, and compensation processes, forming a complete mapping. Simultaneously, it introduces quantitative scoring and a closed-loop self-correction mechanism to improve the reliability and long-term evolutionary adaptability of the automatically constructed results, providing an interpretable and maintainable end-to-end mapping process for component governance of complex business systems. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of an automatic construction method for end-to-end component mapping based on optimal path search, provided as an embodiment of the present invention.

[0018] Figure 2 This invention provides a candidate component relationship diagram for an automatic construction method of full-link component mapping process based on optimal path search, as an embodiment of the present invention.

[0019] Figure 3 This is a flowchart of the process segment splicing and integrity verification of an automatic construction method for end-to-end component mapping based on optimal path search, provided as an embodiment of the present invention. Detailed Implementation

[0020] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0021] Example 1, referring to Figure 1 This is the first embodiment of the present invention, which provides an automatic construction method for a full-link component mapping process based on optimal path search, including: S1: Based on the externally observable feature vectors of the business instance, generate a business process sketch consisting of process stages and determine the necessary set of process stages.

[0022] S2: Based on component events, identify at least one key component anchor point for each necessary process stage, and arrange them in the order of process stages to form a candidate anchor point sequence.

[0023] S3: Based on the relationships between components, establish candidate edges for components, construct a candidate relationship graph for components, calculate the edge cost of candidate edges, and label the edge type.

[0024] S4: Under the constraints of anchor point coverage and anchor point order, perform optimal path search based on the candidate anchor point sequence and the candidate component relationship graph to obtain the set of remaining candidate path segments.

[0025] S5: Divide the remaining candidate path segments into path types, and concatenate them to generate a full-link component mapping process. Perform integrity verification on the full-link component mapping process, and output the process when the verification passes.

[0026] It should be noted that this method can automate, stage, constrain, and interpretably construct the entire component mapping process in complex and ever-changing distributed business systems. Compared with traditional component mapping methods that only rely on call chains, service dependencies, or static topologies, this invention can simultaneously consider business process sketches, key component anchors, candidate path costs, fragment type classification, and process integrity verification, thereby effectively overcoming the shortcomings of existing technologies in process boundary identification, key component coverage, multi-path selection, asynchronous fragment splicing, and mapping result interpretation.

[0027] Example 2, refer to Figure 2 and Figure 3 As an embodiment of the present invention, based on the above embodiment, an automatic construction method for end-to-end component mapping process based on optimal path search is provided.

[0028] Furthermore, in this embodiment of the application, step S1 generates a business process sketch consisting of process stages based on the externally observable feature vector of the business instance, and determines the necessary set of process stages. Specific steps include: S101: Collect business instances within the time window, extract the entry characteristics, business object characteristics, data landing point characteristics, message triggering characteristics, and exit characteristics of each business instance, and form a phase event sequence; Specifically, the input is the set of entry requests. Business object identifier set Data endpoint set Message Topic Collection Callback address set and process end status set The goal of this step is not to directly determine the component-level call path, but rather to generate a process phase-level sketch based on observable external characteristics of the business logic. Its output is a business process sketch. and necessary process stages set .

[0029] Let the time window to be analyzed be: in, Indicates the first The start time of each time window, Indicates the window length.

[0030] In the time window Within this system, the set of business instances collected is as follows: For any business instance The system extracts its entry characteristics, business object characteristics, data endpoint characteristics, message triggering characteristics, and exit characteristics, which are represented as follows: in, Indicates the business entry point. Represents a business object. Indicates the data endpoint. Indicates the message trigger point. Indicates business export.

[0031] To clarify the mapping process from business instance features to process stage-level sketches, the system first maps the externally observable feature vectors of business instance u. Convert to a phased event sequence: Where sq represents the q-th stage event in the business instance (q=1,2,…,n), and the values ​​include process stages such as entry access, identity verification, business processing, data persistence, asynchronous triggering, and result return.

[0032] S102: Perform sequence pattern mining on the phase event sequences of multiple business instances, filter out the stable process phase sequences, calculate the confidence of phase transitions, and generate the business process sketch; the business process sketch includes at least the entry access phase, identity verification phase, business processing phase, data persistence phase, asynchronous triggering phase, and result return phase. Set time window The set of stage sequences corresponding to all business instances within the system is as follows: The system uses a sequence pattern mining method to extract stable process stage sequences from the stage sequence set S.

[0033] For any phase sequence X, its support is defined as: Where X⊆Su indicates that the stage sequence X appears sequentially in the business instance sequence Su.

[0034] When the following conditions are met: At that time, it was assumed that the stage sequence X had sufficient stability.

[0035] Furthermore, for adjacent process stages and The confidence level of its stage transition is defined as: When the following conditions are met: At that time, it was considered a stage To the stage There is a stable process sequence relationship between them.

[0036] in, This represents the minimum support threshold. This represents the minimum confidence threshold.

[0037] Finally, based on the recurring entry points, objects, data endpoints, message topics, and exit statuses in multiple business instances, the system generates a process phase-level sketch: in, Indicates the first Each process stage This indicates the number of process stages. The process stages include at least the entry access stage, identity verification stage, business processing stage, data persistence stage, asynchronous triggering stage, and result return stage.

[0038] It should be noted that, This represents a process stage-level sketch, rather than a specific component call path. This sketch is used to determine the boundaries of process stages that need to be covered later, so that the component mapping process is no longer determined solely by local call edges.

[0039] S103: Based on the stability of each process stage in the business instance, process stages with stability exceeding the threshold are identified as necessary process stages, thus obtaining the set of necessary process stages.

[0040] set up Indicates based on business instance The system identifies the process stages by recognizing the entry point, business object, data destination, message topic, and exit status. To determine whether each process stage is a necessary process stage, the system calculates the stage stability. in, Indicates an indicator function.

[0041] When a certain process stage meets the following: Then add this process stage to the set of necessary process stages: in, This represents the stage stability threshold.

[0042] This step outputs the set of necessary process stages. This will serve as the stage constraint input for identifying key component anchor points.

[0043] Furthermore, in this embodiment, step S2, based on component events, identifies at least one key component anchor point for each necessary process stage and arranges them in the order of the process stages to form a candidate anchor point sequence. The specific steps include: S201: For each necessary process stage, based on the interface call events, data access events, message events and state change events generated by the component, calculate the matching strength of the component's interface behavior with that process stage, the matching strength of data access, the matching strength of message events and the matching strength of state change, and comprehensively obtain the anchoring strength of the component to that process stage. Specifically, the input to this step is the set of necessary process stages output by S1. and component runtime event set API call event collection Data access event set Message event collection and set of state change events Its output is the set of primary anchor points. and candidate anchor sequence .

[0044] Let the system component set be: For components and necessary process stages The anchoring strength of the system's computing components to this stage of the process: in, Representation Component At any moment For process stages The event response strength is such that a larger value indicates that the component event is more consistent with the characteristics of that process stage. This indicates the reliability of the component event evidence, and its value can be determined based on the credibility of the log source, the completeness of the fields, and the completeness of the link identifier.

[0045] The intensity of the event response is determined by interface behavior, data access, message events, and state changes. The interface behavior matching strength is defined as follows: in, Representation Component In the time window The collection of interface call events generated internally. Indicates whether event e is related to the process stage. match.

[0046] Data access matching strength is defined as: The message event matching strength is defined as: The state change matching strength is defined as: in, and Representing components respectively In the time window The set of data access events, message events, and state change events generated within the system; To prevent constants with a denominator of zero; This indicates the strength of the match between the interface behavior and the process stage. This indicates the strength of the match between data object access and process stage. This indicates the strength of the match between the message event and the process stage. This indicates the strength of the match between the state change and the process stage; It is a non-negative adjustment parameter, and satisfies: S202: The component with the highest anchoring strength in each necessary process stage is identified as the main anchor point of that stage, and components with anchoring strength exceeding the preset anchor point identification threshold are included in the candidate anchor point set of that stage. The candidate anchor point sequence is obtained by arranging the candidate anchor point sets of each stage according to the process stage order in the business process sketch.

[0047] Specifically, for each necessary process stage The system selects the component with the highest anchoring strength as the main anchor point for this stage: Simultaneously, the system adds components with anchoring strength exceeding the threshold to the candidate anchor point set for this stage: in, This indicates the anchor point identification threshold. Indicates the first The main anchor point of each process stage Indicates the first A set of candidate anchor points for each process stage.

[0048] like If there is only one component, then that component serves as the anchor point; if If a path contains multiple components, then during the path search process, one of the components will be automatically selected as the actual anchor point based on the total path cost.

[0049] Finally, the system generates a sequence of candidate anchor points according to the order of the process stages: Furthermore, in this embodiment of the application, step S3 establishes candidate edges for components based on the relationships between components, constructs a candidate relationship graph for components, calculates the edge cost of the candidate edges, and labels the edge type. The specific steps include: S301: For any two components in the system, if there is at least one of the following relationships: link tracing relationship, data object sharing relationship, message production and consumption relationship, or state transition relationship, then establish a candidate edge between the two components. Specifically, the input for this step is a collection of components. Link tracing relationship indicator function Data object sharing relationship indicator function Message production and consumption relationship indicator function State transition relationship indicator function The goal of this step is to construct a candidate component graph for pathfinding and to compute edge costs and types for candidate edges. The output is the candidate component graph. Edge cost matrix and edge type marking function .

[0050] in, Representation Component To Component A link tracing relationship exists; otherwise, the value is 0. This indicates that there is a data object sharing relationship between the two; otherwise, it is 0. This indicates that there is a message production and consumption relationship between the two; otherwise, it is 0. This indicates that there is a state transition relationship between the two; otherwise, it is 0.

[0051] System Component Candidate Relationship Diagram: in, Represents a collection of component nodes. Represents the set of candidate component edges. Represents the edge cost matrix. This indicates the function for marking edge types.

[0052] For any two components and Candidate edges are established if any of the following relationships are satisfied. : S302: Based on the main relationship type on which the candidate edge is based, mark its edge type as a call chain edge, data dependency edge, message triggering edge, state transition edge, or mixed edge of multiple types; in, Indicates calling the chain edge, Indicates data dependency edges, Indicates the message-triggered edge, Represents a state transition edge. This represents a mixed edge of multiple types. The edge type labeling function... This will serve as the basis for distinguishing between asynchronous paths, data-dependent paths, and compensation paths.

[0053] S303: Calculate the overall credibility of the candidate edges by combining the evidence strength of the candidate edges in link tracing, data sharing, message consumption, and state transition, and determine the edge cost of the candidate edges based on the overall credibility; wherein, the evidence strength includes time support strength, data support strength, message support strength, and state support strength.

[0054] For candidate edges The system calculates the support strength for different types of evidence. Let... Representation Component To Component If the time interval is a random variable, then the time support strength is defined as: in, Indicates time interval The probability density function, This represents the time decay parameter.

[0055] Specifically, the system is based on components To Component Historical correlation event estimation time interval probability density function Within the time window Inside, if components The incident The time of occurrence is Components Subsequent related events The time of occurrence is And satisfy Then we obtain a sample of one time interval: Where h represents the h-th time interval sample. This yields the component. To Component Time interval sample set: Where H represents the number of samples in the time interval.

[0056] The system will divide the time range Divide into M equal-width sub-intervals, each with a width of: The m-th subinterval is represented as: Count the number of time intervals that fall into the m-th subinterval: Then the time interval probability density function It can be estimated as follows: in, This represents the maximum time interval threshold, and M represents the number of interval divisions. Indicates the width of the sub-interval.

[0057] In one embodiment, This can be the maximum allowed call latency or message processing latency for the business, where M is an integer between 20 and 100. When H=0, it indicates that the component... With components If there are no observable time-related samples within the current time window, then let The corresponding time support strength is 0.

[0058] Data support strength is defined as: in, and Representing components respectively With components The collection of data objects accessed Represents data objects Business weight, To prevent constants with a denominator of zero.

[0059] Message support strength is defined as: in, Indicates that it is composed of components Production and components A collection of consumer messages and These represent the message sets associated with the two components.

[0060] State support strength is defined as: in, Representation Component The associated set of state events Representation Component The associated set of state events Indicates that it is composed of components State event triggering component The set of state transitions for subsequent events.

[0061] To avoid the edge cost degenerating into a simple linear weighting, a multi-source evidence complementary fusion method is used to calculate the edge credibility: This formula indicates that any enhancement of any type of evidence will increase the credibility of the candidate edge, and when multiple types of evidence exist simultaneously, the credibility of the candidate edge is further enhanced.

[0062] Finally, the candidate edge cost is defined as: in, This represents the penalty for abnormal connections. To ensure that this penalty is enforceable, it is defined as follows: in, Indicates a function that indicates a jump across levels. This indicates a missing business identifier function. This indicates the direction reversal indicator function. Indicates a low-trust source indicator function; This is a non-negative penalty parameter.

[0063] Furthermore, in this embodiment of the application, step S4, under the constraints of anchor point coverage and anchor point order, performs an optimal path search based on the candidate anchor point sequence and the candidate component relationship graph to obtain a set of remaining candidate path segments. The specific steps include: S401: Using the set of adjacent candidate anchor points in the candidate anchor point sequence as an interval, search for the locally optimal path segment that satisfies the anchor point coverage constraint and has the minimum path cost within each interval; the anchor point coverage constraint requires that the path passes through at least one component in the start and end candidate anchor point sets of the interval, and the anchor point order constraint requires that the order in which the path passes through the candidate anchor point sets is consistent with the order of the process stages. Let one candidate component path be: This path needs to cover the candidate anchor point set sequentially according to the phase order in the business process sketch. That is, for the candidate anchor point sequence: A valid path should satisfy: S402: Piece together the locally optimal path segments of each interval according to the process stage sequence to form the main process path; wherein, the path cost is determined by the edge cost of the candidate edge, the anchor point coverage, the degree of conformity of the anchor point order and the path length. The anchor point coverage loss of a path is defined as: Among them, if path Covering the Candidate anchor point set for each process stage If the coverage loss is 0, then the coverage loss for that stage is 1; otherwise, it is 1.

[0064] When the candidate anchor set When a set contains multiple components, the system does not require the path to cover all components in the set at the same time. Instead, it treats the set of candidate anchor points as a set of alternative anchor points for the same process stage.

[0065] set up: in, This represents the number of candidate anchor points corresponding to the r-th process stage.

[0066] For adjacent candidate anchor point sets and The system calculates the interval path cost between any pair of anchor points: in, and .

[0067] For any and The system calculates its interval path cost. .

[0068] System selection makes Take the anchor point pair with the minimum value as the actual anchor point passed through, denoted as: in, This anchor point combination will then serve as the starting and ending point for subsequent local optimal path searches.

[0069] Therefore, when the path passes through the set of candidate anchor points When any component in the process is covered, that process stage is considered covered. The corresponding anchor point coverage indicator function is defined as: The corresponding anchor point coverage indicator function is defined as follows: Among them, when When the value is 1; when The value is 0 at that time.

[0070] Therefore, the multiple components in the candidate anchor set are related by "or" rather than "and".

[0071] The anchor order loss of a path is defined as: in, Representing a path The first appearance of the number in China The position of components within a set of candidate anchor points; if the path Not covered Then let If the first The anchor point for the first stage appears in the [missing information]. After each stage anchor point, a sequence loss occurs.

[0072] The total path cost is defined as: The first term represents the sum of edge costs in the path, the second term represents the anchor point missing penalty, the third term represents the anchor point incorrect order penalty, and the fourth term represents the path length penalty. These are non-negative weight parameters.

[0073] The system determines the optimal component mapping path as follows: in, Represents the candidate relationship graph of components The set of candidate paths that meet the start and end conditions.

[0074] To reduce the complexity of the global path search, the system breaks down the complete path search into multiple anchor point interval searches. For adjacent candidate anchor point sets... and The system searches for locally optimal path segments: in, Indicates by the first The candidate anchor point set is added to the first The set of feasible paths between a set of candidate anchor points.

[0075] The system splices the path segments of each interval into the main flow path according to the anchor point order: in, This indicates a path segment concatenation operation.

[0076] S403: The candidate path segments that were not selected into the main process path but are associated with the main process path through shared components, shared data objects or messages are retained as the set of remaining candidate path segments.

[0077] Specifically, the system will not be included. However, candidate path segments that have a start point association, end point association, or shared data object relationship with the main process path are retained as the set of remaining candidate path segments: in, Represents path fragments With the main process path There is a correlation. This correlation includes: and There are shared components, or there are components... Middle component points Candidate edges of components, or between components, there is a shared data object, message triggering relationship, or state transition relationship.

[0078] Furthermore, in this embodiment, step S5 divides the path types of the remaining candidate path segments and concatenates them to generate a full-link component mapping process. The integrity of the full-link component mapping process is then verified, and the process is output when the verification passes. Specific steps include: S501: For each remaining candidate path segment, if its starting point is on the main process path and its ending point rejoins the main process path or reaches the same result state as the main process, then it is divided into a branch path. If the main edge type among the remaining candidate path segments is a message-triggered edge, it is classified as an asynchronous path; if the main edge type is a data-dependent edge, it is classified as a data-dependent path; if the remaining candidate path segments are associated with abnormal state events, timeout events, or rollback events, they are classified as compensation paths. The obtained branch paths, asynchronous paths, data dependency paths, and compensation paths are combined with the main process path to generate a full-link component mapping process that includes the main process, branch processes, asynchronous processes, data dependency processes, and compensation processes.

[0079] Specifically, based on the main process path, branch paths, asynchronous paths, data dependency paths, and compensation paths are identified from the remaining candidate path fragments, and then concatenated to generate a complete end-to-end component mapping process. Its output is the end-to-end component mapping process. Process integrity score and process verification results .

[0080] For any remaining candidate path segment ,set up and Representing path segments The starting and ending components. If its starting point is in the main flow path and its ending point rejoins the main flow path or reaches the same result state as the main flow, it is identified as a branch path: in, Main process path, For the set of remaining candidate path segments, For the set of branch paths, This represents the business state corresponding to the end of a component or path, which can be represented by a set of state change events or a set of process end states. Sure.

[0081] set up Represents path fragments The set of edges contained within. If the primary edge type of a path fragment is a message-triggered edge, it is identified as an asynchronous path: If the primary edge type of a path segment is a data-dependent edge, then it is identified as a data-dependent path: If a path segment primarily appears after an exception status code, timeout event, or rollback event, it is identified as a compensation path: in, , and These represent the threshold values ​​for determining asynchronous paths, data-dependent paths, and compensation paths, respectively. Represents the set of events in the normal process. This represents a set of abnormal status codes, timeout events, rollback events, or failure status events.

[0082] The final generated end-to-end component mapping process is represented as follows: in, For asynchronous paths, For data dependency paths, For compensation path.

[0083] S502: Determine the process integrity score based on the number of covered stages in the necessary process stage set, the number of broken path segments in the full-link component mapping process, the number of reverse order components whose connection direction is opposite to the process stage order, and the number of isolated components not connected to the main process. When the process integrity score is greater than or equal to the preset integrity threshold, the full-link component mapping process is output; when the process integrity score is less than the threshold, the process returns to adjust the identification of the key component anchor points or the construction of the component candidate relationship graph, and regenerates the full-link component mapping process until the verification is passed.

[0084] Specifically, the system further verifies the integrity of the process. Let... Represents the end-to-end component mapping process The number of necessary process stages already covered in the document. If the number of necessary process stages is represented, then the process integrity score is defined as follows: in, Indicates the number of broken segments. Indicates the number of inversions in the direction. Indicates the number of isolated components. This is a non-negative penalty parameter.

[0085] The process validation result is defined as: in, This indicates the threshold for process integrity.

[0086] when When, the system outputs the mapping process of the entire link component; when When this happens, the system returns a set of anchor points for adjusting key components, or a candidate relationship diagram for extended components.

[0087] Example 3 is the third embodiment of the present invention, which differs from the previous two embodiments in that: This embodiment also provides an automatic construction system for full-link component mapping process based on optimal path search, including: a process sketch generation module, an anchor point identification module, a candidate relationship graph construction module, an optimal path search module, and a process segment splicing and verification module; The process sketch generation module is used to generate business process sketches based on the externally observable characteristics of business instances and to determine the necessary set of process stages. Anchor point identification module is used to identify at least one key component anchor point for each necessary process stage based on component events, and arrange them in the order of process stages to form a candidate anchor point sequence; The candidate relationship graph construction module is used to establish candidate edges for components in the system based on the relationships between components, construct the component candidate relationship graph, calculate the edge cost of the candidate edges, and label the edge type; The optimal path search module is used to perform optimal path search based on the candidate anchor sequence and component candidate relationship graph under anchor coverage constraints and anchor order constraints, to obtain the main process path and the set of remaining candidate path segments associated with the main process. The process segment splicing and verification module is used to divide the remaining candidate path segments into branch paths, asynchronous paths, data dependency paths and compensation paths according to the main process path, the set of remaining candidate path segments and the edge type, and splice them to generate a full-link component mapping process; perform integrity verification on the generated full-link component mapping process, and output the process when the verification passes.

[0088] This embodiment also provides an electronic device, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize an automatic construction method for full-link component mapping process based on optimal path search as proposed in the above embodiment.

[0089] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements an automatic construction method for a full-link component mapping process based on optimal path search as proposed in the above embodiment.

[0090] The storage medium proposed in this embodiment and the automatic construction method for a full-link component mapping process based on optimal path search proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0091] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0092] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0093] Example 4 is the first embodiment of the present invention, used to verify an automatic construction method for full-link component mapping process based on optimal path search.

[0094] Taking a company's work order processing system as an example, the system includes a front-end portal component. Application gateway component Authentication service components Work order management component Database components Message queue component and notification service components .

[0095] The system first generates a flowchart based on the business entry point, business object, data endpoint, and message trigger point: Based on statistical results from multiple business instances, the set of necessary process stages is as follows: The system identifies candidate anchor point sequences based on necessary process stages: in, For the entrance access anchor point, As an anchor point for identity verification, As an anchor point for business processing, Used as an anchor point for data persistence.

[0096] Within the time window, the component edge cost matrix constructed by the system is as follows: in, Although direct access exists, it bypasses the identity verification anchor point. Its path cost is high; This indicates that the gateway directly connects to the database component. This connection lacks support during normal business operations, and therefore incurs higher costs.

[0097] The system searches for local path segments according to the anchor point order: The main process path is obtained after splicing: At the same time, the system detected the following path fragments: in, and All edges are message-triggered edges, thus meeting the asynchronous path determination criteria and being identified as asynchronous notification paths. The final process for generating the end-to-end component mapping is as follows: The first element represents the main process path, the second element represents the set of branch paths, the third element represents the set of asynchronous paths, the fourth element represents the set of data dependency paths, and the fifth element represents the set of compensation paths.

[0098] If the system detects a path If a path appears within a short period of time but does not pass through the identity verification anchor point and the business processing anchor point, and does not conform to the stage order in the flowchart, then the path will not be included in the main process, but will enter the set of remaining candidate path segments and be marked as a bypass path to be confirmed.

Claims

1. A method for automatically constructing a full-link component mapping process based on optimal path search, characterized in that, include: Based on the externally observable feature vectors of the business instance, a business process sketch consisting of process stages is generated, and the necessary set of process stages is determined. Based on component events, at least one key component anchor point is identified for each necessary process stage, and the candidate anchor points are arranged in the order of the process stages to form a sequence. Based on the relationships between components, candidate edges for components are established, a candidate relationship graph for components is constructed, the edge cost of the candidate edges is calculated, and the edge type is marked. Under anchor point coverage constraints and anchor point order constraints, optimal path search is performed based on the candidate anchor point sequence and component candidate relationship graph to obtain the set of remaining candidate path segments. The remaining candidate path segments are divided into path types and spliced ​​together to generate a full-link component mapping process. The integrity of the full-link component mapping process is verified, and the process is output when the verification passes.

2. The automatic construction method for end-to-end component mapping process based on optimal path search as described in claim 1, characterized in that, The step of generating a business process sketch consisting of process stages based on the externally observable feature vectors of business instances and determining the necessary set of process stages includes: collecting business instances within a time window, extracting the externally observable features of each business instance, and forming a sequence of stage events; the externally observable features include business entry point, business object, data endpoint, message topic, and process end status; Sequence pattern mining is performed on the phase event sequence of the business instance to filter out the stable process phase sequence, and the confidence of the phase transition is calculated to generate the business process sketch. The business process sketch includes at least the entry access stage, identity verification stage, business processing stage, data persistence stage, asynchronous triggering stage, and result return stage; Based on the stability of each process stage in business instances, process stages with stability exceeding a threshold are identified as necessary process stages.

3. The automatic construction method for end-to-end component mapping process based on optimal path search as described in claim 2, characterized in that, The step of identifying at least one key component anchor point for each necessary process stage based on component events and arranging them into a candidate anchor point sequence according to the process stage order includes: for each necessary process stage, calculating the interface behavior matching strength, data access matching strength, message event matching strength, and state change matching strength of the component with the process stage according to the various component events generated by the component, and comprehensively obtaining the anchoring strength of the component with the process stage. The component events include component runtime events, API call events, data access events, message events, and state change events; The component with the highest anchoring strength in each necessary process stage is identified as the main anchor point of the current stage, and components with anchoring strength exceeding a preset threshold are included in the candidate anchor point set of the current stage. The candidate anchor point sequence is obtained by arranging the candidate anchor point sets of each stage according to the process stage order in the business process sketch.

4. The automatic construction method for end-to-end component mapping process based on optimal path search as described in claim 3, characterized in that, The relationships between components include tracing relationships, data object sharing relationships, message production and consumption relationships, and state transition relationships; For any two components in the system, if at least one of the relationships between the components exists, then a candidate edge is established between them. Based on the main relationship type on which the candidate edges are based, the edge type is marked as a call chain edge, a data dependency edge, a message trigger edge, a state transition edge, or a mixed edge of multiple types. The overall credibility of the candidate edges is calculated by combining the evidence strength of the candidate edges in link tracing, data sharing, message consumption and state transition, and the edge cost of the candidate edges is determined based on the overall credibility. The strength of evidence includes time support strength, data support strength, message support strength, and state support strength.

5. The automatic construction method for end-to-end component mapping process based on optimal path search as described in claim 4, characterized in that, Under anchor point coverage constraints and anchor point order constraints, the optimal path search is performed based on the candidate anchor point sequence and component candidate relationship graph to obtain the set of remaining candidate path segments, which includes: taking the adjacent candidate anchor point sets in the candidate anchor point sequence as intervals, and searching for the local optimal path segment that satisfies the anchor point coverage constraints and has the minimum path cost in each interval. The anchor point coverage constraint requires the path to pass through at least one component in the current interval start and end candidate anchor point set, and the anchor point order constraint requires the path to pass through the candidate anchor point set in the same order as the process stage order. The local optimal path segments in each interval are spliced ​​together in the order of the process stages to form the main process path; the path cost is determined by the edge cost of the candidate edges, the anchor point coverage, the degree of conformity of the anchor point order, and the path length. Candidate path segments that were not selected for the main process path but are associated with the main process path through shared components, shared data objects, or messages are retained as the set of remaining candidate path segments.

6. The automatic construction method for end-to-end component mapping process based on optimal path search as described in claim 5, characterized in that, The process of dividing the remaining candidate path segments into path types and splicing them together to generate a full-link component mapping includes: for each remaining candidate path segment, if the starting point is located on the main process path and the ending point re-enters the main process path or reaches the same result state as the main process, then it is divided into a branch path. If the main edge type among the remaining candidate path segments is a message-triggered edge, it is classified as an asynchronous path; if the main edge type is a data-dependent edge, it is classified as a data-dependent path; if the remaining candidate path segments are associated with abnormal state events, timeout events, or rollback events, they are classified as compensation paths. The obtained branch paths, asynchronous paths, data dependency paths, and compensation paths are combined with the main process path to generate a full-link component mapping process that includes the main process, branch processes, asynchronous processes, data dependency processes, and compensation processes.

7. The method for automatically constructing a full-link component mapping process based on optimal path search as described in claim 6, characterized in that, The integrity verification of the full-link component mapping process, when the verification passes, outputs the process including: determining the process integrity score based on the number of covered stages in the necessary process stage set, the number of broken path segments in the full-link component mapping process, the number of reverse order directions of component connection direction opposite to the process stage order, and the number of isolated components not connected to the main process. When the process integrity score is greater than or equal to the preset integrity threshold, the full-link component mapping process is output; When the process integrity score is less than the threshold, the process returns to adjust the identification of the key component anchor points or the construction of the component candidate relationship graph, and regenerates the full-link component mapping process until the verification passes.

8. An automatic construction system for end-to-end component mapping process based on optimal path search, employing the method for automatic construction of end-to-end component mapping process based on optimal path search as described in any one of claims 1 to 7, characterized in that, include: The module includes a flowchart generation module, an anchor point recognition module, a candidate relationship graph construction module, an optimal path search module, and a process segment splicing and verification module. The process sketch generation module is used to generate a business process sketch based on the externally observable characteristics of the business instance and to determine the necessary set of process stages. The anchor point identification module is used to identify at least one key component anchor point for each necessary process stage based on component events, and arrange them in the order of process stages to form a candidate anchor point sequence. The candidate relationship graph construction module is used to establish candidate edges for components in the system based on the relationships between components, construct a candidate relationship graph of components, calculate the edge cost of candidate edges, and mark the edge type; The optimal path search module is used to perform optimal path search based on the candidate anchor sequence and component candidate relationship graph under anchor coverage constraints and anchor order constraints, to obtain the main process path and the set of remaining candidate path segments associated with the main process. The process segment splicing and verification module is used to divide the remaining candidate path segments into branch paths, asynchronous paths, data dependency paths and compensation paths according to the main process path, the set of remaining candidate path segments and the edge type, and splice them to generate a full-link component mapping process; perform integrity verification on the generated full-link component mapping process, and output the process when the verification passes.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the automatic construction method for full-link component mapping process based on optimal path search as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the automatic construction method for full-link component mapping process based on optimal path search as described in any one of claims 1 to 7.

Images