A self-reinforcing full-element mining method, device, equipment, media, and element mining system based on spiral iteration.

Abstract

This application discloses a self-reinforcing full-element mining method, apparatus, equipment, medium, and element mining system based on spiral iteration, relating to the field of business activity element mining. The method uses the full set of business activity elements as input for the k-th iteration and obtains a completeness score using an element completeness scoring function. If the completeness score is less than a preset threshold and the value of k is less than the maximum number of iterations, missing elements are identified. A confidence evaluation function is used to select the completion scheme with the highest confidence and add it to the full element set, generating a new element set. The value of k is incremented by 1, and the above steps are repeated. If the completeness score is greater than or equal to the preset threshold or the value of k is greater than or equal to the maximum number of iterations, the final full element set is solidified, and a structured and standardized full element set is output. This application enables dynamic, complete, and accurate intelligent mining of all elements.

Description

Technical Field

[0001] This application relates to the field of business activity element mining, and in particular to a self-reinforcing full-element mining method, apparatus, equipment, medium, and element mining system based on spiral iteration. Background Technology

[0002] In the current field of business process modeling and optimization, accurate and comprehensive mining of key elements in business activities is the foundation for achieving intelligent decision-making. However, existing technologies have the following limitations in element mining:

[0003] 1. Static and incomplete: Existing methods mostly rely on manual sorting or static rules, making it difficult to comprehensively and dynamically identify all entities, environmental and tool elements in the business process. In particular, the ability to mine implicit relationships and dynamically changing elements is seriously insufficient.

[0004] 2. Lack of closed-loop optimization mechanism: The traditional factor mining process is usually unidirectional and open-loop. Once completed, it is difficult to self-correct and continuously optimize according to the actual changes in business, which leads to the gradual disconnect between the model and actual business operation.

[0005] 3. Rough definition of tool elements: For functional nodes (i.e. "tool elements") in the process, existing technologies generally lack standardized and structured descriptions of their versions, interfaces, dependencies, and other information, which hinders subsequent automated process orchestration and intelligent analysis. Summary of the Invention

[0006] The purpose of this application is to provide a self-reinforcing full-element mining method, device, equipment, medium, and element mining system based on spiral iteration, which can realize dynamic, complete, and accurate intelligent mining of all elements.

[0007] To achieve the above objectives, this application provides the following solution:

[0008] Firstly, this application provides a self-reinforcing full-element mining method based on spiral iteration, the method being executed by an element mining system, the method comprising:

[0009] Identify structured elements from the raw data set of business activities to obtain an initial set of all elements;

[0010] Initialize the value of k to 0;

[0011] The initial set of all features is used as the input for the k-th iteration, and the integrity score of the initial set of all features is obtained by using the feature integrity score function.

[0012] If the completeness score is less than a preset threshold and the value of k is less than the maximum number of iterations, then the initial set of all elements is identified for missing items, and a structured set of missing items is generated.

[0013] Multiple completion schemes are generated for the structured set of missing items, and the highest confidence completion scheme is selected from the multiple completion schemes using a confidence evaluation function.

[0014] Add the highest confidence completion scheme to the initial full feature set to generate a new feature set;

[0015] Increment the value of k by 1, replace the initial full feature set with the new feature set, and return to the step of using the initial full feature set as input for the k-th iteration and using the feature integrity score function to obtain the integrity score of the initial full feature set.

[0016] If the completeness score is greater than or equal to a preset threshold, or the value of k is greater than or equal to the maximum number of iterations, then the current input set of all elements will be used as the final set of all elements.

[0017] Solidify the final set of all elements and output a structured and standardized set of all elements.

[0018] Secondly, this application provides an element mining system that performs the above-described self-reinforcing full-element mining method based on spiral iteration.

[0019] Thirdly, this application provides a self-reinforcing full-element mining device based on spiral iteration, including: an element identification module, an initialization module, an iteration module, a missing element identification module, a completion module, an addition module, a loop module, a loop termination module, and an output module.

[0020] The feature recognition module is used to identify structured features from the raw data set of business activities and obtain an initialized full feature set.

[0021] An initialization module is used to initialize the value of k to 0;

[0022] The iterative module is used to take the initialized full feature set as input for the k-th iteration and use the feature integrity score function to obtain the integrity score of the initialized full feature set.

[0023] The missing item identification module is used to identify missing items in the initialized full element set if the completeness score is less than a preset threshold and the value of k is less than the maximum number of iterations, and to generate a structured set of missing items.

[0024] The completion module generates multiple completion schemes for a structured set of missing items and uses a confidence evaluation function to select the completion scheme with the highest confidence from the multiple completion schemes;

[0025] The addition module is used to add the completion scheme with the highest confidence to the initial full feature set, generating a new feature set;

[0026] The loop module is used to increment the value of k by 1, replace the initialized full feature set with the new feature set, and call the iteration module;

[0027] The loop termination module is used to take the current input set of all elements as the final set of all elements if the integrity score is greater than or equal to a preset threshold, or if the value of k is greater than or equal to the maximum number of iterations.

[0028] The output module is used to solidify the final set of all elements, outputting a structured and standardized set of all elements.

[0029] Fourthly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to implement the above-described self-reinforcing full-element mining method based on spiral iteration.

[0030] Fifthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described self-reinforcing full-element mining method based on spiral iteration.

[0031] According to the specific embodiments provided in this application, this application has the following technical effects:

[0032] This application provides a spiral iterative self-reinforcing full-element mining method, device, equipment, medium, and element mining system. The cyclic iterative process constitutes the spiral iterative self-reinforcing algorithm framework, which can actively identify missing data and self-drive iterative enhancement, transforming a one-time mining task into a continuous, closed-loop optimization process with dynamic adaptability. It can continuously evolve with changes in business data, completely solving the fundamental problem that traditional static methods cannot cope with the dynamic nature of business. By introducing a formalized element integrity scoring function, an objective and computable metric is provided for element integrity assessment, greatly improving the rigor and reproducibility of the cyclic iterative process, eliminating reliance on fuzzy rules or human experience, and improving the integrity of element mining. By introducing a confidence evaluation function, the completion scheme is strictly screened, ensuring the high accuracy of knowledge enhancement. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] Figure 1 A flowchart illustrating the self-reinforcing full-element mining method based on spiral iteration provided in this application embodiment;

[0035] Figure 2 A schematic diagram of the self-reinforcing algorithm framework for spiral iteration provided in the embodiments of this application;

[0036] Figure 3 A schematic diagram of the functional modules of the self-reinforcing full-element mining device based on spiral iteration provided in the embodiments of this application;

[0037] Figure 4 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0038] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0039] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0040] To achieve dynamic, complete, and accurate intelligent data mining of all elements, in an exemplary embodiment, such as... Figure 1 As shown, a self-reinforcing full-element mining method based on spiral iteration is provided, including the following steps 101 to 109. Wherein:

[0041] Step 101: Identify structured elements from the original data set of business activities to obtain an initialized set of all elements.

[0042] Step 102: Initialize the value of k to 0.

[0043] Step 103: Use the initialized full feature set as input for the k-th iteration, and use the feature integrity score function to obtain the integrity score of the initialized full feature set.

[0044] Step 104: If the completeness score is less than the preset threshold and the value of k is less than the maximum number of iterations, then perform missing item identification on the initialized full element set to generate a structured missing item set.

[0045] Step 105: Generate multiple completion schemes for the structured set of missing items, and use a confidence evaluation function to select the completion scheme with the highest confidence from the multiple completion schemes.

[0046] Step 106: Add the completion scheme with the highest confidence to the initialized full feature set to generate a new feature set.

[0047] Step 107: Increment the value of k by 1, replace the initial full feature set with the new feature set, and return to execute step 103.

[0048] Step 108: If the completeness score is greater than or equal to the preset threshold, or the value of k is greater than or equal to the maximum number of iterations, then the current input set of all elements is taken as the final set of all elements.

[0049] Step 109: Solidify the final set of all elements, and output a structured and standardized set of all elements.

[0050] The execution entity for steps 101 to 109 above is the element mining system (a computing device), which continuously optimizes the entire element set through a series of steps with clear input-output relationships. Steps 102 to 108 constitute a spiral iterative self-reinforcing algorithm framework: modeling the element mining process as a closed-loop iterative cycle including quantitative evaluation, missing element identification, intelligent completion, and set enhancement. This framework transforms the static mining task into a dynamic, continuously evolving self-reinforcing process. Step 103 constitutes a quantifiable element integrity evaluation method: to accurately drive the above cycle, a computable element integrity score function is proposed. This function can objectively quantify the quality of the element set, serving as the core basis for iterative optimization. Step 105 constitutes a confidence-based intelligent completion mechanism: to ensure the convergence and accuracy of the enhancement process, a confidence evaluation function is introduced in the completion stage. This mechanism ensures that only highly reliable updates are adopted, thereby achieving precise element enhancement.

[0051] In another exemplary embodiment of this application, the raw data set of business activities includes: documents, operation manuals, system logs, operation records, and structured data in the database.

[0052] In another exemplary embodiment of this application, the structured elements include: entities, environment, and tools.

[0053] In another exemplary embodiment of this application, the detailed implementation process of step 101 above for constructing the initial full set of elements can be as follows:

[0054] Implementing entity: Element mining system.

[0055] Input: Original data set D raw Home Knowledge Collection K expert .

[0056] Processing procedure: The system performs data preprocessing, converting D... raw The data is then converted into high-quality data. Subsequently, the system uses this high-quality data and expert knowledge K... expert The system comprehensively reviews business processes, initially identifying and defining elements such as Entity, EntityType, and Environment. Simultaneously, it identifies key business functional units that can be abstracted into business tools and records their initial top-level descriptions, thereby constructing an initial set of all elements, A. (0) .

[0057] Output: The initial set of all elements, A (0) .

[0058] In another exemplary embodiment of this application, the element integrity scoring function is:

[0059]

[0060] In the formula, The integrity score for the k-th iteration. Let a be the set of all elements input in the k-th iteration. i w is the i-th element in the full set of elements. i Let V(a) be the weight of the i-th element in the full set of elements. i ) is the internal integrity measure of the i-th element in the full set of elements.

[0061] In another exemplary embodiment of this application, the detailed process of step 104 above to identify missing items in the initialized full feature set and generate a structured missing item set can be as follows: compare the initialized full feature set with the standard feature template, identify missing items in the initialized full feature set, and generate a structured missing item set from all missing items.

[0062] In another exemplary embodiment of this application, the detailed process of generating multiple completion schemes for the structured missing item set in step 105 above can be as follows: using the All-element Modeling Empowering Large Language Model (AELM) to complete the internal structure of the tools in the structured missing item set, and mining and completing the entities and environment to generate multiple completion schemes.

[0063] AELM is a large-scale language model with natural language processing and reasoning capabilities, capable of handling structured elements. AELM, through specific optimizations or by default, possesses the ability to understand and process structured elements such as entities, environments, and tools as defined in this application. It can be a specialized model obtained by domain-adaptive fine-tuning based on a general-purpose large model, or it can be a future general-purpose foundational model capable of directly processing such structured data.

[0064] Reference Figure 2 The spiral iterative self-reinforcing algorithm framework shown above provides a more detailed explanation of the spiral iterative self-reinforcing performed in steps 102 to 108.

[0065] Implementing entity: Element mining system.

[0066] Input: Initialized set of all elements A (0) .

[0067] Processing procedure: The system uses A (0) Using A as the initial input, a loop enhancement process is initiated until a preset termination condition is met. In the k-th iteration of the loop, the system starts with A... (k) For input, the following internal operations are executed sequentially:

[0068] a. Integrity Assessment: The system calls the feature integrity scoring function to assess the current feature set A. (k) The integrity score is calculated.

[0069] b. Termination Judgment: The system determines the integrity score C(A). (k) Is it not less than the preset threshold C? threshold Or, has the current iteration count k reached the maximum limit K? max If any condition is met, the loop terminates; otherwise, execution continues.

[0070] c. Missing Item Identification: The system uses structured analysis to compare standard element templates with domain knowledge to identify missing items (A). (k) To address issues such as missing attributes and broken relationship chains, a structured set of missing items, GapSet, is generated. (k) .

[0071] d. Intelligent completion and filtering: The system is designed for GapSets. (k) The missing items in the tool feature are filled with deep fill, and all generated fill solutions are evaluated for confidence and filtered. Taking an incompletely defined tool feature as an example, the deep fill process includes:

[0072] 1. Tool internal structure completion: Utilize AELM's model-based reasoning capabilities to refine the meta (such as I / O contracts) of a specific version of the tool, propose implementation suggestions for its model, and plan the structure of its data components.

[0073] 2. Related Element Mining and Completion: Based on the tool's preliminarily defined functions, the prior knowledge mining capabilities of AELM are used to reverse mine and complete the entity and environmental elements on which it depends.

[0074] The system ultimately calls the confidence evaluation function Conf(δ|A) to filter out the high-confidence update set ΔA. * .

[0075] e. Set augmentation: The system updates set ΔA with high confidence. * Apply to the current feature set to generate a new version of the feature set A. (k+1) And increment the iteration count k by 1 for the next iteration.

[0076] Output: When the loop terminates, the system outputs the latest version of the complete set of all elements that meets the integrity requirements, denoted as A. final .

[0077] In another exemplary embodiment of this application, step 109 above outputs the final result.

[0078] Implementing entity: Element mining system.

[0079] Input: The final set of all elements, A final .

[0080] Processing procedure: The system solidifies the input set of elements as the final output, producing a structured and standardized set of all elements. The elements and their relationships within this set can then be used to construct a dynamic, enterprise-specific all-element model, providing services to downstream business systems.

[0081] Output: The final optimized, high-quality, structured collection of all elements.

[0082] The technical effects of this application are as follows:

[0083] 1. Achieves dynamic self-evolution in element mining, overcoming the limitations of static methods: The technical effectiveness of this application is rooted in its core "spiral iterative self-reinforcing algorithm framework." This framework transforms a one-time mining task into a continuous, closed-loop optimization process. The system can proactively identify deficiencies in the current knowledge base (ˋc. missing data identification) and self-drive iterative enhancement (ˋe. set enhancement). This mechanism enables the system to dynamically adapt and continuously evolve with changes in business data, completely solving the fundamental problem that traditional static methods cannot cope with the dynamism of business.

[0084] 2. It endows the optimization process with rigorous measurability and improves the reproducibility of the solution: The technical effect of this application stems from its innovative "quantifiable element integrity assessment method." This is achieved by introducing a formalized element integrity scoring function. This application provides an objective and computable metric for the fuzzy problem of "how good is the set of elements". This score not only provides a clear and quantifiable convergence target for the iterative loop (ˋb. termination judgmentˋ), but also makes every step of the optimization process traceable, thereby greatly improving the rigor and reproducibility of the technical solution and eliminating the reliance on fuzzy rules or human experience.

[0085] 3. Ensuring high accuracy of knowledge augmentation and guaranteeing the reliability of the final result: The technical effectiveness of this application relies on its key "confidence-based intelligent completion mechanism." In the intelligent completion stage (ˋd. Intelligent completion and selectionˋ), the system does not blindly accept all potential updates generated by AELM, but instead introduces a confidence evaluation function. A rigorous screening process is implemented. This mechanism ensures that only highly reliable knowledge (ΔA) is used. * Only those elements are integrated into the element set, effectively preventing error accumulation and model contamination, and fundamentally ensuring the final output A. final A dual improvement in both completeness and accuracy.

[0086] The following are some possible alternative technical implementation paths:

[0087] 1. An adaptive feature mining scheme based on evolutionary algorithms: Feature mining is modeled as an evolutionary optimization problem, with each feature set treated as an individual. Iterative optimization of the feature set is achieved through selection, crossover, and mutation operations using a genetic algorithm. A fitness function replaces the integrity evaluation function, achieving a self-reinforcing effect through population evolution.

[0088] 2. Reinforcement Learning-Based Element Discovery Agent Scheme: A reinforcement learning agent is constructed, modeling the element discovery process as a Markov decision process. The agent learns the optimal element discovery strategy through interaction with the environment. The reward function is designed to improve element integrity, and continuous learning and self-improvement are achieved through the policy gradient method.

[0089] 3. Graph Neural Network-Based Element Relationship Modeling Scheme: This scheme models the elements and their relationships in the business process as a dynamic knowledge graph, utilizing graph neural networks to learn the implicit relationships between elements. Through the dynamic evolution of the graph structure and the continuous updating of node embeddings, it achieves automatic element discovery and relationship completion.

[0090] 4. Uncertainty Element Mining Scheme Based on Bayesian Inference: This scheme uses Bayesian networks to model the probabilistic dependencies between elements, and performs probabilistic element discovery and completion through Bayesian inference. A posterior probability update mechanism is used to achieve dynamic knowledge accumulation, and uncertainty quantification ensures the quality of inference.

[0091] 5. Self-Organizing Map-Based Adaptive Feature Space Construction Scheme: This scheme utilizes a self-organizing map neural network to construct a high-dimensional feature space for features, discovering the intrinsic structure and clustering patterns of features through unsupervised learning. Adaptive adjustment of the network topology enables dynamic optimization of the feature space, and a competitive learning mechanism achieves automatic feature classification and completion.

[0092] The aforementioned alternative solutions, based on different technical paradigms, can all achieve dynamic mining, self-reinforcement, and quality assurance of elements, achieving the same technical effects as this application, and should be included within the scope of protection of this application.

[0093] Based on the same inventive concept, this application also provides an element mining system that performs the above-described self-reinforcing full-element mining method based on spiral iteration.

[0094] This application also provides a spiral iteration-based self-reinforcing full-element mining device for implementing the spiral iteration-based self-reinforcing full-element mining method described above. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more spiral iteration-based self-reinforcing full-element mining device embodiments provided below can be found in the limitations of the spiral iteration-based self-reinforcing full-element mining method described above, and will not be repeated here.

[0095] In one exemplary embodiment, such as Figure 3 As shown, a self-reinforcing full-element mining device based on spiral iteration is provided, comprising: an element identification module, an initialization module, an iteration module, a missing element identification module, a completion module, an addition module, a loop module, a loop termination module, and an output module.

[0096] The system comprises the following modules: an element identification module for identifying structured elements from the raw data set of business activities to obtain an initial set of all elements; an initialization module for initializing the value of k to 0; an iteration module for using the initial set of all elements as input for the k-th iteration and obtaining the integrity score of the initial set of all elements using an element integrity score function; a missing item identification module for identifying missing items in the initial set of all elements if the integrity score is less than a preset threshold and the value of k is less than the maximum number of iterations, thereby generating a structured set of missing items; and a completion module for generating multiple completion schemes for the structured set of missing items and enabling... The system uses a confidence evaluation function to select the highest-confidence completion scheme from multiple completion schemes; an addition module adds the highest-confidence completion scheme to the initialized full-feature set, generating a new feature set; a loop module increments the value of k by 1, replaces the initialized full-feature set with the new feature set, and calls the iteration module; a loop termination module uses the current input full-feature set as the final full-feature set if the completeness score is greater than or equal to a preset threshold, or the value of k is greater than or equal to the maximum number of iterations; and an output module solidifies the final full-feature set, outputting a structured and standardized full-feature set.

[0097] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 4 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs in the non-volatile storage media to run. The database stores a structured, standardized set of all features. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a self-reinforcing full-feature mining method based on spiral iteration.

[0098] Those skilled in the art will understand that Figure 4The structures shown are merely block diagrams of some structures related to the present application and do not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than shown in the figures, or combine certain components, or have different component arrangements. In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0099] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0100] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0101] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0102] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0103] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0104] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A self-reinforcing full-element mining method based on spiral iteration, characterized in that, The method is executed by a feature mining system, and the method includes: Identify structured elements from the raw data set of business activities to obtain an initial set of all elements; Initialize the value of k to 0; The initialized full feature set is used as input for the k-th iteration, and the feature integrity score of the initialized full feature set is obtained using the feature integrity score function; the feature integrity score function is: In the formula, The integrity score for the k-th iteration. Let the set of all elements be the input for the k-th iteration. The first element in the total set of elements One element, For the first element in the total set of elements The weight of each element For the first element in the total set of elements Internal integrity measurement of each element; If the completeness score is less than a preset threshold and the value of k is less than the maximum number of iterations, then the initial set of all elements is identified for missing items, and a structured set of missing items is generated. Multiple completion schemes are generated for a structured set of missing items, and a confidence evaluation function is used to select the completion scheme with the highest confidence from the multiple completion schemes. Specifically, generating multiple completion schemes for a structured set of missing items includes: using a full-element large model to complete the internal structure of the tools in the structured set of missing items, and mining and completing entities and the environment to generate multiple completion schemes; the full-element large model is a large-scale language model with natural language processing and reasoning capabilities, and capable of processing structured elements. Add the highest confidence completion scheme to the initial full feature set to generate a new feature set; Increment the value of k by 1, replace the initial full feature set with the new feature set, and return to the step of using the initial full feature set as input for the k-th iteration and using the feature integrity score function to obtain the integrity score of the initial full feature set. If the completeness score is greater than or equal to a preset threshold, or the value of k is greater than or equal to the maximum number of iterations, then the current input set of all elements will be used as the final set of all elements. Solidify the final set of all elements and output a structured and standardized set of all elements.

2. The self-reinforcing full-element mining method based on spiral iteration according to claim 1, characterized in that, The structured elements include: entities, environment, and tools.

3. The self-reinforcing full-element mining method based on spiral iteration according to claim 1, characterized in that, The initial set of all features is used to identify missing items, generating a structured set of missing items, specifically including: The initialized full feature set is compared with the standard feature template to identify missing items in the initialized full feature set, and all missing items are generated into a structured missing item set.

4. The self-reinforcing full-element mining method based on spiral iteration according to claim 1, characterized in that, The original data set for the business activities includes: documents, operation manuals, system logs, operation records, and structured data in the database.

5. A feature mining system, characterized in that, The feature mining system executes the self-reinforcing full feature mining method based on spiral iteration as described in any one of claims 1-4.

6. A self-reinforcing full-element mining device based on spiral iteration, characterized in that, The apparatus is used to implement the self-reinforcing full-element mining method based on spiral iteration as described in any one of claims 1-4, and the apparatus comprises: The feature recognition module is used to identify structured features from the raw data set of business activities and obtain an initialized full feature set. An initialization module is used to initialize the value of k to 0; The iterative module is used to take the initialized full feature set as input for the k-th iteration and use the feature integrity score function to obtain the integrity score of the initialized full feature set. The missing item identification module is used to identify missing items in the initialized full element set if the completeness score is less than a preset threshold and the value of k is less than the maximum number of iterations, and to generate a structured set of missing items. The completion module generates multiple completion schemes for a structured set of missing items and uses a confidence evaluation function to select the completion scheme with the highest confidence from the multiple completion schemes; The addition module is used to add the completion scheme with the highest confidence to the initial full feature set, generating a new feature set; The loop module is used to increment the value of k by 1, replace the initialized full feature set with the new feature set, and call the iteration module; The loop termination module is used to take the current input set of all elements as the final set of all elements if the integrity score is greater than or equal to a preset threshold, or if the value of k is greater than or equal to the maximum number of iterations. The output module is used to solidify the final set of all elements, outputting a structured and standardized set of all elements.

7. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that the processor executes the computer program to implement the self-reinforcing full-element mining method based on spiral iteration as described in any one of claims 1-4.

8. 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 self-reinforcing full-element mining method based on spiral iteration as described in any one of claims 1-4.

Images