Software cost method and system based on dynamic configuration and hierarchical analysis

Abstract

The invention relates to the field of data processing, and provides a software cost method and system based on dynamic configuration and hierarchical analysis. The method comprises the following steps: dynamically loading corresponding configuration parameters and algorithms by evaluating a method identification field; constructing a standardized cost parameter configuration system, including configuring an estimated function point reference value, an estimated function point weight coefficient and a scale adjustment factor; based on the evaluation method, a tree budget configuration table is preset, hierarchical node design is adopted, a budget tree is constructed, and a tree budget configuration and cost declaration / evaluation header is formed; transferring the configuration right of the tree budget configuration and the cost declaration / evaluation header to the back end, so that the back end dynamically loads a preset header template according to the enabled cost standard, and synchronously injects field attribute rules and calculation formula metadata; obtaining an unstructured document uploaded by a user, and analyzing the unstructured document and the header template by adopting a large language model to obtain software cost details; and generating an evaluation report through the expert review software cost details.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a software cost estimation method and system based on dynamic configuration and hierarchical parsing. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Existing software cost estimation schemes have the following problems: 1. The cost assessment model is rigid and lacks dynamic switching capability. Existing technology only supports a single cost method (such as the man-month unit price method or the function point method), and cannot switch assessment models according to project characteristics. For example, the simple man-month unit price method cannot be used when a small software project needs to be quickly estimated, while the function point method cannot be quickly switched for large and complex software projects.

[0004] 2. Fixed parameter configuration affects the accuracy of the assessment. The core parameters of the function point method (such as 4 points for EI and 5 points for EO) are hard-coded and cannot be dynamically adjusted. This will cause the assessment results to contradict the dynamic adaptation requirements of the "size adjustment factor" in the "Software Estimation Process". When the standard is updated, it is necessary to modify the underlying code to adapt again, which violates the "open and closed principle" of software engineering.

[0005] 3. Lack of hierarchical functional point breakdown. Existing technology directly outputs a flat list of functional points, without constructing a tree structure. This deficiency significantly reduces the intuitiveness of the evaluation process.

[0006] 4. The expert review mechanism has gaps. The existing system only provides function point decomposition functionality and has not built a complete expert collaboration loop. The system design focuses on counting the number of function points and lacks a mechanism for experts to deeply evaluate the complexity, quality characteristics, and other dimensions of function points. It cannot use expert experience to improve the accuracy of the decomposition process. Expert review opinions cannot be automatically converted into structured reports. Summary of the Invention

[0007] To address the technical problems mentioned above, this invention provides a software cost estimation method and system based on dynamic configuration and hierarchical analysis. This invention enables dynamic evaluation of software costs and full lifecycle management and collaborative review of the expert database.

[0008] To achieve the above objectives, the present invention adopts the following technical solution: The first aspect of this invention provides a software cost estimation method based on dynamic configuration and hierarchical analysis.

[0009] A software cost estimation method based on dynamic configuration and hierarchical resolution includes: The corresponding configuration parameters and algorithms are dynamically loaded by identifying the fields through the evaluation method. Construct a standardized cost parameter configuration system, including configuring the baseline values ​​of estimated function points, the weight coefficients of estimated function points, and the scale adjustment factors; Based on the evaluation method, a pre-set tree-shaped budget configuration table is used. A hierarchical node design is adopted to construct a budget tree, forming a tree-shaped budget configuration and cost declaration / review header. The configuration authority for tree-structured budgets and cost application / review headers is moved to the backend, enabling the backend to dynamically load pre-set header templates based on the enabled cost standards, and synchronously inject field attribute rules and calculation formula metadata. Get unstructured documents uploaded by users, parse the unstructured documents and header templates using a large language model, and obtain the software cost details; The cost details are reviewed by expert review software to generate an evaluation report.

[0010] Furthermore, in the evaluation method identification field stage, the default configuration and the software project main table configuration support dynamic switching between the man-month unit price method and the function point method.

[0011] Furthermore, in the stage of building a standardized cost parameter configuration system: it supports multiple options for configuration rule adjustment factors; it supports standard intelligent cloning function; it supports the coexistence of multiple versions of standards but only allows a single item to be in the published state, triggering integrity verification and serializing the configuration item to Redis cache when it is published.

[0012] Furthermore, the tree-structured budget configuration table includes: a built-in total budget, a hardware budget, a finished software budget, and a custom development budget; and whether a node is allowed to have child nodes.

[0013] Furthermore, the hierarchical node design includes: setting branch nodes and leaf nodes; if the nodes created by default are all branch nodes, they need to be manually converted into leaf nodes. After conversion, leaf nodes are prohibited from having new child nodes, and key evaluation information is filled in.

[0014] Furthermore, the budget tree is dynamically managed through a leaf node field, a tree depth field, and a leaf node count field. The leaf node field is used for front-end rendering control; the tree depth field is used to record the node hierarchy path and supports recursive calculation during dynamic import and export; and the leaf node count field is used to count the size of the terminal nodes.

[0015] Furthermore, the configuration authority for the tree-structured budget configuration and the cost application / review header is transferred to the backend stage. The backend database has a pre-set standardized header configuration table, which stores the evaluation method identification field, the corresponding header field, display attributes, and validation rules in a structured manner. When a user switches evaluation methods, the backend matches the tree-structured budget configuration rules through the evaluation method identification field and loads the associated header template into the memory cache.

[0016] Furthermore, a large language model is adopted, the process of which includes: by introducing a large language model, pre-setting the cost standard parameter set, functional point decomposition rules, and structured data rules into the cloud-native AI inference engine, and using prompt word engineering to build a dynamic parsing template, which includes a domain knowledge base, parameter mapping relationships, and verification rules; when users upload unstructured documents, the document parsing service is automatically called to extract the text content, and multimodal prompt words are dynamically generated in combination with project context information to drive the large language model to complete the semantic parsing and structured mapping from requirement description to functional points; after the parsing results are verified by confidence, they are fed back to the user in real time through an interactive visualization interface.

[0017] Furthermore, during the cost breakdown stage of the expert review software, it supports initiating multi-expert collaborative meetings through a joint review plan; when the meeting starts, the data in the application budget form is automatically initialized to the expert review form, and an independent review copy is generated for each expert.

[0018] A second aspect of the present invention provides a software cost estimation system based on dynamic configuration and hierarchical parsing.

[0019] A software cost estimation system based on dynamic configuration and hierarchical resolution includes: The first configuration module is configured to dynamically load the corresponding configuration parameters and algorithms by identifying the evaluation method field. The second configuration module is configured to: construct a standardized cost parameter configuration system, including configuring the baseline value of the estimated function point, the weight coefficient of the estimated function point, and the scale adjustment factor; The third configuration module is configured to: based on the evaluation method, pre-set a tree-shaped budget configuration table, adopt hierarchical node design, construct a budget tree, and form a tree-shaped budget configuration and cost declaration / review header; The header dynamic loading module is configured to transfer the configuration authority of the tree-structured budget configuration and the cost application / review header to the backend, so that the backend can dynamically load the pre-set header template according to the enabled cost standard and synchronously inject the field attribute rules and calculation formula metadata. The parsing module is configured to: acquire unstructured documents uploaded by users, parse the unstructured documents and header templates using a large language model, and obtain the software cost details; The evaluation output module is configured to generate an evaluation report based on the cost details reviewed by expert review software.

[0020] A third aspect of the present invention provides a computer device comprising: A processor, adapted to execute computer programs; A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in the software cost estimation method based on dynamic configuration and hierarchical resolution as described in the first aspect above.

[0021] A fourth aspect of the present invention provides a computer-readable storage medium storing a computer program adapted to be loaded by a processor and executed the steps of the software cost estimation method based on dynamic configuration and hierarchical resolution as described in the first aspect above.

[0022] Compared with the prior art, the beneficial effects of the present invention are: 1. In existing software cost estimation systems, the man-month unit price method and the function point method are mutually exclusive, and there is an urgent need to solve the problem of their coexistence through a new architecture design. To this end, this invention supports dynamic switching between the man-month unit price method and the function point method in the system default configuration and the software project main table configuration, so as to meet the evaluation needs of projects of different scales and stages, and realize multi-mode compatibility and dynamic switching of software evaluation methods.

[0023] 2. In response to the rigid constraints of the function point method, it is necessary to break through the traditional hard-coding implementation method. This invention constructs a structured dynamic adaptation mechanism for cost standards, and overcomes the technical bottlenecks of parallel storage of multiple standards, real-time parameter calling and dynamic migration of standard versions.

[0024] 3. In view of the fact that the existing planar function point storage structure is difficult to adapt to the evaluation needs of complex systems, this invention solves the technical problems of multi-level mapping and dynamic expansion of function points by establishing hierarchical storage and visual interaction functions for function points.

[0025] 4. In response to the differentiated requirements of application elements in different review scenarios, it is necessary to solve the problem of dynamic adaptation of the application header field structure and display logic. This invention realizes the dynamic and configurable presentation of the application header, breaking through the limitations of traditional fixed templates.

[0026] 5. Integrate large model technology to automatically break down software costs, supporting adaptation from unstructured text (such as Word descriptions) to structured input, and realizing intelligent parsing of software document descriptions and automatic decomposition of functional points.

[0027] 6. Existing expert management is limited to static information maintenance. A technical solution is needed to build dynamic updates of expert qualifications, random matching of review tasks, and retention of the review process. To this end, this invention improves the full life cycle management and collaborative review mechanism of the expert database and opens up a business closed loop between expert management and review execution.

[0028] 7. To address the limitations of single-round reviews and the technical issues of resolving conflicts in review results and automatically collecting review data, this invention constructs a process control and intelligent report generation capability for multi-round reviews, realizing the automated conversion of review conclusions into structured reports. Attached Figure Description

[0029] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0030] Figure 1 This is a flowchart illustrating a software cost estimation method based on dynamic configuration and hierarchical parsing, as shown in an embodiment of the present invention. Figure 2 This is a structural diagram of a software cost estimation system based on dynamic configuration and hierarchical analysis, as shown in an embodiment of the present invention. Figure 3 This is a structural diagram of a computer device shown in an embodiment of the present invention. Detailed Implementation

[0031] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0032] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0033] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0034] Figure 1 This is a flowchart illustrating a software cost estimation method based on dynamic configuration and hierarchical parsing, as shown in an embodiment of the present invention; the method includes: The corresponding configuration parameters and algorithms are dynamically loaded by identifying the fields through the evaluation method. Construct a standardized cost parameter configuration system, including configuring the baseline values ​​of estimated function points, the weight coefficients of estimated function points, and the scale adjustment factors; Based on the evaluation method, a pre-set tree-shaped budget configuration table is used. A hierarchical node design is adopted to construct a budget tree, forming a tree-shaped budget configuration and cost declaration / review header. The configuration authority for tree-structured budgets and cost application / review headers is moved to the backend, enabling the backend to dynamically load pre-set header templates based on the enabled cost standards, and synchronously inject field attribute rules and calculation formula metadata. Get unstructured documents uploaded by users, parse the unstructured documents and header templates using a large language model, and obtain the software cost details; The cost details are reviewed by expert review software to generate an evaluation report.

[0035] The following is a detailed description of this embodiment: 1. The system default configuration and the software project main table configuration support dynamic switching between the man-month unit price method and the function point method.

[0036] The system backend employs a plug-in architecture with pre-defined default evaluation methods (such as the function point method) to ensure compliance with the requirement that "evaluation methods are indispensable." When a software project is created, the system automatically writes the default evaluation method as the initial review mode into the project's main table, ensuring that each project has at least one valid evaluation method. When creating a table header template, the corresponding evaluation method is marked on the template. When the frontend needs to load the header template and algorithm for the corresponding evaluation method, the backend queries the corresponding header template based on the project's evaluation method, generates the corresponding header based on the template and the cost parameter configuration system, and returns it to the frontend. If it is necessary to modify the project's evaluation method, the data corresponding to the original method must be cleared, and then the corresponding method in the main table must be modified.

[0037] 2. Construct a standardized cost parameter configuration system to support dynamic switching of multiple standards and dynamic maintenance of indicators.

[0038] The system achieves coexistence and adaptation with mainstream function point evaluation standards through parameterized configuration, covering core parameter configurations such as estimated function point baseline values, estimated function point weight coefficients, and scale adjustment factors. It also supports dynamic management of drop-down options (name, value, and remarks triplets) for factors such as EI / EO / EQ. The system provides a standard intelligent cloning function, allowing users to quickly copy and edit existing standards to inherit and adjust differentiated items, reducing configuration workload. The system supports the coexistence of multiple versions of standards but only allows a single standard to be in the published state. During publication, integrity verification is triggered, and configuration items are serialized to the Redis cache to accelerate query response, achieving a unified approach of dynamic adaptation, efficient reuse, and high-performance querying.

[0039] 3. The software's cost tree model structure design supports the tree-like display of budget nodes.

[0040] Construct a tree-structured budget configuration. The system pre-configures a tree-structured budget configuration table, which includes core configuration items such as total budget, hardware budget, finished software budget, and custom development budget. Access control logic is implemented by configuring whether nodes are allowed to have child nodes: by default, only the custom development budget branch requires the software estimation process to be initiated; other nodes are prohibited from adding child nodes. During project initialization, the initial structure is automatically generated based on the budget tree configuration, ensuring that the budget framework aligns with business rules.

[0041] Dynamic node type management mechanism. The budget tree adopts a hierarchical node design. Branch nodes (such as system management) correspond to functional categories and only require macro-level control without specific estimation. Leaf nodes (such as user management, menu management, role management) carry out software estimation tasks. Nodes created by the system by default are all branch nodes. Users must manually convert them to leaf nodes. This operation will trigger two constraints: first, leaf nodes cannot have new child nodes added; second, key evaluation information must be filled in to ensure the integrity of the estimation data.

[0042] The intelligent tree structure storage and display have been optimized. The budget tree table achieves dynamic management through three key fields: a "Whether it is a leaf node" field for front-end rendering control to avoid expanding invalid nodes; a "Tree depth" field to record the node hierarchy path, supporting recursive calculations during dynamic import and export; and a "Number of leaf nodes" field to count the size of the terminal nodes, assisting in the implementation of cell merging logic. This design ensures the flexibility of the tree structure and solves the data consistency problem of traditional static trees when operating across levels through a real-time recursive algorithm.

[0043] 4. Dynamically loading table headers increases the versatility of table display.

[0044] A dynamic header adaptation mechanism with decoupled front-end and back-end is constructed. For the requirement of linked management of tree-structured budget configuration and cost application / review headers, existing technical solutions suffer from high coupling between the front-end and back-end: the front-end must simultaneously handle multiple complexities such as header rendering logic when switching evaluation methods, field attribute configuration (display type / disabled status / dropdown options), and formula binding rules, leading to a surge in maintenance costs and potential compatibility issues. By shifting header configuration authority to the back-end and establishing a mapping relationship of "evaluation method → ​​tree structure → header template," deep decoupling between headers and business logic can be achieved. When creating a header template, the corresponding evaluation method is marked. When the front-end needs to load the header, the back-end retrieves the header template based on the project's evaluation method and matches the corresponding cost standard. The header template contains pre-set replaceable cost parameter placeholders. These placeholders are replaced with actual cost parameter information, including field attributes and calculation formulas, according to the cost standard. The front-end only needs to perform static rendering to complete multi-scenario adaptation.

[0045] Enhance backend-driven configuration management capabilities. The system pre-defines standardized header configuration tables in the backend database, storing the evaluation method identifier along with corresponding header fields, display attributes, and validation rules in a structured manner. When a user switches evaluation methods, the backend service queries the standardized cost parameter configuration matching the evaluation method based on its unique identifier. Simultaneously, it matches the corresponding header template for that evaluation method. The header template is a JSON array consisting of JSON configurations for each column. Each column's JSON configuration includes frontend table rendering information and replaceable formulas. An example header template with JSON configuration for a specific column is shown below.

[0046] { "rules": [ { "trigger": "change", "message": "Cannot be empty", "required": true } ], "title": "Function Points Category", "align": "center", "type": "select", "ellipsis": true, "dictList": "${pointType}", "autoSize": { "minRows": 1, "maxRows": 3 }, "scopedSlots": { "customRender": "pointType" }, "dataIndex": "pointType", "isTrigger": "1", "width": 160, "refFieldCode": "unadjustedNum", "calculateFormula": "${ pointTypeCalculateFormula}", "disabled": false } In this context, ${pointType} will be replaced by an array of selectable function point categories in the cost standard when assembling the table header, and ${pointTypeCalculateFormula} will be replaced with the actual formula.

[0047] After obtaining the header template, the parameters in the standardized parameter configuration are replaced with placeholders to form a usable header for the front end, ensuring that front-end requests can obtain structured header data in real time. This design not only continues the dynamic routing capability of the existing estimation algorithm in the back end (such as the function point method requiring the loading of the adjustment factor calculation module), but also incorporates the header configuration into a unified technical framework. When adding a new evaluation method, only the backend configuration table needs to be expanded instead of the front-end code being refactored, keeping the system complexity within a reasonable threshold.

[0048] This solution achieves dual optimization value in its technical architecture. By centrally managing header configurations on the backend, it significantly reduces the complexity of conditional judgment logic on the frontend. On one hand, it avoids header field conflicts caused by switching evaluation methods in traditional solutions (e.g., the function point method requires displaying the "adjustment factor" column while the man-month unit price method requires hiding it). On the other hand, it supports synchronous header evolution when cost standards are dynamically updated (e.g., after adding a factor, the backend automatically injects the corresponding configuration item into the header). Real-world testing has verified that this architecture reduces the amount of frontend header rendering code by 95% and improves page loading speed, resulting in a significant technical difference.

[0049] 5. Intelligent cost document parsing and structured cost estimation based on generative AI.

[0050] By introducing large language model technology, the cost standard parameter set, functional point decomposition rules, and structured data rules are pre-built into the cloud-native AI inference engine. Dynamic parsing templates are constructed using prompt word engineering, which include a domain knowledge base, parameter mapping relationships, and verification rules. When users upload unstructured documents, the system automatically calls the document parsing service to extract the text content and dynamically generates multimodal prompt words based on project context information (such as the industry and assessment method type). This drives the LLM to complete the semantic parsing and structured mapping from requirement description to functional points. After the parsing results are verified with confidence, they are fed back to the user in real time through an interactive visualization interface.

[0051] 6. Introduce an expert database and conduct joint expert evaluations to rationalize expert review.

[0052] Establish a full lifecycle management system for experts. The system enables structured storage and dynamic maintenance of basic expert information, covering multiple dimensions such as age, education, degree, professional title, professional background, industry and technical fields of expertise, and supports the entire process of expert entry, information update, and expert removal.

[0053] It enables random selection and collaborative review of expert panelists. Based on project needs, joint review plans can be flexibly initiated, supporting random selection from an expert database or selection based on industry sector, technical expertise, professional title, and other criteria, ensuring the professional coverage and authority of the review team. The selection algorithm employs a transparent and fair mechanism, satisfying both single-project review needs and adapting to collaborative review scenarios for complex cross-domain projects, ultimately achieving efficient utilization of expert resources and providing scientific decision-making support for review conclusions.

[0054] 7. Establish a collaborative mechanism for real-time expert review and joint review.

[0055] The system supports real-time online expert review of software cost estimates and allows for the initiation of multi-expert collaborative meetings through joint review plans. When a meeting starts, the system automatically initializes the data from the submitted budget form to the expert review form and generates an independent review copy for each expert, achieving physical isolation of review data and non-interference access control, thus complying with the "one person, one review" compliance requirement.

[0056] The system enables dynamic review adaptation driven by evaluation methods. It can dynamically load the corresponding review header fields, weight allocation rules, and algorithm modules based on the evaluation method selected for the project (such as function point method / person-month unit price method).

[0057] Optimize the redundant storage strategy for non-leaf nodes in the tree-structured budget table. To address the query performance bottleneck of the tree-structured budget table, the system implements a redundant storage design for non-leaf nodes: each non-leaf node pre-stores the aggregated budget value. When a user queries, there is no need to traverse all child nodes in real time for weighted calculations; the result can be obtained directly by reading the redundant fields, thus optimizing the query response time for large-scale node budget tables.

[0058] Establish an event-driven update mechanism for the total budget. The system achieves real-time synchronous updates of the total budget through an event listening architecture: when any expert modifies a sub-item budget, the system automatically triggers a "budget change event," which is broadcast to the total budget calculator to update the total budget value. This ensures the consistency of the total budget in scenarios with concurrent reviews by multiple experts, solves the data latency problem caused by asynchronous updates, and achieves second-level budget linkage response.

[0059] 8. Supports exporting detailed cost estimates and generating review opinion documents after expert review.

[0060] Structured export of cost details. The system supports dynamic visualization and structured export of budget data after expert review. It uses Apache POI technology to parse and calculate the tree structure of the entire budget data. By recursively traversing the budget tree depth and the number of leaf nodes, it precisely controls the row and column merging logic of Excel cells (horizontal merging corresponds to the expanded width of the corresponding level, vertical merging matches the coverage of the parent node). The specific process includes: (1) Export the tree structure to Excel and merge the cells. The core is to calculate the number of rows (rowspan) and columns (colspan) that each node should occupy. Perform a depth-first traversal (DFS) of the entire tree and record two key pieces of information during the traversal: the maximum depth of the tree max_depth and the depth of each tree node node_depth.

[0061] (2) Column merging. Non-leaf nodes do not need to be merged in the column direction, and their colspan is fixed at 1. The colspan of leaf nodes is determined by their depth: colspan = max_depth - node_depth + 1. The data of leaf nodes will span all columns from their own level to the bottom level (maximum depth level).

[0062] (3) Row merging. The number of rows to be merged (rowspan) is equal to the total number of records (record_count) of the node. For leaf nodes, record_count = the number of records (i.e., rowspan = the number of records). For non-leaf nodes, record_count is the sum of the record_counts of all their child nodes, which is also their rowspan. During DFS traversal, a global row counter current_row needs to be maintained: when accessing a node, its starting row number is the current_row; after processing the node and all its subtrees, current_row is incremented by the rowspan of the node to ensure that the node occupies the row range [start_row, start_row + rowspan - 1] and the row numbers are assigned consecutively without conflict.

[0063] Ensure that the exported results form a complete mapping chain with the source identifiers of the budget application records. This design automatically calculates the average value of expert reviews for each application item as the approval value by linking review data with the original application records, thus ensuring data traceability and improving budget verification efficiency.

[0064] Review comment document generation. The system integrates a document template engine to automate the generation of review comments. It uses a pre-set standard Word template (including field placeholders for variable expressions) and dynamically replaces it with actual review data (such as expert conclusions and construction content) after the review is completed. It supports versioned archiving and structured output of multiple rounds of review comments. This mechanism uses the XWPFDocument component to parse the template syntax tree, ensuring that the document format remains consistent during the variable replacement process, while also being compatible with complex typesetting requirements.

[0065] The above combination Figure 1The software cost estimation method based on dynamic configuration and hierarchical parsing provided in the embodiments of the present invention has been described in detail. Next, the software cost estimation system based on dynamic configuration and hierarchical parsing provided in the embodiments of the present invention will be described in conjunction with the accompanying drawings.

[0066] Figure 2 This is a schematic diagram illustrating the structure of a software cost estimation system based on dynamic configuration and hierarchical parsing, as shown in an embodiment of the present invention. Figure 2 The system described in this invention includes: The first configuration module is configured to dynamically load the corresponding configuration parameters and algorithms by identifying the evaluation method field. The second configuration module is configured to: construct a standardized cost parameter configuration system, including configuring the baseline value of the estimated function point, the weight coefficient of the estimated function point, and the scale adjustment factor; The third configuration module is configured to: based on the evaluation method, pre-set a tree-shaped budget configuration table, adopt hierarchical node design, construct a budget tree, and form a tree-shaped budget configuration and cost declaration / review header; The header dynamic loading module is configured to transfer the configuration authority of the tree-structured budget configuration and the cost application / review header to the backend, so that the backend can dynamically load the pre-set header template according to the enabled cost standard and synchronously inject the field attribute rules and calculation formula metadata. The parsing module is configured to: acquire unstructured documents uploaded by users, parse the unstructured documents and header templates using a large language model, and obtain the software cost details; The evaluation output module is configured to generate an evaluation report based on the cost details reviewed by expert review software.

[0067] In some embodiments, during the evaluation method identification field stage, the system default configuration and the software project main table configuration support dynamic switching between the man-month unit price method and the function point method.

[0068] In some embodiments, during the stage of building a standardized cost parameter configuration system: multiple options for configuration rule adjustment factors are supported; standard intelligent cloning function is supported; multiple versions of standards can coexist but only a single item is allowed to be in the published state, and integrity verification is triggered when publishing and the configuration item is serialized to the Redis cache.

[0069] In some embodiments, the tree-shaped budget configuration table includes: a built-in total budget, a hardware budget, a finished software budget, and a custom development budget; and whether a node is allowed to have child nodes.

[0070] In some embodiments, the hierarchical node design includes: setting branch nodes and leaf nodes; if the nodes created by the system by default are all branch nodes, they need to be manually converted into leaf nodes. After conversion, leaf nodes are prohibited from having new child nodes, and key evaluation information is filled in.

[0071] In some embodiments, the budget tree is dynamically managed through a leaf node field, a tree depth field, and a leaf node count field. The leaf node field is used for front-end rendering control; the tree depth field is used to record the node hierarchy path and supports recursive calculation during dynamic import and export; and the leaf node count field is used to count the size of the terminal nodes.

[0072] In some embodiments, the configuration authority for the tree-structured budget configuration and the cost application / review header is transferred to the backend stage. A standardized header configuration table is preset in the backend database, which stores the evaluation method identification field, the corresponding header field, display attributes, and validation rules in a structured manner. When the user switches the evaluation method, the backend matches the tree-structured budget configuration rules through the evaluation method identification field and loads the associated header template into the memory cache.

[0073] In some embodiments, a large language model is employed, and the process includes: by introducing a large language model, pre-setting the cost standard parameter set, functional point decomposition rules, and structured data rules into the cloud-native AI inference engine, and using prompt word engineering to construct a dynamic parsing template, which includes a domain knowledge base, parameter mapping relationships, and verification rules; when a user uploads an unstructured document, the system automatically calls the document parsing service to extract the text content, and dynamically generates multimodal prompt words in combination with project context information, driving the large language model to complete the semantic parsing and structured mapping from requirement description to functional points; after the parsing results are verified by confidence, they are fed back to the user in real time through an interactive visualization interface.

[0074] In some embodiments, during the expert review of the software cost details stage, it supports initiating multi-expert collaborative meetings through a joint review plan; when the meeting starts, the system automatically initializes the data of the application budget form to the expert review form and generates an independent review copy for each expert.

[0075] According to embodiments of the present invention, the software cost estimation system based on dynamic configuration and hierarchical parsing can correspond to the execution of the methods described in the embodiments of the present invention, and the above and other operations and / or functions of each module of the software cost estimation system based on dynamic configuration and hierarchical parsing are respectively for implementing Figure 1 For the sake of brevity, the corresponding processes of each method in the code will not be elaborated here.

[0076] See Figure 3The diagram shows the structure of a computer device, which includes a processor, a communication interface, and a computer-readable storage medium. The processor, communication interface, and computer-readable storage medium are connected via a bus or other means. The communication interface is used to receive and send data. The computer-readable storage medium can be stored in the computer device's memory. The computer-readable storage medium stores computer programs, including program instructions, and the processor executes the program instructions stored in the computer-readable storage medium. The processor (or CPU, Central Processing Unit) is the computing and control core of the computer device, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions to implement the corresponding steps in the software cost estimation method embodiment based on dynamic configuration and hierarchical parsing.

[0077] This embodiment provides a computer-readable storage medium (Memory), which is a memory device in a computer device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the computer device and extended storage media supported by the computer device. The computer-readable storage medium provides storage space that stores the processing system of the computer device. Furthermore, this storage space also contains one or more instructions suitable for loading and execution by the processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be high-speed RAM memory or non-volatile memory, such as at least one disk storage device; optionally, it can also be at least one computer-readable storage medium located remotely from the aforementioned processor.

[0078] In one embodiment, the computer-readable storage medium stores one or more instructions; the processor loads and executes one or more instructions stored in the computer-readable storage medium to implement the corresponding steps in the above embodiment of the software cost estimation method based on dynamic configuration and hierarchical resolution.

[0079] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0080] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0081] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0082] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0083] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0084] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A software cost estimation method based on dynamic configuration and hierarchical parsing, characterized in that, include: The corresponding configuration parameters and algorithms are dynamically loaded by identifying the fields through the evaluation method. Construct a standardized cost parameter configuration system, including configuring the baseline values ​​of estimated function points, the weight coefficients of estimated function points, and the scale adjustment factors; Based on the evaluation method, a pre-set tree-shaped budget configuration table is used. A hierarchical node design is adopted to construct a budget tree, forming a tree-shaped budget configuration and cost declaration / review header. The configuration authority for tree-structured budgets and cost application / review headers is moved to the backend, enabling the backend to dynamically load pre-set header templates based on the enabled cost standards, and synchronously inject field attribute rules and calculation formula metadata. Get unstructured documents uploaded by users, parse the unstructured documents and header templates using a large language model, and obtain the software cost details; The cost details are reviewed by expert review software to generate an evaluation report.

2. The software cost estimation method based on dynamic configuration and hierarchical parsing according to claim 1, characterized in that, In the evaluation method identification field stage, the default configuration and the software project main table configuration support dynamic switching between the man-month unit price method and the function point method.

3. The software cost estimation method based on dynamic configuration and hierarchical parsing according to claim 1, characterized in that, During the construction of a standardized cost parameter configuration system: it supports multiple options for configuration rule adjustment factors; it supports standard intelligent cloning function; it supports the coexistence of multiple versions of standards but only allows a single item to be in the published state. When publishing, it triggers integrity verification and serializes the configuration item to the Redis cache.

4. The software cost estimation method based on dynamic configuration and hierarchical parsing according to claim 1, characterized in that, The tree-structured budget configuration table includes: a built-in total budget, a hardware budget, a finished software budget, and a custom development budget; and whether a node is allowed to have child nodes. or, The hierarchical node design is adopted, including: setting branch nodes and leaf nodes; if the nodes created by default are all branch nodes, they need to be manually converted into leaf nodes. After conversion, leaf nodes are not allowed to have new child nodes, and key evaluation information is filled in. or, The budget tree is dynamically managed through a leaf node field, a tree depth field, and a leaf node count field. The leaf node field is used for front-end rendering control; the tree depth field is used to record the node hierarchy path and supports recursive calculation during dynamic import and export; and the leaf node count field is used to count the size of the terminal nodes.

5. The software cost estimation method based on dynamic configuration and hierarchical parsing according to claim 1, characterized in that, With the configuration authority for tree-structured budgets and cost application / review headers shifted to the backend, a standardized header configuration table is pre-set in the backend database. The evaluation method identifier field, the corresponding header field, display attributes, and validation rules are stored in a structured manner. When a user switches evaluation methods, the backend matches the tree-structured budget configuration rules with the evaluation method identifier field and loads the associated header template into the memory cache.

6. The software cost estimation method based on dynamic configuration and hierarchical parsing according to claim 1, characterized in that, The process employs a large language model, which includes: pre-setting cost standard parameter sets, functional point decomposition rules, and structured data rules into a cloud-native AI inference engine by introducing a large language model; constructing dynamic parsing templates using prompt word engineering, which includes a domain knowledge base, parameter mapping relationships, and verification rules; automatically calling the document parsing service to extract text content when users upload unstructured documents, dynamically generating multimodal prompt words based on project context information, driving the large language model to complete semantic parsing and structured mapping from requirement descriptions to functional points; and providing real-time feedback to users through an interactive visualization interface after the parsing results are verified by confidence.

7. The software cost estimation method based on dynamic configuration and hierarchical parsing according to claim 1, characterized in that, During the cost breakdown stage of the expert review software, it supports initiating multi-expert collaborative meetings through a joint review plan; when the meeting starts, the data in the application budget form is automatically initialized to the expert review form, and an independent review copy is generated for each expert.

8. A software cost estimation system based on dynamic configuration and hierarchical parsing, characterized in that, include: The first configuration module is configured to dynamically load the corresponding configuration parameters and algorithms by identifying the evaluation method field. The second configuration module is configured to: construct a standardized cost parameter configuration system, including configuring the baseline value of the estimated function point, the weight coefficient of the estimated function point, and the scale adjustment factor; The third configuration module is configured to: based on the evaluation method, pre-set a tree-shaped budget configuration table, adopt hierarchical node design, construct a budget tree, and form a tree-shaped budget configuration and cost declaration / review header; The header dynamic loading module is configured to transfer the configuration authority of the tree-structured budget configuration and the cost application / review header to the backend, so that the backend can dynamically load the pre-set header template according to the enabled cost standard and synchronously inject the field attribute rules and calculation formula metadata. The parsing module is configured to: acquire unstructured documents uploaded by users, parse the unstructured documents and header templates using a large language model, and obtain the software cost details; The evaluation output module is configured to generate an evaluation report based on the cost details reviewed by expert review software.

9. A computer device, characterized in that, A processor, adapted to execute computer programs; A computer-readable storage medium storing a computer program, which, when executed by the processor, implements the steps of the software cost estimation method based on dynamic configuration and hierarchical resolution as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program adapted to be loaded by a processor and to execute the steps of the software cost estimation method based on dynamic configuration and hierarchical resolution as described in any one of claims 1-7.

Images