Method, system and device for improving quality of business document based on artificial intelligence, processor and readable storage medium thereof

Abstract

The present application relates to a kind of based on artificial intelligence and realizes the method for improving investment business document quality, comprising the following steps: constructing data base module, interface multi-source data and provide template resource library;Through four synergistic mechanisms in quality collaborative module, investment business document is executed from pre-generation, writing, auditing to the whole process quality control of flow transfer;Through application interaction module, quality control capability is embedded in document editing and business flow whole process.Adopted the method, system, device, processor and its computer readable storage medium of the present application based on artificial intelligence and realizes the method for improving investment business document quality, four mechanisms are not simply superimposed, but interdependent, organic whole of data intercommunication.The present application focuses on how to systematically improve the quality of investment document in the whole business life cycle, and the technical concept is the closed-loop control system of "generation-writing-auditing-flow transfer".

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence, and more particularly to the field of financial technology, specifically to a method, system, device, processor, and computer-readable storage medium for improving the quality of investment banking business documents based on artificial intelligence. Background Technology

[0002] Investment banking documents (prospectuses, project proposals, issuance documents, etc.) are core deliverables of investment banking, characterized by high professionalism, stringent compliance requirements, and complex data relationships. Traditional document processing relies on manual drafting and review, which suffers from inefficiency, difficulty in preventing errors and omissions, and inconsistent standards. Furthermore, existing tools have fragmented functions, making it difficult to form a systematic quality assurance system.

[0003] like Figure 1 As shown, here is an analysis of the difficulties in writing investment banking business documents: The writing and maintenance of investment banking business documents is essentially a process of "completing a logically sound, data-accurate, perfectly formatted, and absolutely compliant business document through collaboration among multiple departments within a very short period of time." The main pain points include: 1. Content quality and data accuracy Data accuracy and consistency: Financial and market data need to be collected from multiple sources (company reports, Wind, Bloomberg, industry think tanks). Project-related and customer information are scattered in various places, requiring switching between different locations to obtain data when writing reports, which is difficult to acquire and maintain.

[0004] Precision and compliance of language: Strict legal and financial terminology must be used, ensuring investors can understand the language while avoiding exaggeration. For example, when describing risk factors, the wording must be watertight.

[0005] 2. Collaboration and Version Management The chaos of multi-threaded collaboration: A transaction team typically involves investment banking, compliance, and corporate management. Everyone is working on the same document simultaneously, and it is easy to make mistakes when merging revisions from all parties, often resulting in situations such as "overwriting someone else's content" or "missing important annotations."

[0006] Version confusion and difficulty in tracing the source: If the current document is used, there will be multiple versions being modified in parallel, making merging and tracing the source extremely difficult.

[0007] High communication costs: A significant amount of time is wasted explaining and revising content. For example, when a document is in the workflow, the entire process is often restarted due to issues with the document's content, resulting in low communication efficiency.

[0008] 3. Technology and Format Formatting issues: Investment banking documents have extremely high requirements for formatting (font, line spacing, page numbers, heading numbering, table of contents). Due to multi-person collaboration, pasting content can easily lead to formatting problems. Many business documents have standard templates that are updated regularly. Modifying older versions can result in differences in formatting and content, making identification difficult and increasing management costs.

[0009] Performance issues with large files: Documents containing numerous high-resolution charts and extensive chapters are huge in size, which can easily cause computers to lag or even crash, resulting in the loss of unsaved work.

[0010] 4. Compliance and Risk Control Regulatory rule changes: Different exchanges have different disclosure rules, and policies are constantly being updated. It is essential to ensure that documents comply with the latest regulatory requirements. This necessitates real-time updates to the template information used by users and global synchronization with relevant colleagues to guarantee that the basic templates meet the latest rules.

[0011] Document retention and retrieval: Every conclusion in the document requires supporting documentation. Maintaining the correspondence between these documentations and the document is a huge undertaking. Furthermore, document modification tracking is also crucial for quickly locating revision records and identifying issues.

[0012] Process flow review: During the workflow of investment banking documents, modification and review comments are often made through multiple channels such as email annotations and verbal communication, making them easily fragmented and overlooked. There is a risk that the entire process will be restarted due to adjustments in document content, significantly slowing down team productivity. Summary of the Invention

[0013] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method, system, device, processor and its computer-readable storage medium for improving the quality of investment banking business documents based on artificial intelligence, which meets the requirements of high data accuracy, good consistency, high precision and wide applicability.

[0014] To achieve the above objectives, the present invention provides a method, system, apparatus, processor, and computer-readable storage medium for improving the quality of investment banking business documents based on artificial intelligence, as follows: The main feature of this method for improving the quality of investment banking business documents based on artificial intelligence is that the method includes the following steps: (1) Construct a data infrastructure module, connect to multi-source data and provide a template resource library to provide data support for document quality; (2) Through the four collaborative mechanisms in the quality collaboration module, the entire process of quality control of investment banking business documents is carried out from pre-generation, writing, review to circulation; the four collaborative mechanisms include template filling-data source collaboration mechanism, writing-search-question and answer collaboration mechanism, multi-dimensional document intelligent review collaboration mechanism, and document-process-tool collaboration mechanism. (3) By applying the interaction module, the quality control capability is embedded into the entire process of document editing and business flow.

[0015] Preferably, the template filling-data source collaboration mechanism specifically includes the following steps: Establish a mapping relationship between multi-source data fields and templates in advance to form a field-template matching matrix; After selecting a template, the system automatically extracts matching fields from multi-source data and intelligently fills them into the corresponding positions in the template to pre-generate the document; Data consistency verification is performed during the data filling process; Once the population is complete, a population report will be generated, showing the populated fields, the matching rate, and a list of unmatched fields.

[0016] Preferably, the data consistency check specifically verifies whether the field type, format, and value range meet the template requirements.

[0017] Preferably, the aforementioned writing-search-question-answering collaborative mechanism specifically includes the following steps: During the document writing process, the write-and-search function is triggered to retrieve internal business data from the project library, customer library, personnel library, and knowledge base in real time and directly insert the search results; With one click, you can access the AI ​​integration unit to obtain real-time answers to professional questions and directly insert the question-and-answer results to form a continuous workflow chain.

[0018] Preferably, the multi-dimensional intelligent document review and collaboration mechanism specifically includes the following steps: The document undergoes multi-dimensional parallel review, including intelligent document verification, regulatory sensitive word checking, general document check, and document comparison. After the document is approved, a quality tag is automatically added to it, which includes compliance status and risk level. The review results are aggregated and displayed, and linked with the knowledge base, Q&A, and search to form a closed loop for review.

[0019] Preferably, the aforementioned regulatory sensitive word check specifically includes: It detects promises and absolute terms, and provides compliance alternatives for detected non-compliant statements.

[0020] Preferably, the intelligent document review specifically involves checking the document format, structure, and content completeness; the general document check specifically involves detecting consistency of company names in headers and footers, typos, and grammatical issues; and the document comparison specifically involves automatically marking the differences between the target document and the comparison document to assist in review decisions.

[0021] Preferably, the document-process-tool collaboration mechanism specifically includes the following steps: The document visually displays the process association status. Documents with associated processes show clickable process titles, while documents without associated processes provide an entry point for process association. Implement access control for documents that have been linked to processes. If the business process moves to a specific node, the document permissions will be automatically adjusted.

[0022] Preferably, the access control is linked to business process nodes. When the business process flows to the corresponding node, the system automatically adjusts the document's viewing, annotation, and editing permissions, and records the modification operations throughout the document's entire lifecycle.

[0023] The main feature of this system for improving the quality of investment banking business documents based on artificial intelligence is that the system includes: The data infrastructure module is used to connect to multi-source data and provide a template resource library to provide data support for document quality. The quality collaboration module, connected to the aforementioned data infrastructure module, includes a template filling-data source collaboration mechanism, a writing-search-question answer collaboration mechanism, a multi-dimensional document intelligent review collaboration mechanism, and a document-process-tool collaboration mechanism, used for full-process quality control of documents; The application interaction module is connected to the quality collaboration module to embed quality assurance capabilities into the entire document editing and circulation process.

[0024] Preferably, the data infrastructure module includes a multi-source data access unit, a template resource library unit, and an AI integration unit. The multi-source data access unit is used to connect to multiple business data sources covering project information, customer information, personnel information, and draft data. The template resource library unit is used to provide templates covering business scenarios such as project initiation, issuance, and ongoing maintenance. The AI ​​integration unit is used to support intelligent question answering, document polishing, and document interpretation.

[0025] The main feature of this device for improving the quality of investment banking business documents based on artificial intelligence is that the device includes: A processor is configured to execute computer-executable instructions; The memory stores one or more computer-executable instructions, which, when executed by the processor, implement the steps of the aforementioned method for improving the quality of investment banking business documents based on artificial intelligence.

[0026] The processor for improving the quality of investment banking business documents based on artificial intelligence is characterized in that the processor is configured to execute computer-executable instructions, and when the computer-executable instructions are executed by the processor, the various steps of the above-mentioned method for improving the quality of investment banking business documents based on artificial intelligence are implemented.

[0027] The main feature of this computer-readable storage medium is that it stores a computer program thereon, which can be executed by a processor to implement the various steps of the above-described method for improving the quality of investment banking business documents based on artificial intelligence.

[0028] This invention employs a method, system, apparatus, processor, and computer-readable storage medium based on artificial intelligence to improve the quality of investment banking business documents. It constructs a closed-loop quality control system covering the entire process, from document generation to writing assistance, intelligent review, and finally, circulation and collaboration. These four mechanisms are not simply superimposed but rather form an interdependent and interconnected organic whole. This invention focuses on how to systematically improve the quality of investment banking documents throughout the entire business lifecycle; its technical concept is a closed-loop control system of "generation-writing-review-circulation." Attached Figure Description

[0029] Figure 1 This diagram illustrates the challenges of investment banking documents in the method for improving the quality of investment banking business documents based on artificial intelligence, as presented in this invention.

[0030] Figure 2 The present invention presents a system architecture and flowchart of four collaborative mechanisms for the method of improving the quality of investment banking business documents based on artificial intelligence.

[0031] Figure 3 This is a schematic diagram of the investment banking business process of the method for improving the quality of investment banking business documents based on artificial intelligence according to the present invention.

[0032] Figure 4 This is a schematic diagram illustrating the implementation effect of the method for improving the quality of investment banking business documents based on artificial intelligence according to the present invention. Detailed Implementation

[0033] To more clearly describe the technical content of the present invention, the following description is provided in conjunction with specific embodiments.

[0034] The method for improving the quality of investment banking business documents based on artificial intelligence according to the present invention includes the following steps: (1) Construct a data infrastructure module, connect to multi-source data and provide a template resource library to provide data support for document quality; (2) Through the four collaborative mechanisms in the quality collaboration module, the entire process of quality control of investment banking business documents is carried out from pre-generation, writing, review to circulation; the four collaborative mechanisms include template filling-data source collaboration mechanism, writing-search-question and answer collaboration mechanism, multi-dimensional document intelligent review collaboration mechanism, and document-process-tool collaboration mechanism. (3) By applying the interaction module, the quality control capability is embedded into the entire process of document editing and business flow.

[0035] As a preferred embodiment of the present invention, the template filling-data source collaboration mechanism specifically includes the following steps: Establish a mapping relationship between multi-source data fields and templates in advance to form a field-template matching matrix; After selecting a template, the system automatically extracts matching fields from multi-source data and intelligently fills them into the corresponding positions in the template to pre-generate the document; Data consistency verification is performed during the data filling process; Once the population is complete, a population report will be generated, showing the populated fields, the matching rate, and a list of unmatched fields.

[0036] In a preferred embodiment of the present invention, the data consistency verification specifically verifies whether the field type, format, and value range meet the template requirements.

[0037] As a preferred embodiment of the present invention, the writing-search-question-answering collaborative mechanism specifically includes the following steps: During the document writing process, the write-and-search function is triggered to retrieve internal business data from the project library, customer library, personnel library, and knowledge base in real time and directly insert the search results; With one click, you can access the AI ​​integration unit to obtain real-time answers to professional questions and directly insert the question-and-answer results to form a continuous workflow chain.

[0038] As a preferred embodiment of the present invention, the multi-dimensional document intelligent review and collaboration mechanism specifically includes the following steps: The document undergoes multi-dimensional parallel review, including intelligent document verification, regulatory sensitive word checking, general document check, and document comparison. After the document is approved, a quality tag is automatically added to it, which includes compliance status and risk level. The review results are aggregated and displayed, and linked with the knowledge base, Q&A, and search to form a closed loop for review.

[0039] In a preferred embodiment of the present invention, the regulatory sensitive word check specifically includes: It detects promises and absolute terms, and provides compliance alternatives for detected non-compliant statements.

[0040] In a preferred embodiment of the present invention, the intelligent document review specifically involves checking the document format, structure, and content completeness; the general document check specifically involves detecting consistency of company names in headers and footers, typos, and grammatical issues; and the document comparison specifically involves automatically marking the differences between the target document and the comparison document to assist in review decisions.

[0041] As a preferred embodiment of the present invention, the document-process-tool collaboration mechanism specifically includes the following steps: The document visually displays the process association status. Documents with associated processes show clickable process titles, while documents without associated processes provide an entry point for process association. Implement access control for documents that have been linked to processes. If the business process moves to a specific node, the document permissions will be automatically adjusted.

[0042] As a preferred embodiment of the present invention, the access control is linked with the business process nodes. When the business process flows to the corresponding node, the system automatically adjusts the document's viewing, annotation, and editing permissions, and records the modification operations throughout the document's entire lifecycle.

[0043] The present invention discloses a system for improving the quality of investment banking business documents based on artificial intelligence, wherein the system includes: The data infrastructure module is used to connect to multi-source data and provide a template resource library to provide data support for document quality. The quality collaboration module, connected to the aforementioned data infrastructure module, includes a template filling-data source collaboration mechanism, a writing-search-question answer collaboration mechanism, a multi-dimensional document intelligent review collaboration mechanism, and a document-process-tool collaboration mechanism, used for full-process quality control of documents; The application interaction module is connected to the quality collaboration module to embed quality assurance capabilities into the entire document editing and circulation process.

[0044] In a preferred embodiment of the present invention, the data infrastructure module includes a multi-source data access unit, a template resource library unit, and an AI integration unit. The multi-source data access unit is used to connect to multiple business data sources covering project information, customer information, personnel information, and draft data. The template resource library unit is used to provide templates covering business scenarios such as project initiation, issuance, and ongoing maintenance. The AI ​​integration unit is used to support intelligent question answering, document polishing, and document interpretation.

[0045] The apparatus of the present invention for improving the quality of investment banking business documents based on artificial intelligence, wherein the apparatus includes: A processor is configured to execute computer-executable instructions; The memory stores one or more computer-executable instructions, which, when executed by the processor, implement the steps of the aforementioned method for improving the quality of investment banking business documents based on artificial intelligence.

[0046] The present invention discloses a processor for improving the quality of investment banking business documents based on artificial intelligence. The processor is configured to execute computer-executable instructions, and when the computer-executable instructions are executed by the processor, the various steps of the above-described method for improving the quality of investment banking business documents based on artificial intelligence are implemented.

[0047] The computer-readable storage medium of the present invention stores a computer program thereon, which can be executed by a processor to implement the various steps of the above-described method for improving the quality of investment banking business documents based on artificial intelligence.

[0048] Based on the unique characteristics of investment banking document writing and its business, this invention provides an integrated method and system for improving the quality of investment banking business documents. Through multi-dimensional AI capabilities, it achieves full-process quality control of documents from generation, assisted writing, review to circulation. The aim is to continuously improve document quality and document collaboration efficiency through system technology support, and build a complete investment banking document quality assurance system.

[0049] The present invention provides an artificial intelligence-based method for improving the quality of investment banking business documents, comprising the following steps: Build a data infrastructure module, connect to multi-source data and provide a template resource library to provide data support for document quality; The quality collaboration module employs four collaborative mechanisms to conduct end-to-end quality control of documents. These four mechanisms include: template filling-data source collaboration mechanism, writing-search-question-answer collaboration mechanism, multi-dimensional intelligent document review collaboration mechanism, and document-process-tool collaboration mechanism. Quality assurance capabilities are embedded into the entire document editing and workflow process through application interaction modules.

[0050] The template-data source collaboration mechanism includes: pre-establishing a mapping relationship between multi-source data fields and templates; after the user selects a template, the system automatically extracts matching fields from the data source for intelligent filling; performing data consistency verification during the filling process; and generating a filling report after the filling is completed, marking unmatched fields for the user to manually supplement.

[0051] The aforementioned writing-search-question-answering collaborative mechanism includes: when users are writing documents, they can trigger the write-and-search function to retrieve project information, customer information, personnel information, and knowledge base content in real time; the search results can be directly inserted at the cursor position; when encountering professional questions, AI question answering can be invoked with one click to obtain real-time answers, and the question answering results can be further inserted into the document.

[0052] The multi-dimensional intelligent document review and collaboration mechanism includes: conducting in-depth document review and automatically labeling documents with quality tags after approval; specifically including intelligent document review, regulatory sensitive word check, general document check, and document comparison; aggregating and displaying review results and linking them with knowledge base, Q&A, and search; supporting re-review and recording the entire problem-solving process.

[0053] The regulatory sensitive word check detects and identifies promises and absolute terms, including prohibited expressions such as "only choice," "best solution," "guaranteed returns," and "will definitely pass," and provides compliant alternative suggestions.

[0054] The document-process-tool collaboration mechanism includes: a document list that intuitively displays the process association status; clickable process titles for documents with associated processes, and an "unassociate" entry for unassociated documents; support for tools such as one-click PDF conversion, OCR recognition, and cross-verification after document finalization; access control for linked documents; and support for multi-user, multi-device collaboration.

[0055] An AI-based document quality improvement system for investment banking includes: The data infrastructure module includes a multi-source data access unit, a template resource library unit, and an intelligent AI integration unit. The quality collaboration module includes a template filling-data source collaboration mechanism, a writing-search-question answer collaboration mechanism, a multi-dimensional intelligent document review collaboration mechanism, and a document-process-tool collaboration mechanism. The multi-source data access unit connects to 7 types of data sources and 129 fields. The template resource library unit contains 127 templates. The system has verified the quality improvement effect based on actual business data of 7,604 documents, 3,685 templates, and 3,633 process attachment documents. The template element filling accuracy rate reached 98.5%, and the process attachment rate reached 98.6%.

[0056] This system consists of three parts: a data infrastructure module, a quality collaboration module, and an application interaction module. These modules work together to improve the quality of investment banking business documents.

[0057] (I) Data Foundation Module – Providing Data Support for Document Quality: Multi-source Data Access Unit: Connects to 129 fields from 7 types of internal business data sources, covering project information, customer information, personnel information, draft data, etc., to ensure the accuracy and reliability of document data.

[0058] Template Resource Library Unit: Contains 127 templates, covering all business scenarios such as project initiation, issuance, and life cycle, ensuring standardized and consistent document formats.

[0059] Intelligent AI Integration Unit: Connects to internal intelligent AI tools, providing support for intelligent agent capabilities such as intelligent question answering, document polishing, and document interpretation.

[0060] (II) Quality Collaboration Module – Four Collaboration Mechanisms to Ensure a Closed Loop of Document Quality Mechanism 1: Template Filling - Data Source Collaboration Mechanism (Document Pre-generation) The system pre-establishes a mapping relationship between 129 fields from 7 data sources and 127 templates, forming a field-template matching matrix. After the user selects a template, the system automatically extracts matching fields from the multi-source data and intelligently fills them into the corresponding positions in the template, achieving automated pre-generation of documents. During the filling process, data consistency checks are performed to ensure that the field types, formats, and value ranges meet the template requirements. If the check fails, it is automatically marked and a prompt is displayed. After the filling is completed, a filling report is generated, clearly showing the number of fields filled, the matching rate, and a list of unmatched fields for the user to manually supplement. Through this mechanism, 3,685 documents were automatically pre-generated, avoiding manual input errors from the source, and the document generation accuracy rate reached 98.5%.

[0061] Mechanism 2: Writing-Searching-Question-Answer Collaborative Mechanism (Document being written) Building a collaborative ecosystem for document writing: Users can trigger the write-and-search function at any time while writing documents, and access ecological data sources such as project library, customer library, personnel library, and knowledge base in real time. Search results can be directly inserted at the cursor position, forming an ecological collaboration of "writing - data source".

[0062] Ecological intelligent assistance embedded: When encountering professional questions, one-click access to Zhijian AI Q&A is provided, and AI capabilities are ecologically embedded into the writing process. Q&A results can be further inserted into documents, building an ecological workflow chain of "search → insert → consult → re-insert".

[0063] By directly inserting search results, we ensure the accuracy of cited information, and by using AI Q&A, we guarantee the standardization of professional expressions, thus avoiding manual input errors and professional biases from the source.

[0064] Mechanism 3: Multi-dimensional intelligent document review and collaboration mechanism (document review) This mechanism performs in-depth document review and automatically assigns a quality tag to documents upon approval, enabling rapid identification and accurate processing in subsequent workflows. Specific functions include: Intelligent document review: Create review tasks to conduct multi-dimensional checks on document format, structure, content completeness, etc. General document checks: Detects issues such as consistency of company name in headers and footers, typos, and grammatical errors; Document comparison: If the user selects documents to compare, the system automatically marks the differences to assist in the review decision-making; Document tag generation: After the document is approved, the system automatically assigns tags such as compliance status and risk level to the document based on the review results, providing a basis for identification in subsequent processes (such as process linkage, access control, and version tracking).

[0065] The aforementioned multi-dimensional review process covers four major dimensions: sensitive words, format, content, and consistency, ensuring a comprehensive and thorough review. Simultaneously, the system provides original regulatory rules through a review-knowledge base linkage, answers user questions about the review results in real time through a review-Q&A linkage, and retrieves standardized expressions from draft documents or historical cases for reference through a review-search linkage, forming a complete review loop of "identification → answering → verification → correction."

[0066] Mechanism 4: Document-Process-Tool Collaboration Mechanism (Document Flow) Workflow status visualization: The document list visually displays the workflow association status (associated / not associated) of each document.

[0067] Facilitate workflow navigation: Documents with associated workflows display clickable workflow titles, allowing users to directly navigate to the workflow details page upon clicking; documents without associated workflows provide an "Unassociate" button, enabling users to jump directly to the studio's workflow list to complete the association process.

[0068] Cross-validation of multi-source data: Perform consistency verification on key data in the document.

[0069] Access control: Linked documents are subject to strict access control to ensure that only authorized personnel can view and operate them.

[0070] Multi-user, multi-device collaboration: Supports multi-user, multi-device collaboration, allowing team members to view document status and provide feedback in real time.

[0071] Tool plug-and-play collaboration mechanism: After the document is finalized, it supports one-click conversion to PDF to ensure that the document format is standardized and consistent; for scanned PDF documents, OCR tools can be used to extract text information for the review process; other tools support plug-and-play and flexible expansion.

[0072] The system architecture and flowchart of the four major collaboration mechanisms are as follows: Figure 2 As shown.

[0073] like Figure 2 As shown, the main process is described below: 1) A method and system for improving the quality of investment banking business documents based on artificial intelligence, consisting of five aspects: intelligent writing, writing assistance, writing proofreading, process document control, and intelligent tool invocation.

[0074] 2) The key aspects of intelligent writing include template creation and data source setup.

[0075] 3) Writing assistance focuses on providing users with accurate data assistance during the writing process.

[0076] 4) Writing and proofreading mainly involves reviewing and verifying articles that have been basically completed, and further identifying potential problems in the documents.

[0077] 5) Workflow document management focuses on documents that have been written and associated with the workflow. It records and tracks documents, locks files, and ensures file consistency in relation to workflow flow and access control.

[0078] 6) Intelligent tool invocation mainly refers to the fact that the current system contains a wealth of document-related intelligent tools that can be plugged in and called at any time during document writing.

[0079] like Figure 3 As shown, the main process is described below: Currently, relevant business documents are generated at each stage of the investment banking business process. The system currently covers the writing of some documents and the control of related processes at the stages of solicitation, execution, underwriting, and follow-up supervision.

[0080] The following are specific embodiments of the present invention.

[0081] Example 1: The Entire Process of Improving the Quality of Bond Project Initiation Reports Step 1: Document pre-generation Users access the web-based editor and select a bond project initiation report template. The system initiates a template population-data source collaboration mechanism, automatically extracting fields such as project information, issuer information, and financial data from seven types of data sources. It automatically generates and pre-generates documents based on the template. During the population process, data consistency checks are performed to ensure that field types and formats meet requirements. After population is complete, successfully populated fields and fields that cannot be confirmed are displayed, prompting the user for confirmation. After the user completes the missing fields, a first draft is obtained.

[0082] Step 2: Writing Support During the writing process, users need to reference detailed customer information. This triggers various functions on the right panel, allowing real-time retrieval of the customer database. After selecting the target customer, the information is directly inserted at the cursor position. When encountering uncertainties regarding professional terms like "debt repayment guarantee measures," users can use the Zhijian AI Q&A tool with a single click to obtain standard definitions and common expressions. The Q&A results can then be further inserted into the document, forming a continuous workflow of "retrieval → insertion → consultation → re-insertion."

[0083] Step 3: Intelligent Review After the initial draft is completed, the user initiates a multi-dimensional intelligent review. The system performs intelligent document review, cross-verification, and general document checks in parallel. It also detects an error in the company name in the header and prompts for correction. The user clicks on the issue to jump directly to the corresponding location, accepts the suggestion, and completes the modification. After modification, the system re-reviews the document, confirms its approval, and automatically generates a "Compliance - Low Risk" label.

[0084] Step 4: Call other functions Once the document is finalized, users can convert it to PDF, compare it with other documents, or use other document-related tools. To integrate the workflow, proceed to step 5.

[0085] Step 5: Process Integration and Tool Application Once the document is finalized, users can click the "Unlink" button on the document list page to navigate to the process list and select "Bond Project Initiation Process" to link the document to the process. Strict access control and document logging will be implemented throughout the process to reduce the difficulty of the review process. If any content needs to be modified during the process, other members of the project team can view the document status and review comments in real time via mobile devices and collaboratively confirm via the comment function. The document will be locked in real time after the process is completed to prevent unauthorized alteration.

[0086] like Figure 4 As shown, document generation quality is guaranteed at the source: through the template filling-data source collaboration mechanism, 129 fields from 7 types of data sources are intelligently matched to the template, and 3,685 documents are automatically generated with an accuracy rate of 98.5%, avoiding manual input errors from the source.

[0087] Document review quality has been comprehensively improved: multi-dimensional parallel review covers four major dimensions: sensitive words, format, content, and consistency. The accuracy rate of problem identification is over 97%, and the review time has been significantly shortened.

[0088] The collaborative review process forms a closed loop: after the review is approved, the document is automatically assigned quality tags such as compliance status and risk level, and the review conclusion is transformed into an identifiable and traceable structured identifier; the review-knowledge base linkage provides regulatory basis, the review-Q&A linkage resolves user questions, the review-search linkage provides historical reference, and the review results are traceable and verifiable.

[0089] Document flow is standardized and controllable: 3,633 documents were successfully linked to business processes, with a linking rate of 98.6%. Linked documents are traceable, and unlinked documents can be guided, ensuring standardized document flow.

[0090] The document data is accurate and verifiable: supported by 7 categories and 129 fields of multi-source data, the system searches while writing to ensure accurate references and avoids errors from manual data entry.

[0091] Efficient and smooth team collaboration: Document comments are synchronized in both directions, supporting multi-device collaboration, ensuring that no team member's opinion is missed, and improving the quality of collaboration.

[0092] End-to-end quality closed loop: Four collaborative mechanisms run through all stages of template filling, writing, review, process, and tools, forming a complete document quality assurance system.

[0093] This invention introduces proactive, interactive, and ecological collaborative capabilities.

[0094] This invention features a collaborative writing-search-question-answering mechanism. This is not simply "filling in" information, but rather providing real-time, proactive knowledge services to users during the writing process. Users can trigger a search at any point in the document, and the system retrieves information in real-time from multiple ecosystems such as the "project library, client library, and knowledge base," supporting one-click insertion. Simultaneously, AI question-answering capabilities are "ecologically embedded" into the writing process, forming a continuous workflow of "search → insertion → consultation → re-insertion." This deep collaboration of "searching while writing and using immediately after asking questions" is something existing technologies lack.

[0095] Deep Embedding: Search and Q&A features are deeply embedded in the document writing flow (cursor position), so users do not need to switch applications or contexts.

[0096] Ecological data sources: The objects retrieved are not general information from the public internet, but core business data such as project databases, customer databases, and knowledge bases within the enterprise, ensuring the accuracy, relevance, and security of the information.

[0097] Results are what you get: Whether it's search results or AI Q&A results, they all support one-click insertion at the current cursor position, achieving a frictionless operation from "acquiring information" to "applying information".

[0098] A continuous workflow chain has been established, forming a closed-loop workflow of "retrieval → insertion → consultation → re-insertion," which is missing in conventional document tools. This is not simply a collection of functions, but a reconstruction and optimization of the document writing workflow, which greatly improves the efficiency and accuracy of professional document writing. This is not a "piecemeal" solution that those skilled in the art would easily think of when faced with the problem of "low efficiency."

[0099] This invention features a multi-dimensional intelligent document review and collaboration mechanism. It conducts multi-dimensional parallel reviews (format, content, consistency, and sensitive words) and automatically generates structured quality tags (such as "compliant - low risk") upon approval. This invention establishes a closed-loop linkage between review and knowledge base, review and Q&A, and review and search, ensuring that review results are verifiable, questions can be answered in real time, and corrections can be referenced from history. This transcends the scope of "detection" and rises to the level of "intelligent diagnosis and collaborative correction."

[0100] Multi-dimensional parallel review: Simultaneously conducts checks on multiple complex dimensions such as format, content completeness, data consistency, and regulatory sensitive words, especially the identification of financial regulatory sensitive words such as "commitment statements," which has a high degree of domain expertise.

[0101] Structured quality tags: The audit results are not simply a list of issues, but automatically generate structured quality tags such as "compliance status" and "risk level", which users can further view for intelligent audit results.

[0102] Workflow Integration -> Permissions / Approval: The document and workflow tool collaboration mechanism relies on the system's strict permission control and collaboration mechanisms to ensure both the integrity of documents and their collaborative nature while maintaining security. For example, when a business process reaches the "core" node, the system automatically adjusts the document's permission control (allowing only core committee members to view / annotate) and records the document's activity. Even when spanning different business processes, the system maintains controllable permissions for individual documents. This deep integration and mutual triggering of the document lifecycle and the business process lifecycle is a cross-domain technical problem completely unaddressed by existing technologies. This invention breaks away from traditional business document workflow processes, providing a novel and highly reliable solution.

[0103] This invention creatively solves the problems existing throughout the entire lifecycle of investment banking business documents, from generation to archiving, through the deep coupling and collaborative operation of four major collaborative mechanisms. While improving work efficiency, it significantly enhances the quality of business documents (especially those disclosed externally), thereby reducing the professional risks for business personnel. Its overall architecture, core mechanisms, and the resulting collaborative technical effects are detailed below. For the specific implementation scheme of this embodiment, please refer to the relevant descriptions in the above embodiments, which will not be repeated here.

[0104] It is understood that the same or similar parts in the above embodiments can be referred to each other, and the contents not described in detail in some embodiments can be referred to the same or similar contents in other embodiments.

[0105] It should be noted that in the description of this invention, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this invention, unless otherwise stated, "a plurality of" means at least two.

[0106] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of the preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

[0107] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution device. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0108] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The corresponding program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

[0109] Furthermore, the functional units in the various embodiments of the present invention can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0110] The storage media mentioned above can be read-only memory, disk, or optical disk, etc.

[0111] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0112] This invention employs a method, system, apparatus, processor, and computer-readable storage medium based on artificial intelligence to improve the quality of investment banking business documents. It constructs a closed-loop quality control system covering the entire process, from document generation to writing assistance, intelligent review, and finally, circulation and collaboration. These four mechanisms are not simply superimposed but rather form an interdependent and interconnected organic whole. This invention focuses on how to systematically improve the quality of investment banking documents throughout the entire business lifecycle; its technical concept is a closed-loop control system of "generation-writing-review-circulation."

[0113] In this specification, the invention has been described with reference to specific embodiments thereof. However, it will be apparent that various modifications and variations can be made without departing from the spirit and scope of the invention. Therefore, the specification and drawings should be considered illustrative rather than restrictive.

Claims

1. A method for improving the quality of investment banking business documents based on artificial intelligence, characterized in that, The method includes the following steps: (1) Construct a data infrastructure module, connect to multi-source data and provide a template resource library to provide data support for document quality; (2) Through the four collaborative mechanisms in the quality collaboration module, the entire process of quality control of investment banking business documents is carried out from pre-generation, writing, review to circulation; the four collaborative mechanisms include template filling-data source collaboration mechanism, writing-search-question and answer collaboration mechanism, multi-dimensional document intelligent review collaboration mechanism, and document-process-tool collaboration mechanism. (3) By applying the interaction module, the quality control capability is embedded into the entire process of document editing and business flow.

2. The method for improving the quality of investment banking business documents based on artificial intelligence according to claim 1, characterized in that, The template-data source collaboration mechanism includes the following steps: Establish a mapping relationship between multi-source data fields and templates in advance to form a field-template matching matrix; After selecting a template, the system automatically extracts matching fields from multi-source data and intelligently fills them into the corresponding positions in the template to pre-generate the document; Data consistency verification is performed during the data filling process; Once the population is complete, a population report will be generated, showing the populated fields, the matching rate, and a list of unmatched fields.

3. The method for improving the quality of investment banking business documents based on artificial intelligence according to claim 2, characterized in that, The data consistency verification specifically verifies whether the field type, format, and value range meet the template requirements.

4. The method for improving the quality of investment banking business documents based on artificial intelligence according to claim 1, characterized in that, The aforementioned writing-search-question-answering collaborative mechanism specifically includes the following steps: During the document writing process, the write-and-search function is triggered to retrieve internal business data from the project library, customer library, personnel library, and knowledge base in real time and directly insert the search results; With one click, you can access the AI ​​integration unit to obtain real-time answers to professional questions and directly insert the question-and-answer results to form a continuous workflow chain.

5. The method for improving the quality of investment banking business documents based on artificial intelligence according to claim 1, characterized in that, The aforementioned multi-dimensional intelligent document review and collaboration mechanism specifically includes the following steps: The document undergoes multi-dimensional parallel review, including intelligent document verification, regulatory sensitive word checking, general document check, and document comparison. After the document is approved, a quality tag is automatically added to it, which includes compliance status and risk level. The review results are aggregated and displayed, and linked with the knowledge base, Q&A, and search to form a closed loop for review.

6. The method for improving the quality of investment banking business documents based on artificial intelligence according to claim 5, characterized in that, The aforementioned check for sensitive regulatory words specifically includes: It detects promises and absolute terms, and provides compliance alternatives for detected non-compliant statements.

7. The method for improving the quality of investment banking business documents based on artificial intelligence according to claim 5, characterized in that, The aforementioned intelligent document review specifically checks the document format, structure, and content completeness; the aforementioned general document check specifically detects consistency of company names in headers and footers, typos, and grammatical issues; the aforementioned document comparison specifically marks the differences between the target document and the comparison document to assist in review decisions.

8. The method for improving the quality of investment banking business documents based on artificial intelligence according to claim 1, characterized in that, The document-process-tool collaboration mechanism specifically includes the following steps: The document visually displays the process association status. Documents with associated processes show clickable process titles, while documents without associated processes provide an entry point for process association. Implement access control for documents that have been linked to processes. If the business process moves to a specific node, the document permissions will be automatically adjusted.

9. The method for improving the quality of investment banking business documents based on artificial intelligence according to claim 8, characterized in that, The aforementioned access control is linked to business process nodes. When the business process flows to the corresponding node, the system automatically adjusts the document's viewing, annotation, and editing permissions, and records all modification operations throughout the document's lifecycle.

10. A system for improving the quality of investment banking business documents based on artificial intelligence, characterized in that, The system includes: The data infrastructure module is used to connect to multi-source data and provide a template resource library to provide data support for document quality. The quality collaboration module, connected to the aforementioned data infrastructure module, includes a template filling-data source collaboration mechanism, a writing-search-question answer collaboration mechanism, a multi-dimensional document intelligent review collaboration mechanism, and a document-process-tool collaboration mechanism, used for full-process quality control of documents; The application interaction module is connected to the quality collaboration module to embed quality assurance capabilities into the entire document editing and circulation process.

11. The system for improving the quality of investment banking business documents based on artificial intelligence according to claim 1, characterized in that, The data infrastructure module includes a multi-source data access unit, a template resource library unit, and an AI integration unit. The multi-source data access unit is used to connect to various business data sources covering project information, customer information, personnel information, and draft data. The template resource library unit is used to provide templates covering business scenarios such as project initiation, issuance, and ongoing maintenance. The AI ​​integration unit is used to support intelligent question answering, document polishing, and document interpretation.

12. A device for improving the quality of investment banking business documents based on artificial intelligence, characterized in that, The device includes: A processor is configured to execute computer-executable instructions; The memory stores one or more computer-executable instructions, which, when executed by the processor, implement the steps of the method for improving the quality of investment banking business documents based on artificial intelligence, as described in any one of claims 1 to 9.

13. A processor for improving the quality of investment banking business documents based on artificial intelligence, characterized in that, The processor is configured to execute computer-executable instructions, which, when executed by the processor, implement the steps of the method for improving the quality of investment banking business documents based on artificial intelligence as described in any one of claims 1 to 9.

14. A computer-readable storage medium, characterized in that, It stores a computer program that can be executed by a processor to implement the steps of the method for improving the quality of investment banking business documents based on artificial intelligence, as described in any one of claims 1 to 9.

Images