Method and system for automatically generating bond project implementation scheme

Abstract

The invention provides a bond project implementation scheme automatic generation method and system. The method comprises the steps that bond project key information input by a user side is acquired; according to the bond project key information, performing data matching on the bond project key information by using a pre-constructed bond business large model, and outputting a project initial scheme corresponding to the bond project key information; extracting business data information from the project initial scheme, and based on the business data information, performing data identification by using a matching algorithm to obtain a data table and business description corresponding to the business data information; according to the data table and the service description, generating a project implementation scheme corresponding to the bond project key information; according to the method, through automatic acquisition of bond project key information, intelligent matching and data identification, the accuracy of information extraction and the accuracy of scheme matching are remarkably improved, and a standardized project implementation scheme can be quickly generated, so that manual intervention is reduced, errors are reduced, and the efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method and system for automatically generating a bond project implementation plan. Background Art

[0002] At present, in the existing bond project management and review process, the key information collection, data matching and business data identification of bond projects still mainly rely on manual operations or simple rule-based algorithms. These traditional methods have significant deficiencies in information extraction accuracy, data matching accuracy and the efficiency of generating project implementation plans.

[0003] Specifically, there are multiple problems in the preparation of traditional special bond project implementation plans. First, the workload of preparing implementation plans is very large. The content involved in the project is complicated, including a large amount of financial data, market information, risk assessment, etc., and there are complex logical and computational relationships between these contents. Because these plans are complex and lengthy, they usually need to be written manually, resulting in extremely low work efficiency. For example, the plan may contain hundreds of pages of content. When writing, it is necessary to pay attention to the consistency of data in different parts at all times to ensure the accuracy and logical correctness of various types of information. This huge manual writing process is not only time-consuming, but also prone to errors. Especially during the modification process, errors and deviations in logical relationships are difficult to avoid, which seriously affects the quality and efficiency of the plan.

[0004] Secondly, although the existing bond business model can provide some rules and templates, it lacks flexibility and adaptability due to the lack of intelligent data matching and analysis, and cannot efficiently meet the needs of complex bond projects. The initial plan generation process of bond projects often relies on manual adjustments and modifications. Frequent manual operations not only increase the workload, but also lead to inefficiency and high error rates of the plan, especially in terms of business data extraction, table generation and accuracy of business descriptions, which still lack effective automation support.

[0005] In addition, the current special bond project plan modification process lacks effective inspection and verification methods, and the implementation plan content and data are prone to inconsistencies or logical errors. The manually written implementation plan has no unified review mechanism, which makes it difficult to ensure the quality of the implementation plan, affecting the project's application effect and actual implementation effect.

[0006] In summary, the degree of automation in bond project management and review in the existing technology is low, the information processing accuracy is insufficient, the plan matching is inaccurate, and the generated project implementation plans often have deviations. Summary of the invention

[0007] In order to solve the problems in the prior art that the key information collection, data matching and business data identification of bond projects are of low automation and often rely on manual intervention, resulting in inaccurate information extraction, inaccurate matching of business data and solutions, deviations in the generated project implementation plans, low efficiency and prone to errors, the present invention proposes a method for automatically generating a bond project implementation plan, comprising:

[0008] Obtain key information of bond projects input by the user;

[0009] According to the key information of the bond project, the pre-built bond business big model is used to perform data matching on the key information of the bond project, and an initial project plan corresponding to the key information of the bond project is output;

[0010] Extracting business data information from the initial project plan, and performing data identification using a matching algorithm based on the business data information to obtain a data table and business description corresponding to the business data information;

[0011] Generate a project implementation plan corresponding to the key information of the bond project according to the data table and the business description;

[0012] The bond business big model is obtained based on data extraction and text parsing of a preset bond model library.

[0013] Optionally, the bond business model includes the following construction process:

[0014] Using OCR technology, extract data from each project in the preset bond model library to obtain project content information corresponding to each project;

[0015] Performing text analysis on the project content information corresponding to each project to obtain semantic analysis information of each project;

[0016] According to the semantic analysis information of each project, the LLM big model is trained to obtain the bond business big model.

[0017] Optionally, the step of performing data matching on the key information of the bond project using a pre-built bond business model according to the key information of the bond project, and outputting an initial project plan corresponding to the key information of the bond project, includes:

[0018] Performing data matching on the bond project key information and the semantic parsing information of each project in the bond business macro model to obtain the project in the bond business macro model corresponding to the bond project key information;

[0019] According to the project corresponding to the key information of the bond project in the bond business big model, the project plan corresponding to the project is output as the initial project plan corresponding to the key information of the bond project.

[0020] Optionally, extracting business data information from the initial project plan includes:

[0021] Classify data according to the business content in the initial project plan to obtain business content corresponding to different business categories in the initial project plan;

[0022] According to the business contents corresponding to the different business categories, a preset information extraction model is used to extract data to obtain business data information corresponding to the initial project plan;

[0023] The business data information at least includes: business content category and business data elements.

[0024] Optionally, the performing data identification based on the business data information by using a matching algorithm to obtain a data table and a business description corresponding to the business data information includes:

[0025] Performing full word matching, word vector similarity matching, and unit feature matching on the business data information to obtain an operation formula operator corresponding to the business data information;

[0026] According to the operation formula operator, the business data information is matched with an operator to obtain an operator corresponding to the business data information;

[0027] According to the operator, a construction formula corresponding to the business data information is generated;

[0028] Generate a data table corresponding to the business data information according to the operation relationship in the construction formula corresponding to the business data information and the business logic sequence in the business data information;

[0029] Extracting target business indicators according to the data table corresponding to the business data information, and using the target business indicators as new business content in the initial project plan according to a preset indicator code;

[0030] Generate a corresponding business description according to the business data information and in a preset data format;

[0031] The data tables include one or more of the following: project annual investment plan table, project financing status table, project income calculation table, project cost calculation table, project principal and interest payment table, project capital balance table and project stress test table;

[0032] The target business indicators include one or more of the following: the total investment in the investment estimate table, the start and completion time in the basic project information, and the amounts of various types of financing funds in the project financing information;

[0033] The business description includes one or more of the following: the construction content of the project, the economic and social benefits of the project, and the construction objectives of the project.

[0034] Optionally, generating a construction formula corresponding to the business data information according to the operator includes:

[0035] According to the operator, selecting a formula matching algorithm corresponding to the business data information;

[0036] Generate a construction formula corresponding to the business data information according to a formula matching algorithm corresponding to the business data information;

[0037] The formula matching algorithm includes at least one or more of the following: an operator level priority algorithm, a formula matching priority algorithm, and a formula matching exclusion algorithm.

[0038] Optionally, after generating the project implementation plan corresponding to the key information of the bond project according to the data table and the business description, the method further includes:

[0039] Conducting a project evaluation on the project implementation plan based on the business data information in the project implementation plan to determine whether the project implementation plan needs to be adjusted;

[0040] If yes, the key information of the bond project input by the user is retrieved, and the data in the project implementation plan is adjusted according to the retrieved key information of the bond project to obtain a new project implementation plan.

[0041] Optionally, the key information of the bond project includes one or more of the following: project name, project region, project construction period, time to be put into operation, project type field, project investment field, user region, project investment information, project construction content and income description, special bond issuance period, income type and total project investment amount.

[0042] Based on the same inventive concept, the present invention also provides a bond project implementation plan automatic generation system, including:

[0043] The information acquisition module is used to obtain key information of the bond project input by the user;

[0044] An initial plan generation module is used to perform data matching on the key information of the bond project using a pre-built bond business model according to the key information of the bond project, and output an initial project plan corresponding to the key information of the bond project;

[0045] A data identification module is used to extract business data information from the initial project plan, and based on the business data information, perform data identification using a matching algorithm to obtain a data table and business description corresponding to the business data information;

[0046] A target solution generation module, used to generate a project implementation plan corresponding to the key information of the bond project according to the data table and the business description;

[0047] The bond business big model is obtained based on data extraction and text parsing of a preset bond model library.

[0048] Optionally, the generation system further includes: a model building module, which is used to:

[0049] Using OCR technology, extract data from each project in the preset bond model library to obtain project content information corresponding to each project;

[0050] Performing text analysis on the project content information corresponding to each project to obtain semantic analysis information of each project;

[0051] According to the semantic analysis information of each project, the LLM big model is trained to obtain the bond business big model.

[0052] Optionally, the initial solution generation module includes:

[0053] A semantic analysis submodule, used for matching the key information of the bond project with the semantic analysis information of each project in the bond business macromodel, and obtaining the project in the bond business macromodel corresponding to the key information of the bond project;

[0054] The project matching submodule is used to output a project plan corresponding to the project as an initial project plan corresponding to the key information of the bond project based on the project corresponding to the key information of the bond project in the bond business big model.

[0055] Optionally, the data identification module includes:

[0056] A business classification submodule, used to classify data according to the business content in the initial project plan, and obtain the business content corresponding to different business categories in the initial project plan;

[0057] A data extraction submodule is used to extract data according to the business content corresponding to the different business categories using a preset information extraction model to obtain business data information corresponding to the initial project plan;

[0058] The business data information at least includes: business content category and business data elements.

[0059] Optionally, the data identification module further includes:

[0060] An operator matching submodule, used to perform full word matching, word vector similarity matching and unit feature matching on the business data information to obtain an operation formula operator corresponding to the business data information;

[0061] An operator matching submodule, used to perform operator matching on the business data information according to the operation formula operator to obtain the operator corresponding to the business data information;

[0062] A formula generation submodule, used to generate a construction formula corresponding to the business data information according to the operator;

[0063] A table output submodule, used to generate a data table corresponding to the business data information according to the operation relationship in the construction formula corresponding to the business data information and the business logic sequence in the business data information;

[0064] A business content replacement submodule is used to extract target business indicators according to the data table corresponding to the business data information, and use the target business indicators as new business content in the initial plan of the project according to a preset indicator code;

[0065] A business description output submodule, used to generate a corresponding business description according to the business data information and a preset data format;

[0066] The data tables include one or more of the following: project annual investment plan table, project financing status table, project income calculation table, project cost calculation table, project principal and interest payment table, project capital balance table and project stress test table;

[0067] The target business indicators include one or more of the following: the total investment in the investment estimate table, the start and completion time in the basic project information, and the amounts of various types of financing funds in the project financing information;

[0068] The business description includes one or more of the following: the construction content of the project, the economic and social benefits of the project, and the construction objectives of the project.

[0069] Optionally, the formula generation submodule includes:

[0070] A priority matching unit, configured to select a formula matching algorithm corresponding to the business data information according to the operator;

[0071] A formula output unit, used to generate a construction formula corresponding to the business data information according to a formula matching algorithm corresponding to the business data information;

[0072] The formula matching algorithm includes at least one or more of the following: an operator level priority algorithm, a formula matching priority algorithm, and a formula matching exclusion algorithm.

[0073] Optionally, the generation system further includes: a scheme adjustment module, which is used to:

[0074] Conducting a project evaluation on the project implementation plan based on the business data information in the project implementation plan to determine whether the project implementation plan needs to be adjusted;

[0075] If yes, the key information of the bond project input by the user is retrieved, and the data in the project implementation plan is adjusted according to the retrieved key information of the bond project to obtain a new project implementation plan.

[0076] Optionally, the key information of the bond project includes one or more of the following: project name, project region, project construction period, time to be put into operation, project type field, project investment field, user region, project investment information, project construction content and income description, special bond issuance period, income type and total project investment amount.

[0077] In yet another aspect, the present invention further provides a computer device, comprising: one or more processors;

[0078] A memory for storing one or more programs;

[0079] When the one or more programs are executed by the one or more processors, a method for automatically generating a bond project implementation plan as described above is implemented.

[0080] On the other hand, the present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed, the method for automatically generating a bond project implementation plan as described above is implemented.

[0081] Compared with the prior art, the present invention has the following beneficial effects:

[0082] The present invention provides a method and system for automatically generating a bond project implementation plan, comprising: obtaining key information of a bond project input by a user end; performing data matching on the key information of the bond project using a pre-built bond business big model according to the key information of the bond project, and outputting an initial project plan corresponding to the key information of the bond project; extracting business data information from the initial project plan, and performing data identification using a matching algorithm based on the business data information to obtain a data table and a business description corresponding to the business data information; generating a project implementation plan corresponding to the key information of the bond project according to the data table and the business description; the present application extracts data from the key information of the bond project in an automated manner and performs intelligent matching and identification, thereby significantly improving the generation efficiency and accuracy of the bond project implementation plan, and utilizing the pre-built bond business big model and matching algorithm to quickly and accurately generate a preliminary plan, and automatically extract key business data and descriptions, thereby reducing manual intervention and errors, and finally, through the combination of data tables and business descriptions, a high-quality, standardized project implementation plan is generated, thereby improving the intelligence and automation level of bond project management. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 A schematic diagram of a flow chart of a method for automatically generating a bond project implementation plan provided by the present invention;

[0084] Figure 2 A flowchart of automatically generating a project implementation plan in a method for automatically generating a bond project implementation plan provided by the present invention;

[0085] Figure 3 The overall flow chart is automatically generated by the calculation formula in the automatic generation method of the bond project implementation plan provided by the present invention;

[0086] Figure 4 A schematic diagram of a formula tree structure in a method for automatically generating a bond project implementation plan provided by the present invention;

[0087] Figure 5 A schematic diagram of an implementation plan adjustment process of a method for automatically generating a bond project implementation plan provided by the present invention;

[0088] Figure 6 A schematic diagram of operator matching for a method for automatically generating a bond project implementation plan provided by the present invention;

[0089] Figure 7 A schematic diagram of the structural composition of a bond project implementation plan automatic generation system provided by the present invention. DETAILED DESCRIPTION

[0090] The present invention proposes a method, system, device and medium for automatically generating a bond project implementation plan. The specific implementation methods of the present invention are further described in detail below in conjunction with the accompanying drawings.

[0091] Embodiment 1:

[0092] The present invention provides a method for automatically generating a bond project implementation plan, the flow chart of which is as follows: Figure 1 As shown, including:

[0093] Step 1: Obtain key information of the bond project input by the user;

[0094] Step 2: Based on the key information of the bond project, use the pre-built bond business model to perform data matching on the key information of the bond project, and output the initial project plan corresponding to the key information of the bond project;

[0095] Step 3: Extract business data information from the initial project plan, and use matching algorithms to identify data based on the business data information to obtain data tables and business descriptions corresponding to the business data information;

[0096] Step 4: Generate a project implementation plan corresponding to the key information of the bond project based on the data table and business description;

[0097] Among them, the bond business big model is obtained based on data extraction and text parsing of the preset bond model library.

[0098] For example, the key information of the bond project in step 1 above may include one or more of the following: project name, project region, project construction period, time to be put into operation, project type field, project investment field, user region, project investment information, project construction content and income description, special bond issuance period, income type and total project investment amount.

[0099] In one implementation, Figure 2 As shown, the process of performing data matching on the key information of the bond project using the pre-built bond business model according to the key information of the bond project in step 2 above and outputting the initial project plan corresponding to the key information of the bond project may include:

[0100] Data matching is performed between the key information of the bond project and the semantic parsing information of each project in the bond business big model to obtain the project in the bond business big model corresponding to the key information of the bond project;

[0101] According to the project corresponding to the key information of the bond project in the bond business big model, the project plan corresponding to the project is output as the initial project plan corresponding to the key information of the bond project;

[0102] In this implementation, users interact with the system through interactive information entry. The system uses semantic recognition and information extraction technology to automatically obtain key information such as project name, location, construction period, time to be put into operation, and project type from the input provided by the user. Subsequently, the system accurately matches these key information with the semantic analysis information in the pre-built bond business big model, thereby automatically generating the initial plan for the bond project. This process not only significantly improves the efficiency of bond project plan generation, but also greatly reduces the need for manual intervention and reduces errors caused by manual input. Through semantic analysis and key information matching, it is possible to efficiently extract the core elements related to the project from the complex project data, ensuring that the generated initial plan is highly consistent with the project requirements and accurate. In addition, based on the matching ability of the big model, the plan generation process is more intelligent and flexible, and can quickly respond to the needs of different types of bond projects, improve the overall processing efficiency and plan accuracy. Finally, this implementation not only optimizes the preparation process of the bond project implementation plan, but also effectively improves the work quality, reduces the deviation caused by human operation errors, and ensures the high quality and consistency of the bond project plan.

[0103] In one implementation, the above bond business model may include the following construction process:

[0104] Using OCR technology, extract data from each project in the preset bond model library to obtain the project content information corresponding to each project;

[0105] Perform text analysis on the project content information corresponding to each project to obtain semantic analysis information of each project;

[0106] According to the semantic analysis information of each project, the LLM big model is trained to obtain the bond business big model;

[0107] In this implementation, by combining OCR technology with the LLM big model, the construction of the bond business big model was successfully realized, greatly improving the automated processing and intelligent analysis capabilities of bond project data. Specifically, the system first uses OCR technology (such as Baidu Paddle's Paddle-OCR4 model) to extract data from projects in the preset bond model library, and can automatically extract printed text and table information from files of various formats (such as PDF). The Paddle-OCR4 model performs particularly well in Chinese text parsing and table content recognition. It can accurately identify the table position, content and text coordinates. The parsing accuracy meets high standards and has high processing efficiency under GPU acceleration. On the basis of data extraction, the content information of each project is deeply parsed, and the LLM big model (such as Mass Spectrum Qingyan's Chat-GLM4 model) is used for semantic analysis and key information extraction. By collecting and analyzing a large number of historically public special bond project implementation plans and other disclosure documents, an efficient implementation plan model for different bond project scenarios and investment areas can be trained. This model can not only extract the core content of the bond project, but also self-learn and optimize according to the new implementation plan and user feedback, gradually improving the accuracy and adaptability of the model. Compared with traditional manual processing methods, this implementation method achieves efficient extraction and intelligent matching of bond project data through the synergy of OCR and LLM large models, greatly reducing manual intervention and improving the accuracy and efficiency of information processing. In addition, the self-learning ability of the model enables the system to continuously optimize in the ever-changing bond project needs, ensuring the high accuracy and reliability of the project implementation plan. This implementation method not only improves the processing efficiency of bond projects, but also greatly reduces the error rate in manual operations, ensuring the high quality and consistency of bond project implementation plans.

[0108] In one implementation, the process of extracting business data information from the initial project plan in step 3 above may include:

[0109] Classify data according to the business content in the initial project plan to obtain the business content corresponding to different business categories in the initial project plan;

[0110] According to the business content corresponding to different business categories, the preset information extraction model is used to extract data to obtain the business data information corresponding to the initial project plan;

[0111] The business data information may include at least: business content categories and business data elements;

[0112] In this implementation, the automation and intelligence level of data processing are greatly improved by accurately classifying and extracting the business content in the initial plan of the bond project. By classifying the content of different business categories in the initial plan of the project, the structure and characteristics of various business data can be automatically identified, laying a solid foundation for subsequent data extraction. Using the preset information extraction model, these business contents are further analyzed in depth, and business data information including business content categories and key data elements are automatically extracted. In this process, the efficient use of the information extraction model enables complex business data to be extracted quickly and accurately, ensuring the integrity and consistency of the data. Compared with the traditional manual extraction method, this data classification and extraction method based on the intelligent model not only greatly improves work efficiency, but also reduces errors and deviations caused by human operations. Especially when facing large-scale and complex bond project data, it can quickly and accurately complete the analysis and extraction of massive information, greatly improving the accuracy and intelligence level of data processing. In this way, various types of business information involved in the bond project can be efficiently structured, providing reliable data support for subsequent project analysis, risk assessment and program optimization. In addition, automated information extraction has strong adaptability and can be flexibly adjusted according to the characteristics and needs of different projects, thereby ensuring the efficient processing of various bond projects and the accurate generation of implementation plans.

[0113] In this implementation, if Figure 3 As shown, the above process of performing data identification based on the business data information by using a matching algorithm to obtain a data table and a business description corresponding to the business data information may include:

[0114] Perform full word matching, word vector similarity matching and unit feature matching on the business data information to obtain the operation formula operator corresponding to the business data information;

[0115] According to the operation formula operator, the business data information is matched with the operator to obtain the operator corresponding to the business data information;

[0116] Generate construction formulas corresponding to business data information based on operators;

[0117] Generate a data table corresponding to the business data information according to the operation relationship in the construction formula corresponding to the business data information and the business logic sequence in the business data information;

[0118] Extract target business indicators from the data table corresponding to the business data information, and use the target business indicators as new business content in the initial project plan according to the pre-set indicator code;

[0119] Generate corresponding business description according to business data information and preset data format;

[0120] The data tables include one or more of the following: annual project investment plan table, project financing status table, project income calculation table, project cost calculation table, project principal and interest payment table, project capital balance table and project stress test table;

[0121] The target business indicators include one or more of the following: the total investment in the investment estimate table, the start and completion time in the basic project information, and the amounts of various types of financing funds in the project financing information;

[0122] The business description includes one or more of the following: the construction content of the project, the economic and social benefits of the project, and the construction objectives of the project.

[0123] In this implementation, the process of generating a construction formula corresponding to the business data information according to the operator may include:

[0124] According to the operator, select the formula matching algorithm corresponding to the business data information;

[0125] Generate a construction formula corresponding to the business data information according to a formula matching algorithm corresponding to the business data information;

[0126] The formula matching algorithm includes at least one or more of the following: an operator level priority algorithm, a formula matching priority algorithm, and a formula matching exclusion algorithm.

[0127] This implementation includes three items: first, extracting data and identifying business data elements; second, generating formulas; and third, generating data tables and business descriptions, and generating implementation plans. The specific contents are as follows:

[0128] (1) Extract data and identify business data elements

[0129] You can choose to use Baidu PaddlePaddle's ERNIE model to classify the business content that users have completed and mark it as a business category. Through the pre-trained Baidu PaddlePaddle UIE model, you can extract business data information from the classified business content, identify the elements of different business data, and mark them as different business data elements.

[0130] (2) Generate formula

[0131] Based on the data accumulated from previous years of special bond projects, fine-tune the general large model to form a large model for the proprietary business field, strengthen the business data extraction effect in the generation of implementation plan fields, and follow the following steps: Figure 3The business data elements are intelligently matched in the described way to automatically generate the required formula. Specifically, according to the obtained formula template, the operators in the business formula template are matched with the business data elements extracted in step 2 through three methods: full word matching, word vector similarity matching, and unit feature matching. The matching methods and usage order of the three matching methods are shown in Table 1:

[0132] Table 1 Operator and business data information matching method and sequence diagram

[0133]

[0134] For those that the model fails to automatically identify, the matching results can be improved through interactive information entry;

[0135] Operator matching is performed based on the business feature rules of the calculation formula operator. Common operators are shown in Table 2:

[0136] Table 2 Schematic diagram of common operators

[0137]

[0138]

[0139] When the system recursively generates a complete calculation formula, the algorithm shown in Table 3 is used to implement recursive construction of the calculation formula to ensure the accuracy and efficiency of formula generation.

[0140] Table 3 Formula generation algorithm

[0141]

[0142] The process of recursively generating a formula tree is as follows:

[0143] 1) According to the above steps, a preliminary calculation formula is formed from the formula template.

[0144] 2) Starting from the top level, determine whether the operator has a corresponding value or formula, as well as a corresponding operator. If so, it is used as an operator in the formula. If there is no corresponding numerical formula, determine whether it can be composed of other operators through the historical project calculation formula knowledge base and domain model. Continue down level by level until all formulas can be determined as fixed operators.

[0145] 3) According to different levels, the formula is composed to form the final calculation formula. The formula tree is shown as follows: Figure 4 shown.

[0146] Generate data tables and business descriptions, and simultaneously generate implementation plans, which is mainly divided into three steps:

[0147] First, update the data table in the implementation plan: automatically generate data tables according to the business logic sequence and the operation relationship in the calculation formula. The tables automatically generated by the system mainly include project investment estimation table, project financing situation table, project income estimation table, project cost estimation table, project principal and interest payment table, project capital balance table, project stress test table, etc. After generating the relevant tables, the system automatically updates the data table in the implementation plan through the table synchronization tool (such as poi-tl), displays the data according to the table style set by the model, and automatically adjusts the column width, row height, etc. according to the data.

[0148] Second, adjust the business content of the implementation plan based on the data in the data table: extract the required business indicators from the generated data table, and replace the corresponding indicator content in the implementation plan model according to the indicator code pre-set in the implementation plan model. The main business indicators include the total investment in the investment estimate table, the start and completion time in the basic situation of the project, and the amount of various types of financing funds in the project financing situation. To ensure that the table data and business content description are consistent.

[0149] The third is to adjust the business description information in the implementation plan according to the project situation. The business domain model generates description information in a fixed format based on the project data information obtained from the system and various types of information obtained from the Internet or historical project databases. The business description that needs to be generated mainly includes the construction content of the project, the economic and social benefits of the project, the construction goals of the project, etc. Replace the corresponding business description code in the implementation plan model based on the generated content to ensure the consistency of the table data and the business content description.

[0150] Through the implementation of the above steps, the business content and business description of the implementation plan are automatically generated and pushed according to the matching implementation plan model, which comprehensively improves the automatic processing and intelligent analysis capabilities of the bond project business data information. First, according to the region to which the project belongs, the business description and business data are indexed, and the corresponding business content is divided and identified through index coding. This process adopts a multi-level matching algorithm and operator recognition mechanism, which can accurately identify and parse complex business data, ensure that each data element and its related operation logic are correctly reflected, and extract key operation formula operators from the project's business data information through methods such as full word matching, word vector similarity matching, and unit feature matching, and generate corresponding structured data tables and detailed business descriptions. Then, combined with the project investment field, region and other information entered by the user, and combined with the historical project database, a comprehensive search is carried out. According to the public time, region, project investment field and other dimensions of the historical implementation plan, the recommendation algorithm is used to generate specific business content, including project investment status, construction content and main income description, and these business contents are pushed to users. This intelligent recommendation not only improves the accuracy of the business content, but also ensures that the implementation plan can be dynamically adjusted and customized according to the specific needs of the project. Finally, the user further improves the business description and data based on the pushed business content, such as modifying the total investment amount of the project, adjusting the income type, and increasing the number of special bond issuance periods. Through the automated calculation and generation process of intelligent algorithms, the system can quickly adapt to the needs of different projects and generate business tables and descriptions that meet actual needs. This adaptive capability and efficient processing method not only greatly improves the processing efficiency of bond projects, but also reduces manual intervention and errors, ensuring the accuracy of data and the high quality of the plan. Overall, this technical solution not only optimizes the implementation plan preparation process of bond projects and improves work efficiency by combining intelligent recommendations with data matching algorithms, but also provides more efficient and intelligent data support for bond project management, and promotes project management to develop in the direction of higher accuracy and adaptive capabilities.

[0151] In one implementation, after the above-mentioned project implementation plan corresponding to the key information of the bond project is generated according to the data table and the business description, it may also include:

[0152] Based on the business data information in the project implementation plan, conduct a project evaluation on the project implementation plan to determine whether the project implementation plan needs to be adjusted;

[0153] If yes, re-acquire the key information of the bond project input by the user, and adjust the data in the project implementation plan according to the re-acquired key information of the bond project to obtain a new project implementation plan;

[0154] In this implementation, the system can automatically determine whether the solution meets the needs of the project and decide whether it needs to be adjusted by evaluating the business data information in the project implementation plan. Specifically, users can evaluate the pre-generated implementation plan to confirm the accuracy and feasibility of the solution. If it is confirmed that the solution does not need to be modified, the generated implementation plan can be directly downloaded to complete the preparation of the solution. However, if the data or content in the solution does not meet the actual needs of the project, the user can make further adjustments through interactive information entry. The system will judge the modification scenario based on the description content entered by the user, including adding model content or adjusting the numerical value in the model. After the judgment is completed, the system will automatically jump to the corresponding step, re-adjust the module content or numerical adjustment, and regenerate the data table based on the adjusted data, thereby updating the project implementation plan. This process ensures the flexibility and intelligence of the project implementation plan, and can make timely adjustments according to actual needs. Through this automated adjustment and optimization mechanism, not only the efficiency of solution generation and modification is improved, but also the accuracy and consistency of the solution are guaranteed. Users can quickly modify the implementation plan through interactive entry, avoiding solution deviations caused by manual operation or manual input errors. At the same time, through dynamic evaluation and adjustment, the system has achieved a high degree of adaptability in the project implementation plan, and can continuously optimize according to user needs and feedback, and finally generate an implementation plan that meets the actual situation of the project. Overall, this intelligent evaluation and adjustment mechanism has significantly improved the efficiency of the preparation of bond project implementation plans, reduced manual intervention, and improved the quality and adaptability of the plans, promoting the intelligent development of bond project management.

[0155] In summary, the present invention aims at the problems in the prior art that the key information collection, data matching and business data identification of bond projects are of low automation and often rely on manual intervention, resulting in inaccurate information extraction, inaccurate matching of business data and solutions, deviations in the generated project implementation plans, low efficiency and error-proneness. By integrating big data, artificial intelligence and business domain models, the present invention provides an automated and efficient method for generating bond project implementation plans. Figure 5As shown in the figure, first, the user provides the key information of the bond project through the interactive information entry function, and then uses the pre-built bond business model to match the data based on this information, so as to automatically generate the initial plan of the project. Based on the data extraction and text parsing of the preset bond model library, the bond business model can accurately identify the business elements related to the project, providing a reliable basis for the subsequent plan generation. After the initial plan is generated, the detailed data table and business description related to the project are obtained by extracting the business data information and combining the matching algorithm for data identification. This process not only improves the automatic processing capability of data, but also ensures the accuracy and structure of the data, avoiding errors and inefficiencies in manual processing. Subsequently, the system automatically generates a complete project implementation plan based on the generated data table and business description, and supports users to modify the business data and description according to actual needs. Through intelligent data recognition and calculation formula generation, the system can automatically generate business data tables related to the project, and synchronously update the implementation plan to ensure the consistency and accuracy of the business description and data content. Finally, users can optimize the project implementation plan according to the modification requirements, and improve the quality of the plan and the operability of implementation.

[0156] Embodiment 2:

[0157] A specific example is used to illustrate a method for automatically generating a bond project implementation plan provided by the present invention, comprising:

[0158] Step S1 matches the project implementation plan model through the business domain big model;

[0159] 1.1 Forming a model library through large models:

[0160] The implementation plan information of special bond projects disclosed in historical years was collected, the implementation plan content was extracted through OCR technology, and the implementation plan model was trained using the LLM large model. The final implementation plan model structure is shown in Table 4:

[0161] Table 4 Example of a special bond project implementation plan model

[0162]

[0163] 1.2 Model Matching

[0164] The user inputs "I want to build a parking project", and the system generates the following information and recommends it to the user based on historical data and the user's region and time. The user completes the content of the key information according to the situation of the proposed project to form the basic project information corresponding to Table 5.

[0165] Table 5 Example of basic information description of the project

[0166] Key information types Information content Project Name xx District Urban Management Parking Integration Project (Phase I) Project Area xx District, xx City Project construction period February 2023 - December 2024 Project investment areas City parking lot

[0167] The system identifies the user's registration area as xx District, xx City, and identifies the project as an urban parking project based on the investment area. Based on other basic information descriptions, it automatically matches in the system model library and finds the special bond project implementation plan model that best suits the project description, which is the urban parking model.

[0168] Step S2: Interactively improve business description and business data

[0169] According to the determined implementation plan and the investment field of the project, the system recommends the implementation plan of the project, including modules such as static investment situation, self-raised funds situation, income situation and cost situation, and automatically generates the recommended business plan content for each business module. The content of the business plan pushed by the system is shown in Table 6.

[0170] Table 6 Example of recommended content of business plan

[0171]

[0172] The user adjusts the description of the plan according to the pushed business plan content, such as modifying the total investment and income description. The modification example is shown in Table 7.

[0173] Table 7 Example of business content classification

[0174]

[0175]

[0176] The adjusted business content description is directly transmitted to the system in its entirety through interactive information entry.

[0177] Step S3 identifies the business data elements to generate data tables and business descriptions and generates implementation plans:

[0178] 3.1 Extract data and identify business data elements

[0179] Baidu PaddlePaddle’s ERNIE model is used to classify the user’s completed business content and mark various business content categories. Baidu PaddlePaddle’s UIE model is then used to extract business data information from the classified business content, identifying business data elements of seven types of business content, including project annual investment plans, project financing status, and project revenue estimates. The specific content identified is shown in Table 8.

[0180] Table 8 Schematic diagram of project business data extraction

[0181]

[0182] 3.2 Generating formula

[0183] 3.2.1 Use the business domain big model to confirm the applicable formula template

[0184] By using a large model in a professional field and intelligently matching business data elements, the formulas required for common business data elements are automatically generated based on various business data elements.

[0185] For example:

[0186] Annual investment plan of the project:

[0187] [Formula: Total investment = static total investment + construction period interest]

[0188] [Formula: Static total investment = project cost + other project construction costs + reserve fund]

[0189] [Formula: Construction period interest = issuance amount 1*issuance interest rate 1*issuance construction period 1+……]

[0190] Project income estimation:

[0191]

Formula: Total income = income 1 + income 2 + ...

[0192] [Formula: Income 1 = unit price * quantity * ...]

[0193] [Formula: Income 1 = Charge price * Quantity * Monthly subscription rate * …]

[0194] 3.2.2 Operation formula operator matching

[0195] According to the obtained formula template, the operators in the business formula template are matched with the business data elements extracted in step S2 through three methods: full word matching, word vector similarity matching, and unit feature matching. The matching diagram is shown in the figure below: Figure 6 .

[0196] 3.2.3 Operator Matching

[0197] The system matches operators according to the business characteristics of the calculation formula operators by taking the operators at the corresponding positions in the operator matching formula. The following operator results are generated.

[0198] Annual investment plan of the project:

[0199] [Formula: Static total investment = project cost + other project construction costs + reserve fund]

[0200] Total investment: RMB 182.8175 million

[0201] Project cost: 80 million yuan - Addition operator

[0202] Other construction costs: RMB 40 million - Addition operator

[0203] Reserve: 72.8175 million yuan - Addition operator

[0204] Project income estimation:

[0205]

Formula: Total income = income 1 + income 2 + ...

[0206] [Formula: Income 1 = unit price * quantity * ...]

[0207] [Formula: Income 1 = Charge price * Quantity * Monthly subscription rate * …]

[0208] Parking Lot Revenue: - Addition Operator

[0209] 2000 parking spaces - multiplication operator

[0210] Monthly fee: 3,000 yuan / month - Multiplication operator

[0211] Monthly subscription rate 35% - multiplication operator

[0212] Fast charging pile service fee - addition operator

[0213] 800 fast charging piles - multiplication operator

[0214] 350 days of operation throughout the year - multiplication operator

[0215] The charge is RMB 0.60 per kWh - multiplication operator

[0216] Average charging 80 degrees - multiplication operator

[0217] 3.2.4 Calculation formula construction

[0218] Through the algorithm, the composition of operators and operators is identified, and recursion is performed level by level to finally form a complete formula tree.

[0219] Annual investment plan of the project:

[0220] [Formula 1: Static total investment = project cost + other project construction costs + reserve fund]

[0221] [Formula 1: Total investment = static total investment + issuance amount * bond term * interest rate]

[0222] Project income estimation:

[0223] [Formula 1: Operating income = parking lot income + fast charging pile income]

[0224] [Formula 2: Parking lot income = parking spaces * monthly fee (3,000 yuan / month) * 12 (months)) * monthly rate]

[0225] [Formula 3: Fast charging pile income = fast charging pile * annual operation time * charging standard * average charging]

[0226] 3.3 Generate data tables and business descriptions, and simultaneously generate implementation plans

[0227] The corresponding data tables are generated according to the generated calculation formula. The tables automatically generated by the system mainly include the project annual investment plan table, project financing situation table, project income calculation table, project cost calculation table, project principal and interest payment table, project capital balance table, project stress test table, etc. As shown in the figure, here, according to the implementation plan model of xx city, the schematic diagram of the investment plan table and the project operation income forecast table is generated, as shown in Table 9-Table 10.

[0228] Table 9 Project annual investment plan Unit: 10,000 yuan

[0229] Serial number project 2023 2024 total 1 Construction and installation engineering 8,000.00 0.00 8,000.00 2 Equipment investment 4,000.00 4,000.00 3 Reserve 7,281.75 7,281.75 4 Other construction project costs 0.00 0.00 one Static total investment 8,000.00 11,281.75 19,281.75 two Construction period interest 160.00 520.00 680.00 three Bond issuance costs 8.00 6.00 14.00 Four Total Investment 8,168.00 11,807.75 19,975.75

[0230] Table 10 Project operating income forecast

[0231] Unit: Ten thousand yuan

[0232] Serial number Income Category unit 2025 2026 2027 2028 2029 2030 1 Parking lot revenue Ten thousand yuan 2,520.00 2,520.00 2,520.00 2,520.00 2,520.00 2,520.00 Off-street parking indivual 2,000.00 2,000,00 2,000.00 2,000.00 2,000.00 2,000.00 TOLL Yuan / month 3,000.00 3,000.00 3,000.00 3,000.00 3,000.00 3,000.00 Monthly rate ％ 35.00 35.00 35.00 35.00 35.00 35.00 Month moon 12.00 12.00 12.00 12.00 12.00 12.00 2 Fast charging pile service fee Ten thousand yuan 3,810.00 3.84 3,810.00 3,840.00 3,840.00 3,840.00 Fast charging pile indivual 80.00 800.00 80.00 80.00 80.00 80.00 Charging Standards Yuan / degree 0.60 0.60 0.60 0.60 0.60 0.60 Average daily charge Degree / day 80.00 80.00 80.00 80.00 80.00 80.00 Total income Ten thousand yuan 2,520.38 2,523.84 2,520.38 2,520.38 2,520.38 2,520.38

[0233] The system automatically generates numerical indicators of the business content in the implementation plan based on the generated data table. The main business indicators include the total investment in the investment estimate table, the start and completion time in the basic project situation, and the amount of various types of financing funds in the project financing situation.

[0234] Replace the corresponding content according to the parameter indicators in the parking lot model.

[0235] The following are some examples of adjustments:

[0236] Before adjustment:

[0237] Project name: {{0001}}

[0238] Project unit: {{0002}}

[0239] Project area: {{0003}}

[0240] After adjustment:

[0241] Project name: xx District Urban Management Parking Integration Project (Phase I)

[0242] Project unit: xx District Urban Management Committee of xx City

[0243] Project Division: xx District, xx City

[0244] Based on the historical special bond project data, a business description knowledge base system is formed. By providing basic information and descriptions of the basic situation of the project, the big model can generate the required business description information of various types. The business description that needs to be generated mainly includes the construction content of the project, the economic and social benefits of the project, and the construction goals of the project.

[0245] The following are some examples of adjustments:

[0246] Before adjustment:

[0247] Project size: {{1001}}

[0248] After adjustment:

[0249] Project scale: This project includes 6 sub-projects, including the construction of a smart parking command center for the entire district, with a total construction area of ​​1,685 square meters, and the intelligent transformation of roadside parking on 99 sections of the district, setting up 3,000 roadside parking spaces, and building 800 car charging piles.

[0250] Step S4: Improve the implementation plan

[0251] The user can evaluate the pre-generated implementation plan to confirm whether the business data and content of the plan need to be adjusted. If modification is required, the user can continue to adjust the data through interactive information entry. The system will automatically jump to step S2 for re-identification and generation. If no modification is required, the generated implementation plan can be downloaded to complete the generation of the implementation plan.

[0252] The user inputs "the unit price of parking lot income is uniformly changed to 3,500 yuan / month, and the monthly subscription rate is changed to 40%". The system determines the content that the user needs to adjust based on the semantics of the user's input content, updates the original business description, and forms a new business description information, as shown in Table 11.

[0253] Table 11 Project business description update

[0254]

[0255]

[0256] After the update, the system automatically jumps to step S3, continues the corresponding analysis and generation, and finally updates to the implementation plan, as shown in Table 12.

[0257] Table 12 Adjusted Project Operating Income Forecast Unit: 10,000 Yuan

[0258] Serial number Income Category unit 2025 2026 2027 2028 2029 2030 1 Parking lot revenue Ten thousand yuan 3,360.00 3,360.00 3,360.00 3,360.00 3,360.00 3,360.00 Off-street parking indivual 2,000.00 2,000.00 2,000.00 2,000.00 2,000.00 2,000.00 TOLL Yuan / month 3,500.00 3,500.00 3,500.00 3,500.00 3,500.00 3,500.00 Monthly rate ％ 40.00 40.00 40.00 40.00 40.00 40.00 Month moon 12.00 12.00 12.00 12.00 12.00 12.00 2 Fast charging pile service fee Ten thousand yuan 3,840.00 3.84 3,840.00 3,840.00 3,840.00 3,840.00 Fast charging pile indivual 80.00 800.00 80.00 80.00 80.00 80.00 Charging Standards Yuan / degree 0.60 0.60 0.60 0.60 0.60 0.60 Average daily charge Degree / day 80.00 80.00 80.00 80.00 80.00 80.00 Total income Ten thousand yuan 3,360.38 3,363.84 3,360.38 3,360.38 3,360.38 3,360,38

[0259] This example illustrates the technical advantages of the method for automatically generating a bond project implementation plan provided by the present invention in the process of writing a bond project implementation plan. During the specific implementation process, the system's automated calculation and data processing not only improve the efficiency of plan writing and reduce manual intervention, but also ensure that the generated implementation plan has high accuracy and logical consistency. Ultimately, this intelligent method makes the writing, modification, and adjustment of bond project implementation plans more efficient and accurate, greatly optimizing the problems of low efficiency and frequent errors in traditional manual operations. Therefore, the technical effect of the present invention in practical applications not only optimizes the plan writing process, but also improves the quality and executability of the plan.

[0260] Embodiment 3:

[0261] The present invention based on the same inventive concept also provides a bond project implementation plan automatic generation system, the structural composition diagram is as follows Figure 7 As shown, including:

[0262] The information acquisition module is used to obtain key information of the bond project input by the user;

[0263] The initial plan generation module is used to match the key information of the bond project with the pre-built bond business model and output the initial plan of the project corresponding to the key information of the bond project;

[0264] The data identification module is used to extract business data information from the initial project plan, and based on the business data information, use the matching algorithm to perform data identification to obtain the data table and business description corresponding to the business data information;

[0265] The target plan generation module is used to generate a project implementation plan corresponding to the key information of the bond project based on the data table and business description;

[0266] Among them, the bond business big model is obtained based on data extraction and text parsing of the preset bond model library.

[0267] For example, the key information of the bond project mentioned above may include one or more of the following: project name, project region, project construction period, time to be put into operation, project type field, project investment field, user region, project investment information, project construction content and income description, special bond issuance period, income type and total project investment amount.

[0268] In one implementation, the initial solution generation module may include:

[0269] The semantic analysis submodule is used to match the key information of the bond project with the semantic analysis information of each project in the bond business model, and obtain the project in the bond business model that corresponds to the key information of the bond project;

[0270] The project matching submodule is used to output the project plan corresponding to the project as the initial project plan corresponding to the key information of the bond project according to the project corresponding to the key information of the bond project in the bond business big model.

[0271] In one implementation, the generation system may further include: a model building module, specifically configured to:

[0272] Using OCR technology, extract data from each project in the preset bond model library to obtain the project content information corresponding to each project;

[0273] Perform text analysis on the project content information corresponding to each project to obtain semantic analysis information of each project;

[0274] Based on the semantic analysis information of each project, the LLM big model is trained to obtain the bond business big model.

[0275] In one implementation, the data identification module may include:

[0276] The business classification submodule is used to classify data according to the business content in the initial project plan, and obtain the business content corresponding to different business categories in the initial project plan;

[0277] The data extraction submodule is used to extract data according to the business content corresponding to different business categories using a preset information extraction model to obtain business data information corresponding to the initial project plan;

[0278] Among them, the business data information at least includes: business content category and business data elements.

[0279] In one implementation, the data identification module may further include:

[0280] The operator matching submodule is used to perform full word matching, word vector similarity matching and unit feature matching on the business data information to obtain the operation formula operator corresponding to the business data information;

[0281] The operator matching submodule is used to match the operator of the business data information according to the operation formula operator to obtain the operator corresponding to the business data information;

[0282] The formula generation submodule is used to generate the construction formula corresponding to the business data information according to the operator;

[0283] The table output submodule is used to generate a data table corresponding to the business data information according to the operation relationship in the construction formula corresponding to the business data information and the business logic sequence in the business data information;

[0284] The business content replacement submodule is used to extract the target business indicators according to the data table corresponding to the business data information, and use the target business indicators as the new business content in the initial project plan according to the pre-set indicator code;

[0285] The business description output submodule is used to generate corresponding business descriptions according to the business data information and the preset data format;

[0286] The data tables include one or more of the following: annual project investment plan table, project financing status table, project income calculation table, project cost calculation table, project principal and interest payment table, project capital balance table and project stress test table;

[0287] The target business indicators include one or more of the following: the total investment in the investment estimate table, the start and completion time in the basic project information, and the amounts of various types of financing funds in the project financing information;

[0288] The business description includes one or more of the following: the construction content of the project, the economic and social benefits of the project, and the construction objectives of the project.

[0289] In one implementation, the above formula generation submodule may include:

[0290] A priority matching unit, used to select a formula matching algorithm corresponding to the business data information according to the operator;

[0291] A formula output unit, used to generate a construction formula corresponding to the business data information according to a formula matching algorithm corresponding to the business data information;

[0292] The formula matching algorithm includes at least one or more of the following: an operator level priority algorithm, a formula matching priority algorithm, and a formula matching exclusion algorithm.

[0293] In one implementation, the above generation system may further include: a scheme adjustment module, specifically configured to:

[0294] Based on the business data information in the project implementation plan, conduct a project evaluation on the project implementation plan to determine whether the project implementation plan needs to be adjusted;

[0295] If yes, the key information of the bond project input by the user is retrieved, and the data in the project implementation plan is adjusted according to the retrieved key information of the bond project to obtain a new project implementation plan.

[0296] Embodiment 4:

[0297] Based on the same inventive concept, the present invention also provides a computer device, which includes a processor and a memory, the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in a computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions in a computer storage medium to implement the corresponding method flow or corresponding function, so as to implement the steps of the automatic generation method of a bond project implementation plan in the above embodiment.

[0298] Embodiment 5:

[0299] Based on the same inventive concept, the present invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device for storing programs and data. It is understandable that the computer-readable storage medium here can include both built-in storage media in a computer device and, of course, extended storage media supported by the computer device. The computer-readable storage medium provides a storage space, which stores the operating system of the terminal. In addition, one or more instructions suitable for being loaded and executed by a processor are also stored in the storage space, and these instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the steps of a method for automatically generating a bond project implementation plan in the above embodiment.

[0300] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0301] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0302] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0303] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0304] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit its protection scope. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that after reading the present invention, those skilled in the art can still make various changes, modifications or equivalent substitutions to the specific implementation methods of the application, but these changes, modifications or equivalent substitutions are all within the protection scope of the claims to be approved.

Claims

1. A method for automatically generating a bond project implementation plan, characterized in that: include: Obtain key information of bond projects input by the user; According to the key information of the bond project, the pre-built bond business big model is used to perform data matching on the key information of the bond project, and an initial project plan corresponding to the key information of the bond project is output; Extracting business data information from the initial project plan, and performing data identification using a matching algorithm based on the business data information to obtain a data table and business description corresponding to the business data information; Generate a project implementation plan corresponding to the key information of the bond project according to the data table and the business description; The bond business big model is obtained by extracting data and parsing text from a preset bond model library.

2. The method according to claim 1, characterized in that The bond business model includes the following construction process: Using OCR technology, extract data from each project in the preset bond model library to obtain project content information corresponding to each project; Performing text analysis on the project content information corresponding to each project to obtain semantic analysis information of each project; According to the semantic analysis information of each project, the LLM big model is trained to obtain the bond business big model.

3. The method according to claim 2, characterized in that According to the key information of the bond project, the pre-built bond business model is used to perform data matching on the key information of the bond project, and an initial project plan corresponding to the key information of the bond project is output, including: Performing data matching on the bond project key information and the semantic parsing information of each project in the bond business macro model to obtain the project in the bond business macro model corresponding to the bond project key information; According to the project corresponding to the key information of the bond project in the bond business big model, the project plan corresponding to the project is output as the initial project plan corresponding to the key information of the bond project.

4. The method according to claim 1, characterized in that The extracting of business data information from the initial project plan includes: Classify data according to the business content in the initial project plan to obtain business content corresponding to different business categories in the initial project plan; According to the business contents corresponding to the different business categories, a preset information extraction model is used to extract data to obtain business data information corresponding to the initial project plan; The business data information at least includes: business content category and business data elements.

5. The method according to claim 4, characterized in that The method of performing data identification based on the business data information by using a matching algorithm to obtain a data table and a business description corresponding to the business data information includes: Performing full word matching, word vector similarity matching, and unit feature matching on the business data information to obtain an operation formula operator corresponding to the business data information; According to the operation formula operator, the business data information is matched with an operator to obtain an operator corresponding to the business data information; According to the operator, a construction formula corresponding to the business data information is generated; Generate a data table corresponding to the business data information according to the operation relationship in the construction formula corresponding to the business data information and the business logic sequence in the business data information; Extracting target business indicators according to the data table corresponding to the business data information, and using the target business indicators as new business content in the initial project plan according to a preset indicator code; Generate a corresponding business description according to the business data information and in a preset data format; The data tables include one or more of the following: project annual investment plan table, project financing status table, project income calculation table, project cost calculation table, project principal and interest payment table, project capital balance table and project stress test table; The target business indicators include one or more of the following: the total investment in the investment estimate table, the start and completion time in the basic project information, and the amounts of various types of financing funds in the project financing information; The business description includes one or more of the following: the construction content of the project, the economic and social benefits of the project, and the construction objectives of the project.

6. The method according to claim 5, characterized in that Generating a construction formula corresponding to the business data information according to the operator includes: According to the operator, selecting a formula matching algorithm corresponding to the business data information; Generate a construction formula corresponding to the business data information according to a formula matching algorithm corresponding to the business data information; The formula matching algorithm includes at least one or more of the following: an operator level priority algorithm, a formula matching priority algorithm, and a formula matching exclusion algorithm.

7. The method according to claim 1, characterized in that After generating the project implementation plan corresponding to the key information of the bond project according to the data table and the business description, the method further includes: Conducting a project evaluation on the project implementation plan based on the business data information in the project implementation plan to determine whether the project implementation plan needs to be adjusted; If yes, the key information of the bond project input by the user is retrieved, and the data in the project implementation plan is adjusted according to the retrieved key information of the bond project to obtain a new project implementation plan.

8. The method according to claim 1, characterized in that The key information of the bond project includes one or more of the following: project name, project region, project construction period, time to be put into operation, project type field, project investment field, user region, project investment information, project construction content and income description, special bond issuance period, income type and total project investment amount.

9. A bond project implementation plan automatic generation system, characterized in that: include: The information acquisition module is used to obtain key information of the bond project input by the user; An initial plan generation module is used to perform data matching on the key information of the bond project using a pre-built bond business model according to the key information of the bond project, and output an initial project plan corresponding to the key information of the bond project; A data identification module is used to extract business data information from the initial project plan, and based on the business data information, perform data identification using a matching algorithm to obtain a data table and business description corresponding to the business data information; A target solution generation module, used to generate a project implementation plan corresponding to the key information of the bond project according to the data table and the business description; The bond business big model is obtained based on data extraction and text parsing of a preset bond model library.

10. The system according to claim 9, characterized in that The generation system further comprises: a model building module, which is used to: Using OCR technology, extract data from each project in the preset bond model library to obtain project content information corresponding to each project; Performing text analysis on the project content information corresponding to each project to obtain semantic analysis information of each project; According to the semantic analysis information of each project, the LLM big model is trained to obtain the bond business big model.

Images