Intelligent logistics solution generation method, system, device and storage medium
By constructing a structured knowledge base and using natural language processing models to automatically parse and assemble solution documents, the problem of difficult reuse of knowledge assets in unstructured documents for logistics companies has been solved, achieving efficient and accurate solution generation and rapid response.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SINOTRANS
- Filing Date
- 2026-02-06
- Publication Date
- 2026-06-26
Smart Images
Figure CN122285755A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of artificial intelligence technology, and in particular to a method, system, device, and storage medium for generating intelligent logistics solutions. Background Technology
[0002] With the deepening of global industrial upgrading and digital transformation, the logistics industry is facing unprecedented pressure and opportunities for change. Large logistics companies, such as Sinotrans, have accumulated a large number of historical service solutions (usually in the form of unstructured documents such as PowerPoint presentations) over the course of their long-term service. These solutions are valuable knowledge assets of the company, containing valuable experience in serving different industries and scenarios. However, currently, the use of these knowledge assets to respond to new customer needs mainly relies on manual operation. When new customer needs arise, business personnel need to manually search and filter for potentially relevant solutions from a massive amount of historical PowerPoint files, relying on personal memory and experience. This process is time-consuming and labor-intensive, and it is easy to miss the best solution, making it difficult to respond quickly to market demands. Historical solutions are stored in the form of unstructured PowerPoint documents, and computers cannot directly understand and process the semantic content within them. This results in the inability of the company's knowledge assets to be effectively analyzed, managed, and utilized by intelligent systems. Even if relevant materials are found, assembling a new, complete solution PowerPoint presentation still requires a lot of manual copying, pasting, and formatting work. This is not only inefficient, but also makes it difficult to ensure the consistency and professionalism of the output solution in terms of format and style. Summary of the Invention
[0003] This invention provides a method, system, device, and storage medium for generating intelligent logistics solutions, which addresses the shortcomings of relying on manual retrieval and assembly of solutions from massive amounts of unstructured documents, resulting in low efficiency, insufficient matching accuracy, and difficulty in effectively reusing knowledge assets.
[0004] This invention provides a method for generating an intelligent logistics solution, comprising: Analyze historical solution documents, extract text content, and use a natural language processing model to add multi-dimensional tags to the text content. Construct a structured knowledge base based on the text content and the multi-dimensional tags. The system receives query conditions input by the user, matches the query conditions with multi-dimensional tags in the structured knowledge base, and retrieves target solution materials that meet the conditions from the structured knowledge base. The target solution materials are combined and arranged with a preset template to generate a solution document.
[0005] According to the intelligent logistics solution generation method provided by the present invention, the step of parsing historical solution documents and extracting text content includes: The python-pptx library is used to parse the historical scheme documents in PPTX format page by page to extract the text content in the text boxes and tables of each page of the historical scheme documents.
[0006] According to the intelligent logistics solution generation method provided by the present invention, the step of using a natural language processing model to assign multi-dimensional tags to the text content and constructing a structured knowledge base based on the text content and the multi-dimensional tags includes: Construct structured cue words for multi-label classification, wherein the cue words contain text to be labeled extracted from the text content and predefined label descriptions; The structured prompt words are input into the natural language processing model, and the structured response data returned by the natural language processing model is received. From the structured response data, parse and extract label results of multiple dimensions corresponding to the predefined label descriptions; The text content, the corresponding multi-dimensional tag results, and their index information in the original document are associated and stored in the database to obtain a structured knowledge base.
[0007] According to the intelligent logistics solution generation method provided by the present invention, the multi-dimensional tags include transportation mode tags, industry classification tags, commodity type tags, transportation information tags, and transportation location tags. The step of matching the query conditions with the multi-dimensional tags in the structured knowledge base and retrieving target solution materials that meet the conditions from the structured knowledge base includes: Parse the query conditions and extract the business requirement parameters from them; Based on the transportation mode tags, industry classification tags, commodity type tags, transportation information tags, and transportation location tags, the solution materials in the structured knowledge base are initially filtered; Calculate the matching score of each scheme material after the initial filtering, and select the high-scoring scheme set that reaches the preset threshold; Semantic similarity is calculated on the set of high-scoring schemes, and schemes with similar semantics are deduplicated. Structured prompt words are constructed based on the high-scoring scheme after deduplication and its business context set; The structured prompts are input into the natural language processing model, which outputs target solution materials that meet the conditions.
[0008] According to the intelligent logistics solution generation method provided by the present invention, the calculation of the matching degree score of each solution material after preliminary filtering includes: Determine the multi-dimensional labels to participate in the matching calculation, and assign a weight coefficient to each dimension label, where the sum of the weight coefficients is 1; For each solution material in the structured knowledge base, the values of its dimension tags are compared with the query conditions: if they match, the corresponding dimension score is 1; if they do not match, the corresponding dimension score is 0. The matching score of each solution's materials is calculated using a weighted summation formula.
[0009] According to the intelligent logistics solution generation method provided by the present invention, the step of splicing and arranging the target solution materials with a preset template to generate a solution document includes: Call the document manipulation interface of the Spire.Presentation library to create a new blank PPT document; According to the preset assembly logic, the pre-stored fixed chapter templates and the target solution materials are automatically integrated into the new blank PPT document. During the integration process, based on the preset format specifications, the page layout, text style and image insertion are uniformly controlled through the Spire.Presentation library to generate the solution document.
[0010] According to the intelligent logistics solution generation method provided by the present invention, the step of uniformly controlling page layout, text style, and image insertion through the Spire.Presentation library based on preset format specifications includes: Apply a consistent background, color scheme, and placeholder layout based on the slide master template defined in the format specification. Iterate through all text boxes in the integrated PPT document and set the font, font size, color, and alignment in batches according to the preset title and body text style rules; Extract the image elements contained in the target solution material, and insert them into the specified area of the corresponding slide according to the preset image frame size and position parameters; The inserted image is adaptively scaled to ensure it maintains its original aspect ratio and does not exceed the preset boundary range.
[0011] This invention also provides a system for generating intelligent logistics solutions, comprising: The parsing module is used to parse historical solution documents, extract text content, and use a natural language processing model to add multi-dimensional tags to the text content, and construct a structured knowledge base based on the text content and the multi-dimensional tags. The retrieval module is used to receive query conditions input by the user, match the query conditions with multi-dimensional tags in the structured knowledge base, and retrieve target solution materials that meet the conditions from the structured knowledge base. The generation module is used to combine and arrange the target solution materials with the preset template to generate a solution document.
[0012] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement a method for generating an intelligent logistics solution as described in any of the preceding claims.
[0013] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method for generating the intelligent logistics solution described in any of the preceding claims.
[0014] The intelligent logistics solution generation method, system, equipment, and storage medium provided by this invention analyze historical solution documents, extract text content, and use a natural language processing model to tag the text content with multi-dimensional labels. A structured knowledge base is then constructed based on the text content and the multi-dimensional labels. User-input query conditions are received, and the query conditions are matched with the multi-dimensional labels in the structured knowledge base to retrieve target solution materials that meet the conditions. The target solution materials are then combined and formatted with a preset template to generate a solution document. This invention automatically analyzes historical solutions and uses a large model for tagging, transforming historical solutions into a structured knowledge base, thus improving tagging efficiency. The use of multi-dimensional labels combined with a quantitative scoring mechanism enables high-quality solution pre-screening, effectively supporting rapid service delivery in business scenarios and improving service matching efficiency and accuracy. Attached Figure Description
[0015] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0016] Figure 1 This is one of the flowcharts for the method of generating an intelligent logistics solution provided in an embodiment of the present invention; Figure 2 This is the second flowchart of the method for generating an intelligent logistics solution provided in this embodiment of the invention; Figure 3 This is the third flowchart of the method for generating the intelligent logistics solution provided in this embodiment of the invention; Figure 4 This is a schematic diagram of the functional structure of the intelligent logistics solution generation system provided in this embodiment of the invention; Figure 5 This is a functional structure diagram of the electronic device provided in an embodiment of the present invention. Detailed Implementation
[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.
[0018] Figure 1 A flowchart of the method for generating an intelligent logistics solution provided in an embodiment of the present invention is shown below. Figure 1 As shown, the method for generating an intelligent logistics solution provided in this embodiment of the invention is applicable to the receiving end and includes: Step 101: Parse historical solution documents, extract text content, and use a natural language processing model to add multi-dimensional tags to the text content. Construct a structured knowledge base based on the text content and the multi-dimensional tags. Step 102: Receive the query conditions input by the user, match the query conditions with the multi-dimensional tags in the structured knowledge base, and retrieve the target solution materials that meet the conditions from the structured knowledge base; Step 103: Combine and arrange the target solution materials with the preset template to generate a solution document.
[0019] Traditional intelligent logistics solutions primarily rely on manual processing. When new customer needs arise, business personnel must manually search and filter for potentially relevant solutions from a vast database of historical PowerPoint presentations, relying on personal memory and experience. This process is time-consuming, labor-intensive, and prone to overlooking optimal solutions, hindering rapid response to market demands. Historical solutions are stored as unstructured PowerPoint documents, meaning computers cannot directly understand or process their semantic content. This prevents intelligent systems from effectively analyzing, managing, and utilizing the company's knowledge assets. Even when relevant materials are found, assembling a new, complete solution PowerPoint presentation still requires significant manual copying, pasting, and formatting. This is not only inefficient but also makes it difficult to ensure consistency and professionalism in the format and style of the output solutions.
[0020] The intelligent logistics solution generation method provided in this embodiment of the invention involves parsing historical solution documents, extracting text content, and using a natural language processing model to assign multi-dimensional tags to the text content. A structured knowledge base is then constructed based on the text content and the multi-dimensional tags. User-input query conditions are received, and these conditions are matched against the multi-dimensional tags in the structured knowledge base to retrieve target solution materials that meet the conditions. The target solution materials are then combined and formatted with a preset template to generate a solution document. This embodiment of the invention improves annotation efficiency by automatically parsing historical solutions and using a large model for tagging, transforming historical solutions into a structured knowledge base. The use of multi-dimensional tags combined with a quantitative scoring mechanism enables high-quality solution pre-screening, effectively supporting rapid service delivery in business scenarios and improving service matching efficiency and accuracy.
[0021] Based on any of the above embodiments, the process of parsing historical solution documents and extracting text content includes: The python-pptx library is used to parse the historical scheme documents in PPTX format page by page to extract the text content in the text boxes and tables of each page of the historical scheme documents.
[0022] Based on any of the above embodiments, the step of using a natural language processing model to assign multi-dimensional tags to the text content, and constructing a structured knowledge base based on the text content and the multi-dimensional tags, includes: Step 201: Construct structured prompt words for multi-label classification, wherein the prompt words contain the text to be labeled extracted from the text content and predefined label descriptions; Step 202: Input the structured prompt words into the natural language processing model and receive the structured response data returned by the natural language processing model; Step 203: Parse and extract multiple dimensions of label results corresponding to the predefined label descriptions from the structured response data; Step 204: Associate and store the text content, the corresponding multi-dimensional tag results, and their index information in the original document in the database to obtain a structured knowledge base.
[0023] like Figure 2As shown, the python-pptx library is used to extract the text and table content of PPT page by page, preserving the original structural information; a structured prompt is constructed and a large model API is called to perform multi-label classification. The key labels include: type (page type) (8 categories such as company introduction / product solution / industry analysis, etc.), company_names (company name), trans_type (transportation mode) (air / water / land / road / warehousing), industry_type (industry classification) (9 major fields such as automotive / medical / new energy / mining, etc.), good_type (product information), trans_info (address information), and trans_lines (address routes); the labeling results are structured and stored in the database to build a searchable knowledge asset.
[0024] Based on any of the above embodiments, the multi-dimensional tags include transportation mode tags, industry classification tags, product type tags, transportation information tags, and transportation location tags. The step of matching the query conditions with the multi-dimensional tags in the structured knowledge base and retrieving target solution materials that meet the conditions from the structured knowledge base includes: Step 301: Parse the query conditions and extract the business requirement parameters from the query conditions; Step 302: Based on the transportation mode label, industry classification label, commodity type label, transportation information label, and transportation location label, perform preliminary filtering on the solution materials in the structured knowledge base; Step 303: Calculate the matching score of each scheme material after the initial filtering, and select the high-scoring scheme set that reaches the preset threshold; Step 304: Calculate the semantic similarity of the high-scoring scheme set and deduplicate schemes with similar semantics; Step 305: Construct structured prompt words based on the deduplicated high-scoring scheme and its business context set; Step 306: Input the structured prompts into the natural language processing model and output the target solution materials that meet the conditions.
[0025] like Figure 3As shown, retrieving target solution materials that meet the conditions from the structured knowledge base specifically includes: parsing the business requirement parameters in the user query; filtering data based on five-dimensional tags: trans_type (transportation mode), industry_type (industry), trans_info (transportation information), good_type (product type), and trans_loc (transportation location); calculating the matching degree (out of 100 points) based on the number of intersections between the query conditions and tags, using the formula: Matching degree score = (Tag 1 score × Weight 1 + Tag 2 score × Weight 2 + ... + Tag n score × Weight n) × 100; filtering high-quality solutions with a score ≥ 22.5; using word vector technology (such as BERT) to perform semantic similarity clustering on high-scoring solutions to eliminate content redundancy; constructing a structured prompt to call a large model, and intelligently sorting the final results based on business logic.
[0026] In this embodiment of the invention, calculating the matching score of each scheme material after preliminary filtering includes: Step 3031: Determine the multi-dimensional labels participating in the matching calculation, and assign a weight coefficient to each dimension label, wherein the sum of the weight coefficients is 1; Step 3032: For each solution material in the structured knowledge base, compare its values on each dimension label with the query conditions: if they match, the corresponding dimension score is 1; if they do not match, the corresponding dimension score is 0. Step 3033: Calculate the matching score of each solution material according to the weighted summation formula.
[0027] Based on any of the above embodiments, the step of splicing and arranging the target solution materials with a preset template to generate a solution document includes: Step 401: Call the document manipulation interface of the Spire.Presentation library to create a new blank PPT document; Step 402: According to the preset assembly logic, the pre-stored fixed chapter templates and the target solution materials are automatically integrated into the new blank PPT document. During the integration process, based on the preset format specifications, the page layout, text style and image insertion are uniformly controlled through the Spire.Presentation library to generate the solution document.
[0028] In this embodiment of the invention, the unified control of page layout, text style, and image insertion through the Spire.Presentation library based on preset format specifications includes: Step 4021: Apply a uniform background, color scheme, and placeholder layout according to the slide master template defined in the format specification; Step 4022: Traverse all text boxes in the integrated PPT document, and set the font, font size, color and alignment in batches according to the preset title and body text style rules; Step 4023: Extract the image elements contained in the target solution material and insert them into the specified area of the corresponding slide according to the preset image frame size and position parameters; Step 4024: Perform adaptive scaling on the inserted image to ensure it maintains its original aspect ratio and does not exceed the preset boundary range.
[0029] In this embodiment of the invention, based on the thumbnail of the material search results selected by the user in the front-end interactive interface, high-quality content resources that have been tagged and categorized and stored in the knowledge base are accurately located. Simultaneously, combined with standardized sections such as company introduction and service commitments required for the project, automated integration is achieved using Python's Spire.Presentation library. This library possesses powerful PPT document manipulation capabilities, allowing flexible control over page layout, text styles, image insertion, and other elements. By writing Python scripts, following preset logic and formatting specifications, the selected materials and fixed sections are intelligently spliced and optimized for layout, ultimately quickly generating a service solution PPT document that is structurally complete, stylistically consistent, and meets business requirements.
[0030] The intelligent logistics solution generation method provided in this invention is based on parsing historical PPTX files page by page, extracting text from text boxes and tables, performing accurate tagging through a large model, storing the tags and file indexes in a database, quickly filtering when the user inputs tag conditions, scoring and sorting based on a tag matching weighted model, and finally automatically assembling the content using a PPTX library combined with the Spire.presentation library, generating a PPTX that meets the requirements after format and logic validation. This document addresses the issues of low knowledge reuse and inefficient retrieval caused by traditional manual methods. When users input their needs, the system uses a multi-dimensional tagging system and a weighted matching model for rapid filtering and quantitative scoring. This allows for the accurate and rapid identification of the highest-quality alternatives that best meet customer needs from a vast pool of options. It overcomes the inherent subjectivity, inaccuracy, and tendency to miss optimal solutions in manual matching, ensuring the objectivity and scientific rigor of the recommended solutions. Furthermore, the system utilizes an automated PPT library to intelligently combine and format selected high-quality materials and standard templates, automatically generating a complete, consistent, and professional final solution document. This not only improves solution creation efficiency but also ensures consistent quality and professional standards through standardized processes, thereby enhancing the company's responsiveness and core competitiveness in a rapidly changing market.
[0031] Figure 4 This is a schematic diagram of the structure of the intelligent logistics solution generation system provided in the embodiments of the present invention, as shown below. Figure 4 As shown, the intelligent logistics solution generation system provided in this embodiment of the invention includes: The parsing module 401 is used to parse historical solution documents, extract text content, and use a natural language processing model to add multi-dimensional tags to the text content, and construct a structured knowledge base based on the text content and the multi-dimensional tags. The retrieval module 402 is used to receive query conditions input by the user, match the query conditions with multi-dimensional tags in the structured knowledge base, and retrieve target solution materials that meet the conditions from the structured knowledge base. The generation module 403 is used to combine and arrange the target solution materials with the preset template to generate a solution document.
[0032] The intelligent logistics solution generation system provided in this invention can accurately describe capabilities and efficiently match resources to meet the needs of different customers. For example, when a new energy vehicle customer submits a transportation request, the system can quickly generate one or more optimal logistics solutions based on information such as the type and quantity of goods to be transported, the origin and destination, and time requirements, combined with the company's vehicle resources, route planning experience, and relevant professional talent reserves in new energy vehicle transportation. The system also details the company's professional capabilities in the field of new energy vehicle logistics, such as the use of special transportation equipment and strict adherence to safety regulations for the transportation of new energy vehicle batteries. The system possesses several key functions to ensure its effective operation. In terms of solution creation, the intelligent template and data-driven generation mechanism can quickly generate professional, detailed logistics solutions that meet the personalized needs of customers. The solution update function ensures that the system can optimize and adjust the generated solutions in real time based on market dynamics, changes in company resources, and customer feedback, ensuring that the solutions remain timely and competitive.
[0033] The intelligent logistics solution generation system provided in this embodiment of the invention also includes an access control / isolation module. Through an access control system, it ensures data security and strict division of operational permissions among different departments, personnel at different levels, and different customers, preventing information leakage and unauthorized operations, and maintaining enterprise operational order and customer information security. By developing this system, Sinotrans is expected to further consolidate its leading position in the complex and ever-changing logistics market, providing customers in various industries with higher-quality, more efficient, and personalized intelligent supply chain solutions.
[0034] The intelligent logistics solution generation system provided in this invention analyzes historical solution documents, extracts text content, and uses a natural language processing model to add multi-dimensional tags to the text content. A structured knowledge base is then constructed based on the text content and the multi-dimensional tags. The system receives user-input query conditions, matches these conditions with the multi-dimensional tags in the structured knowledge base, and retrieves target solution materials that meet the conditions. These target solution materials are then combined and formatted with a preset template to generate a solution document. By automatically analyzing historical solutions and using a large model for tagging, historical solutions are transformed into a structured knowledge base, improving tagging efficiency. The system employs a multi-dimensional tagging combined with a quantitative scoring mechanism to achieve high-quality solution pre-screening, effectively supporting rapid service delivery in business scenarios and improving service matching efficiency and accuracy.
[0035] Figure 5 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 5As shown, the electronic device may include a processor 510, a communication interface 520, a memory 530, and a communication bus 540. The processor 510, communication interface 520, and memory 530 communicate with each other via the communication bus 540. The memory 530 includes computer programs, an operating system, and acquired data. The processor 510 can call logical instructions in the memory 530 to execute a method for generating an intelligent logistics solution. This method includes: parsing historical solution documents, extracting text content, and using a natural language processing model to add multi-dimensional tags to the text content; constructing a structured knowledge base based on the text content and the multi-dimensional tags; receiving user-input query conditions, matching the query conditions with the multi-dimensional tags in the structured knowledge base, and retrieving target solution materials that meet the conditions from the structured knowledge base; and splicing and formatting the target solution materials with a preset template to generate a solution document.
[0036] Furthermore, the logical instructions in the aforementioned memory 530 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to related technologies, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0037] On the other hand, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements a method for generating intelligent logistics solutions provided by the methods described above. This method includes: parsing historical solution documents, extracting text content, and using a natural language processing model to assign multi-dimensional tags to the text content; constructing a structured knowledge base based on the text content and the multi-dimensional tags; receiving query conditions input by a user, matching the query conditions with the multi-dimensional tags in the structured knowledge base, and retrieving target solution materials that meet the conditions from the structured knowledge base; and splicing and formatting the target solution materials with a preset template to generate a solution document.
[0038] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.
[0039] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the related technology, can be embodied in the form of software products. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.
[0040] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for generating an intelligent logistics solution, characterized in that, include: Analyze historical solution documents, extract text content, and use a natural language processing model to add multi-dimensional tags to the text content. Construct a structured knowledge base based on the text content and the multi-dimensional tags. The system receives query conditions input by the user, matches the query conditions with multi-dimensional tags in the structured knowledge base, and retrieves target solution materials that meet the conditions from the structured knowledge base. The target solution materials are combined and arranged with a preset template to generate a solution document.
2. The method for generating an intelligent logistics solution according to claim 1, characterized in that, The extracted text content from the historical parsing document includes: The python-pptx library is used to parse the historical scheme documents in PPTX format page by page to extract the text content in the text boxes and tables of each page of the historical scheme documents.
3. The method of generating a smart logistics solution of claim 1, wherein, The step of using a natural language processing model to assign multi-dimensional labels to the text content and constructing a structured knowledge base based on the text content and the multi-dimensional labels includes: Construct structured cue words for multi-label classification, wherein the cue words contain text to be labeled extracted from the text content and predefined label descriptions; The structured prompt words are input into the natural language processing model, and the structured response data returned by the natural language processing model is received. From the structured response data, parse and extract label results of multiple dimensions corresponding to the predefined label descriptions; The text content, the corresponding multi-dimensional tag results, and their index information in the original document are associated and stored in the database to obtain a structured knowledge base.
4. The method of claim 1, wherein, The multi-dimensional tags include transportation mode tags, industry classification tags, product type tags, transportation information tags, and transportation location tags. The process involves matching the query conditions with the multi-dimensional tags in the structured knowledge base to retrieve target solution materials that meet the conditions from the structured knowledge base, including: Parse the query conditions and extract the business requirement parameters from them; Based on the transportation mode tags, industry classification tags, commodity type tags, transportation information tags, and transportation location tags, the solution materials in the structured knowledge base are initially filtered; Calculate the matching score of each scheme material after the initial filtering, and select the high-scoring scheme set that reaches the preset threshold; Semantic similarity is calculated on the set of high-scoring schemes, and schemes with similar semantics are deduplicated. Structured prompt words are constructed based on the high-scoring scheme after deduplication and its business context set; The structured prompts are input into the natural language processing model, which outputs target solution materials that meet the conditions.
5. The method of generating a smart logistics solution according to claim 4, wherein, The calculation of the matching score for each scheme material after the initial filtering includes: Determine the multi-dimensional labels to participate in the matching calculation, and assign a weight coefficient to each dimension label, where the sum of the weight coefficients is 1; For each solution material in the structured knowledge base, the values of its dimension tags are compared with the query conditions: if they match, the corresponding dimension score is 1; if they do not match, the corresponding dimension score is 0. The matching score of each solution's materials is calculated using a weighted summation formula.
6. The method of generating a smart logistics solution of claim 1, wherein, The step of combining and arranging the target solution materials with a preset template to generate a solution document includes: Call the document manipulation interface of the Spire.Presentation library to create a new blank PPT document; According to the preset assembly logic, the pre-stored fixed chapter templates and the target solution materials are automatically integrated into the new blank PPT document. During the integration process, based on the preset format specifications, the page layout, text style and image insertion are uniformly controlled through the Spire.Presentation library to generate the solution document.
7. The method of generating a smart logistics solution of claim 6, wherein, The aforementioned format specifications, based on the Spire.Presentation library, provide unified control over page layout, text styles, and image insertion, including: Apply a consistent background, color scheme, and placeholder layout based on the slide master template defined in the format specification. Iterate through all text boxes in the integrated PPT document and set the font, font size, color, and alignment in batches according to the preset title and body text style rules; Extract the image elements contained in the target solution material, and insert them into the specified area of the corresponding slide according to the preset image frame size and position parameters; The inserted image is adaptively scaled to ensure it maintains its original aspect ratio and does not exceed the preset boundary range.
8. A system for generating intelligent logistics solutions, characterized in that, include: The parsing module is used to parse historical solution documents, extract text content, and use a natural language processing model to add multi-dimensional tags to the text content, and construct a structured knowledge base based on the text content and the multi-dimensional tags. The retrieval module is used to receive query conditions input by the user, match the query conditions with multi-dimensional tags in the structured knowledge base, and retrieve target solution materials that meet the conditions from the structured knowledge base. The generation module is used to combine and arrange the target solution materials with a preset template to generate a solution document.
9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method for generating an intelligent logistics solution as described in any one of claims 1 to 7.
10. A non-transitory readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the method for generating an intelligent logistics solution as described in any one of claims 1 to 7.