Intelligent online document comparison method and system under supply chain purchase scene

Abstract

The invention relates to the technical field of supply chain management, and discloses an intelligent online document comparison method and system in a supply chain purchase scenario, and the method comprises the following steps: receiving a document in a client, carrying out the preprocessing of the document, carrying out the intelligent comparison and analysis through a preprocessing engine, carrying out the semantic analysis of a structured document output by the preprocessing engine, and carrying out the analysis of the structured document; the intelligent comparison and analysis module is used for identifying and extracting key business entities, establishing a field mapping relation with a supply chain field lexicon according to a structured document content type and business logic, and carrying out cross-document comparison; and the report generator generates difference analysis and report generation. The working efficiency can be greatly improved, and the comparison accuracy is improved; the intelligent level is enhanced; the purchasing risk is reduced; the integration is easy.

Description

Technical Field

[0001] This invention relates to the field of supply chain management technology, specifically to an intelligent online document comparison method and system for supply chain procurement scenarios. Background Technology

[0002] In enterprise supply chain management (SCM), the procurement process involves massive amounts of document processing, such as tender documents, supplier quotations, purchase contracts, orders, technical specifications, and business instructions. These documents are typically generated by different systems or uploaded by different personnel, exhibiting diverse formats and complex content. Currently, document comparison mainly relies on manual work, which presents the following significant problems:

[0003] 1. Inefficiency: Manual line-by-line comparison is time-consuming and labor-intensive, especially when dealing with large contracts or multiple supplier quotations, resulting in long evaluation cycles and slow response times.

[0004] 2. High error rate: The human eye is easily fatigued and it is difficult to detect subtle differences (such as decimal points, units, and wording of clauses), leading to contract disputes or procurement losses.

[0005] 3. Lack of intelligence: Traditional comparison tools (such as the comparison function built into Word) can only perform mechanical comparisons at the text level, and cannot understand the semantics of supply chain business, nor can they identify "5kg" and "5000g" as equivalent information.

[0006] 4. No optimization suggestions: After the comparison is completed, the system cannot proactively provide decision support, such as price anomaly warnings or delivery cycle risk alerts.

[0007] While existing technologies include general document comparison tools and some natural language processing (NLP) applications, they lack in-depth optimization for supply chain procurement scenarios and cannot meet enterprises' needs for accuracy, efficiency, and intelligence in procurement document processing. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this invention provides an intelligent online document comparison method for supply chain procurement scenarios, comprising the following steps:

[0009] Document preprocessing: Users upload procurement documents, and the system converts the procurement documents into a processable text stream.

[0010] Intelligent comparative analysis performs semantic analysis on the preprocessed text stream, identifies and extracts key business entities, establishes field mapping relationships based on the document type and business logic of the text stream, performs cross-document comparison, and uses semantic similarity algorithms to process equivalent expressions in the documents.

[0011] Difference analysis: The system summarizes all comparison results, automatically identifies differences, and generates optimization suggestions based on preset business rules.

[0012] The report generation process generates a visual comparison report based on the comparison results. It uses a side-by-side comparison view, marks the differences, and embeds contextual information. Users can click on the differences to view the original paragraphs, and the report is then pushed to relevant procurement personnel.

[0013] Furthermore, in the document preprocessing, if the purchased document is an image document, the recognition engine performs text recognition and combines layout analysis technology to distinguish document regions.

[0014] Furthermore, the document preprocessing also includes data cleaning, removing irrelevant characters, watermarks, headers, and footers, standardizing key fields into a unified format, and generating structured intermediate data.

[0015] Furthermore, the methods for marking discrepancies in the generated report include, but are not limited to, adding red highlighting, strikethrough, or underline.

[0016] An intelligent online document comparison system for supply chain procurement scenarios includes:

[0017] Document receiving client, used for users to upload procurement documents;

[0018] The preprocessing module is used to convert the procurement document format, specifically optical character recognition image-based procurement documents, by standardizing the data in the procurement documents into a unified format and generating structured documents.

[0019] The intelligent comparison and analysis module includes an analysis engine and an intelligent comparison engine. The analysis engine loads a pre-trained natural language processing model and a supply chain domain lexicon. The pre-trained natural language processing model performs semantic analysis on the structured documents output by the pre-processing engine to identify and extract key business entities. The intelligent comparison engine establishes field mapping relationships between the structured document content type and business logic and the supply chain domain lexicon, and performs cross-document comparison. It also uses a semantic similarity algorithm to process equivalent expressions based on the supply chain domain lexicon.

[0020] The report generator produces visual reports based on the output of the intelligent comparative analysis module.

[0021] The rules engine has built-in procurement business rules and outputs optimization suggestions;

[0022] The persistent storage module is used to store the original documents, intermediate data, and comparison results.

[0023] Furthermore, the preprocessing module includes,

[0024] The format conversion submodule is used to convert unstructured or semi-structured documents such as PDF, Word, and scanned documents into a processable text format.

[0025] The text extraction submodule uses optical character recognition technology to process scanned images and combines it with layout analysis technology to extract text content;

[0026] The data standardization submodule cleans and removes noise from the extracted text, and converts key fields into a unified data format.

[0027] Furthermore, the supply chain domain thesaurus in the intelligent comparison and analysis module includes at least professional terms, material coding systems, supplier names, and contract term templates in the supply chain procurement scenario, and supports dynamic updates and expansions.

[0028] Furthermore, the intelligent comparison and analysis module uses a pre-trained language model based on the Transformer architecture as its machine learning model, and fine-tunes it using an annotated procurement document dataset.

[0029] Furthermore, the visual comparison report adopts a two-column or multi-column comparison view.

[0030] Furthermore, the optimization suggestions are generated based on a preset rule engine, which includes price deviation threshold, delivery cycle reasonableness, and supplier qualification compliance. The rule engine automatically matches rules based on the differences and outputs optimization prompts.

[0031] Compared with the prior art, the present invention provides an energy storage cabinet with replaceable batteries, which has the following advantages:

[0032] 1. Significantly improve efficiency: Automated processing replaces manual comparison, reducing document comparison time from hours to minutes.

[0033] 2. High accuracy: Pre-trained natural language preprocessing models can accurately identify semantic differences using machine learning techniques, reducing errors caused by human negligence.

[0034] 3. Enhance intelligence: The system can not only "see" the differences, but also "understand" the business implications of the differences and provide decision support.

[0035] 4. Reduce procurement risks: promptly identify issues such as changes in contract terms and price anomalies to prevent supply chain risks.

[0036] 5. Easy to integrate: It can be used as a standalone service or embedded in existing supply chain management systems (SCM) / procurement management systems (SRM), achieving seamless integration with systems such as enterprise resource planning systems (ERP). Attached Figure Description

[0037] Figure 1 This is a schematic diagram illustrating the architecture and workflow of the present invention; Detailed Implementation

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

[0039] Please see Figure 1 A method for intelligent online document comparison in a supply chain procurement scenario includes the following steps:

[0040] Document preprocessing (DP) involves users uploading procurement documents to the document receiving client (UC). The preprocessing engine converts the procurement documents into a format that is processed into a text stream. If the procurement document is an image document, an optical character recognition (OCR) engine is used for text recognition, combined with layout analysis technology to distinguish document areas. The process also includes data cleaning, which removes irrelevant characters, watermarks, headers, and footers, standardizes key fields into a unified format, and generates structured intermediate data.

[0041] Intelligent Comparison Analysis: The Intelligent Comparison Analysis (ICA) module loads a pre-trained Natural Language Processing (NPL) model and a supply chain domain lexicon. The NPL model performs semantic analysis on the structured documents output by the preprocessing engine to identify and extract key business entities.

[0042] The Intelligent Comparison Engine (IC Engine) establishes field mapping relationships with the supply chain domain thesaurus based on the document type and business logic of the text stream, performs cross-document comparison, and processes equivalent expressions using semantic similarity algorithms based on the supply chain domain thesaurus.

[0043] The Difference Analysis and Report Generation (DARG) module uses a report generator to summarize the output results of the intelligent comparison analysis module and generate a visual report in HTML or PDF format. It adopts a side-by-side comparison view, marks the differences, and the marking methods for the differences include, but are not limited to, adding red highlighting, strikethrough, or underline; embeds contextual information, and allows you to view the original paragraph by clicking on the difference item; and enables the Business Rule Engine to generate optimization suggestions based on preset business rules and push them to relevant procurement personnel.

[0044] The original documents, intermediate data, and comparison results are stored using persistent storage.

[0045] According to the accompanying drawings, the specific workflow of the present invention is as follows:

[0046] 1. The purchasing agent executes step 0101 and uploads two purchase contracts;

[0047] 2. The system automatically performs preprocessing 0201. If the preprocessing fails, it returns to 0203 to perform preprocessing again. If the preprocessing is successful, it extracts structured data and enters the Natural Language Processing (NLP) model 03.

[0048] 3. The Natural Language Processing (NLP) model starts recognition at 0301. If recognition fails at 0303, it returns to 03. If recognition is successful, it proceeds to 0302. For example, if the unit price of "[Material A]" in two contracts is identified as "100 yuan" and "105 yuan" respectively, it proceeds to 04.

[0049] 4. The comparison engine (IC Engine) starts the comparison at 0401. If the comparison fails at 0403, it returns to 04. If the comparison succeeds at 0402, it marks the price difference and enters the intelligent comparison engine (IC Engine) at 05.

[0050] 5. The Report Generator executes step 0501 to create a comparison report. If creation fails, step 0503 returns to the Report Generator (05). If creation is successful, step 0502 highlights the price difference and quotes the original text, proceeds to step 06, and returns the report (0504) to the client (UC) (01).

[0051] 6. Rule Engine 06 executes 0601 to make rule judgments, for example: judging that the price has increased by 5%, triggering a "price deviation" warning, suggesting review, executing 0602 to return to the client (UC) 01, and synchronously executing 0601 to record to persistent storage 07.

[0052] 7. The purchasing staff can view the report and recommendations in the system and make a decision.

[0053] This invention automates document comparison, reducing the time from hours to minutes and significantly improving efficiency. The pre-trained natural language processing model uses machine learning to accurately identify semantic differences, reducing human error and improving accuracy. The natural language processing model continuously trains and expands its lexicon, enabling it not only to "see" differences but also to "understand" their business implications and provide decision support, enhancing its intelligence. Visualized reports mark discrepancies, promptly identifying contract changes, price anomalies, and other issues, mitigating supply chain risks and reducing procurement risks. This system can be used as a standalone service or embedded in existing supply chain management (SCM) / procurement management (SRM) systems, achieving seamless integration with enterprise resource planning (ERP) systems and other similar systems.

[0054] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0055] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent online document comparison in a supply chain procurement scenario, characterized in that, Includes the following steps: Document preprocessing: Users upload procurement documents, and the system converts the procurement documents into a processable text stream. Intelligent comparative analysis performs semantic analysis on the preprocessed text stream, identifies and extracts key business entities, establishes field mapping relationships based on the document type and business logic of the text stream, performs cross-document comparison, and uses semantic similarity algorithms to process equivalent expressions in the documents. Difference analysis: The system summarizes all comparison results, automatically identifies differences, and generates optimization suggestions based on preset business rules. The report generation process generates a visual comparison report based on the comparison results. It uses a side-by-side comparison view, marks the differences, and embeds contextual information. Users can click on the differences to view the original paragraphs, and the report is then pushed to relevant procurement personnel.

2. The intelligent online document comparison method in a supply chain procurement scenario as described in claim 1, characterized in that, If the purchased document is an image document during the document preprocessing, the recognition engine performs text recognition and combines layout analysis technology to distinguish document regions.

3. The intelligent online document comparison method in a supply chain procurement scenario as described in claim 1, characterized in that, The document preprocessing also includes data cleaning, removing irrelevant characters, watermarks, headers, and footers, standardizing key fields into a unified format, and generating structured intermediate data.

4. The intelligent online document comparison method in a supply chain procurement scenario as described in claim 1, characterized in that, The methods for marking discrepancies in the generated report include, but are not limited to, adding red highlighting, strikethrough, or underline.

5. An intelligent online document comparison system for supply chain procurement scenarios, characterized in that, include: Document receiving client, used for users to upload procurement documents; The preprocessing module is used to convert the procurement document format, specifically optical character recognition image-based procurement documents, by standardizing the data in the procurement documents into a unified format and generating structured documents. The intelligent comparison and analysis module includes an analysis engine and an intelligent comparison engine. The analysis engine loads a pre-trained natural language processing model and a supply chain domain lexicon. The pre-trained natural language processing model performs semantic analysis on the structured documents output by the pre-processing engine to identify and extract key business entities. The intelligent comparison engine establishes field mapping relationships between the structured document content type and business logic and the supply chain domain lexicon, and performs cross-document comparison. It also uses a semantic similarity algorithm to process equivalent expressions based on the supply chain domain lexicon. The report generator produces visual reports based on the output of the intelligent comparative analysis module. The rules engine has built-in procurement business rules and outputs optimization suggestions; The persistent storage module is used to store the original documents, intermediate data, and comparison results.

6. The intelligent online document comparison system for supply chain procurement scenarios as described in claim 5, characterized in that, The preprocessing module includes, The format conversion submodule is used to convert unstructured or semi-structured documents such as PDF, Word, and scanned documents into a processable text format. The text extraction submodule uses optical character recognition technology to process scanned images and combines it with layout analysis technology to extract text content; The data standardization submodule cleans and removes noise from the extracted text, and converts key fields into a unified data format.

7. The intelligent online document comparison system for supply chain procurement scenarios as described in claim 5, characterized in that, The intelligent comparison and analysis module's supply chain domain thesaurus includes at least professional terms, material coding systems, supplier names, and contract term templates for supply chain procurement scenarios, and supports dynamic updates and expansion.

8. The intelligent online document comparison system for supply chain procurement scenarios as described in claim 5, characterized in that, The intelligent comparison and analysis module uses a pre-trained language model based on the Transformer architecture as its machine learning model, and fine-tunes it using an annotated procurement document dataset.

9. The intelligent online document comparison system for supply chain procurement scenarios as described in claim 5, characterized in that: The visual comparison report uses a two-column or multi-column comparison view.

10. The intelligent online document comparison system for supply chain procurement scenarios as described in claim 5, characterized in that: The optimization suggestions are generated based on a preset rule engine, which includes price deviation threshold, delivery cycle reasonableness, and supplier qualification compliance. The rule engine automatically matches rules based on the differences and outputs optimization prompts.

Images