Intelligent purchase contract comparison method and device in cloud environment

Abstract

The invention relates to the technical field of contract processing, and particularly provides an intelligent purchase contract comparison method and device in a cloud environment, and the method comprises the steps: firstly carrying out the data acquisition in a data storage layer, carrying out the format conversion, text preprocessing, feature extraction and similarity calculation in a data processing layer, and finally carrying out the similar contract screening in an application layer. Compared with the prior art, the method can help enterprises to optimize purchasing strategies and reduce purchasing risks, and is of great significance to digital transformation and fine management of enterprise purchasing services.

Description

Technical Field

[0001] This invention relates to the field of contract processing technology, specifically providing a method and apparatus for intelligent procurement contract comparison in a cloud environment. Background Technology

[0002] In the corporate procurement process, the procurement contract, as a key document clarifying the rights and obligations of both parties, directly impacts the company's operating costs, risk management, and economic benefits through its accuracy and efficiency. Currently, traditional manual comparison methods for processing new procurement contracts require a significant amount of manpower and time for word-by-word review. This is not only extremely inefficient but also prone to errors such as omissions or misunderstandings due to human negligence, making it difficult to guarantee the reasonableness and completeness of the contract terms.

[0003] With the rapid development and widespread application of cloud computing technology, enterprises are increasingly storing massive amounts of procurement contract data in the cloud, achieving centralized management and convenient storage of contracts. However, the current lack of efficient and intelligent contract comparison technology makes it difficult for enterprises to quickly and accurately retrieve similar cases from the vast amounts of historical contracts stored in the cloud. This situation prevents enterprises from fully utilizing historical experience during contract review, increasing the difficulty and risk of contract management. Therefore, there is an urgent need for an intelligent procurement contract comparison method based on a cloud environment to meet the needs of enterprises' digital and intelligent procurement management. Summary of the Invention

[0004] This invention addresses the shortcomings of the prior art by providing a highly practical intelligent procurement contract comparison method in a cloud environment.

[0005] A further technical objective of this invention is to provide a reasonably designed, safe, and applicable intelligent procurement contract comparison device in a cloud environment.

[0006] The technical solution adopted by this invention to solve its technical problem is:

[0007] A method for intelligent procurement contract comparison in a cloud environment involves first acquiring data at the data storage layer, then performing format conversion, text preprocessing, feature extraction, and similarity calculation at the data processing layer, and finally filtering similar contracts at the application layer.

[0008] Furthermore, the data storage layer utilizes the cloud platform interface to obtain structured and unstructured new and historical procurement contract data from the data storage layer. New procurement contracts are electronic documents newly signed or received by the enterprise, while historical procurement contracts are contract data previously stored in the cloud and stored in the cloud platform database.

[0009] Furthermore, the format conversion involves transmitting the acquired contract documents in different formats to the data processing layer for format conversion, transforming them into text format;

[0010] The converted text undergoes text preprocessing, including text cleaning and word segmentation. The text cleaning removes special characters, garbled text, and blank lines, unifies full-width and half-width characters, and standardizes the encoding format.

[0011] The word segmentation process uses a natural language processing algorithm to divide the text into words or phrases according to semantic and grammatical rules.

[0012] Furthermore, feature extraction is divided into keyword extraction and semantic feature extraction. The keyword extraction uses the TF-IDF algorithm and combines it with a keyword library of the procurement industry to extract keywords that can represent the core content of the contract.

[0013] The semantic feature extraction process involves inputting the contract text into the deep learning model BERT, using semantic analysis of the text to capture word context relationships and deep semantic information, and mapping the contract text into a high-dimensional semantic feature vector.

[0014] Furthermore, the similarity calculation includes:

[0015] (1) Keyword similarity calculation;

[0016] (2) Semantic similarity calculation;

[0017] (3) Calculation of overall similarity.

[0018] Furthermore, in step (1), the keywords extracted from the new contract and the historical contract are converted into vector form, and the cosine similarity algorithm is used to calculate the cosine value of the angle between the keyword vectors to measure the similarity of the keyword set.

[0019] In step (2), the cosine similarity algorithm is also used to calculate the semantic similarity of the semantic feature vectors of the new contract and the historical contract to determine the degree of similarity of the contracts at the semantic level.

[0020] In step (3), the keyword similarity and semantic similarity are weighted and summed in the similarity calculation according to the preset keyword similarity weight and semantic similarity weight to obtain the comprehensive similarity between the new contract and the historical contract.

[0021] Furthermore, the application layer similar contract filtering includes:

[0022] A. Receive comprehensive similarity data between the new contract and all historical contracts from the similarity calculation output. After receiving the data, the application layer will parse and verify it to ensure the integrity and accuracy of the data.

[0023] B. The application layer has a built-in rule engine-based filtering mechanism that compares the overall similarity of all historical contracts one by one according to a pre-set similarity threshold.

[0024] C. Application layer sets a dynamic adjustment mechanism for similarity thresholds.

[0025] A cloud-based intelligent procurement contract comparison device includes: at least one memory and at least one processor;

[0026] The at least one memory is used to store a machine-readable program;

[0027] The at least one processor is used to call the machine-readable program to execute a smart procurement contract comparison method in a cloud environment.

[0028] Compared with existing technologies, the intelligent procurement contract comparison method and apparatus in a cloud environment of the present invention have the following outstanding advantages:

[0029] This invention leverages the distributed computing and parallel processing capabilities of cloud computing to rapidly process massive amounts of contract data. This significantly shortens the contract processing cycle, enabling enterprises to accelerate the procurement process more promptly and improve overall business efficiency.

[0030] This invention reduces reliance on manual labor through an automated intelligent comparison process. Enterprises no longer need to employ a large number of staff for tedious contract comparison work, thus lowering labor costs. Simultaneously, the rapid comparison speed saves significant time, allowing enterprises to invest more resources and energy in core business development, optimizing resource allocation and improving economic efficiency. Employing multi-dimensional intelligent comparison technologies such as keyword extraction and semantic feature analysis, it can comprehensively and deeply analyze contract content. Whether it's key data or complex semantic logic in the contract, it can accurately identify and compare it, avoiding errors caused by human negligence, ensuring the accuracy of contract comparison results, and providing a reliable basis for enterprise contract review.

[0031] This invention, through precise contract comparison, helps companies promptly identify potential risks in new contracts. By referencing the successes and failures of similar historical contracts, companies can optimize and improve the terms of new contracts, reducing contract risks and protecting their legitimate rights and economic interests.

[0032] By screening and analyzing similar contracts, companies can summarize procurement patterns, understand the cooperation status of different suppliers, and grasp information such as market price fluctuation trends. This data provides strong support for companies to formulate scientific procurement strategies, optimize supplier management, and reasonably control procurement costs, helping them achieve refined procurement management and improve the scientific and rational nature of procurement decisions.

[0033] This invention can fully tap into the potential information within contract data, providing valuable decision-making references for enterprises through the analysis and comparison of historical contracts. Furthermore, combining this data with other business data can further expand data application scenarios, enhance the value of enterprise data assets, and drive digital transformation and innovative development. Attached Figure Description

[0034] To more clearly illustrate the technical solutions in the embodiments of the present 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 the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 This is a schematic diagram of the data storage layer in an intelligent procurement contract comparison method in a cloud environment.

[0036] Figure 2 This is a schematic diagram of the data processing layer in an intelligent procurement contract comparison method in a cloud environment.

[0037] Figure 3 This is a schematic diagram of the application layer in an intelligent procurement contract comparison method in a cloud environment. Detailed Implementation

[0038] To enable those skilled in the art to better understand the present invention, the present invention will be further described in detail below with reference to specific embodiments. Obviously, the described embodiments are merely 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] The following is a preferred embodiment:

[0040] This embodiment describes an intelligent procurement contract comparison method in a cloud environment. First, data is acquired at the data storage layer, and then format conversion, text preprocessing, feature extraction, and similarity calculation are performed at the data processing layer. Finally, similar contracts are screened at the application layer.

[0041] like Figure 1 As shown, data acquisition at the data storage layer: During the contract acquisition phase, the cloud platform interface is used to retrieve new and historical procurement contract data from the data storage layer. New procurement contracts are electronic documents newly signed or received by the enterprise, while historical procurement contracts are contract data previously stored in the cloud, encompassing various formats such as structured (e.g., database table records) and unstructured (e.g., original documents), and are stored in the cloud platform database.

[0042] like Figure 2 As shown, the acquired contract documents in different formats are transmitted to the data processing layer for format conversion. Through the integrated format conversion tool, these documents are uniformly converted into text format. For example, PDF to text tools are used to process PDF contracts, and text content is extracted from Word documents to lay the foundation for subsequent processing.

[0043] The converted text enters the text preprocessing stage. Text cleaning removes special characters, garbled text, blank lines, and other interfering information, unifies full-width and half-width characters, and standardizes the encoding format. Word segmentation uses a natural language processing algorithm to divide the text into words or phrases according to semantic and grammatical rules, facilitating subsequent feature extraction.

[0044] Feature extraction includes keyword extraction and semantic feature extraction;

[0045] Keyword extraction: After word segmentation, the text enters the feature extraction module. Using the TF-IDF algorithm and combined with the procurement industry keyword library, keywords that can represent the core content of the contract are extracted to ensure that the extracted keywords accurately reflect the key information of the contract, such as the name, quantity, and price of the purchased products.

[0046] Semantic feature extraction: The contract text is input into the deep learning model BERT, which performs semantic analysis on the text, captures word context relationships and deep semantic information, and maps the contract text into a high-dimensional semantic feature vector.

[0047] Similarity calculation includes:

[0048] (1) Keyword similarity calculation: Convert the keywords extracted from the new contract and the historical contract into vector form, and use the cosine similarity algorithm in the similarity calculation module to calculate the cosine value of the angle between the keyword vectors, thereby measuring the similarity of the keyword set.

[0049] (2) Semantic similarity calculation: For the semantic feature vectors of new contracts and historical contracts, the cosine similarity algorithm is also used to calculate the semantic similarity and determine the degree of similarity between the contracts at the semantic level.

[0050] (3) Comprehensive similarity calculation: Based on the preset keyword similarity weight and semantic similarity weight, the keyword similarity and semantic similarity are weighted and summed in the similarity calculation module to obtain the comprehensive similarity between the new contract and the historical contract.

[0051] like Figure 3 As shown, the application layer similar contract filtering includes:

[0052] A. Through the system's internal data transmission interface, it receives comprehensive similarity data between new contracts and all historical contracts output from the similarity calculation module. This data is presented in a structured format, including the unique identifier of each historical contract, its corresponding comprehensive similarity value, and a summary of basic contract information. Upon receiving the data, the application layer parses and verifies it to ensure its integrity and accuracy.

[0053] B. The application layer has a built-in rule engine-based filtering mechanism that compares the overall similarity of all historical contracts one by one according to a pre-set similarity threshold.

[0054] C. Considering the complexity and diversity of enterprise procurement operations, the application layer has designed a comprehensive dynamic adjustment mechanism for similarity thresholds.

[0055] Based on the above method, a cloud-based intelligent procurement contract comparison device in this embodiment includes: at least one memory and at least one processor;

[0056] The at least one memory is used to store a machine-readable program;

[0057] The at least one processor is used to call the machine-readable program to execute a smart procurement contract comparison method in a cloud environment.

[0058] 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 or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for intelligent comparison of procurement contracts in a cloud environment, characterized in that, Firstly, data acquisition is performed at the data storage layer, format conversion, text preprocessing, feature extraction and similarity calculation are performed at the data processing layer, and finally, similar contract screening is performed at the application layer. 2.The intelligent comparison method of procurement contract in cloud environment according to claim 1, characterized in that, The data storage layer utilizes a cloud platform interface to obtain structured and unstructured form new procurement contract and historical procurement contract data from the data storage layer. The new procurement contract is an electronic document newly signed or received by the enterprise, and the historical procurement contract is the contract data stored in the cloud by the enterprise in the past. The data is stored in the cloud platform database. 3.The intelligent comparison method of procurement contract in cloud environment according to claim 2, characterized in that, The format conversion is to transmit the obtained contract documents in different formats to the data processing layer for format conversion into text format. The converted text is subjected to text preprocessing, including text cleaning and word segmentation processing. The text cleaning removes special characters, garbled characters and blank lines, unifies full-width and half-width characters, and standardizes the encoding format. The word segmentation processing adopts a natural language processing word segmentation algorithm to divide the text into words or phrases according to semantic and grammatical rules.

4. The intelligent comparison method of the contract of purchase in the cloud environment according to claim 3, characterized in that, In feature extraction, keyword extraction and semantic feature extraction are included. The keyword extraction uses the TF-IDF algorithm and combines with the procurement industry keyword library to extract keywords that can represent the core content of the contract. In the semantic feature extraction, the contract text is input into the deep learning model BERT to perform semantic analysis on the text, capture the context relationship and deep semantic information of the words, and map the contract text into a high-dimensional semantic feature vector.

5. The intelligent comparison method of the contract of purchase in the cloud environment according to claim 4, characterized in that, In the similarity calculation, the following are included: (1) keyword similarity calculation; (2) semantic similarity calculation; (3) comprehensive similarity calculation. 6.The intelligent comparison method of procurement contract in cloud environment according to claim 5, characterized in that, In step (1), the keywords extracted from the new contract and the historical contract are converted into vector form, and the cosine similarity algorithm is used to calculate the cosine value of the included angle between the keyword vectors to measure the similarity of the keyword set. In step (2), the semantic feature vectors of the new contract and the historical contract are also subjected to cosine similarity algorithm to calculate the semantic similarity and determine the similarity degree of the contract at the semantic level. In step (3), the keyword similarity and the semantic similarity are weighted and summed according to the preset keyword similarity weight and semantic similarity weight in the similarity calculation to obtain the comprehensive similarity of the new contract and the historical contract.

7. The intelligent comparison method of a procurement contract in a cloud environment according to claim 6, characterized in that, In the application layer similar contract screening, the following are included: A. Receive the comprehensive similarity data of the new contract and all historical contracts from the similarity calculation output. The application layer will analyze and verify the data after receiving it to ensure the integrity and accuracy of the data. B. The application layer is built-in with a screening mechanism based on a rule engine, which compares the comprehensive similarity of all historical contracts one by one according to the pre-set similarity threshold. C. The application layer sets a dynamic adjustment mechanism for the similarity threshold.

8. An intelligent procurement contract comparison device under a cloud environment, characterized in that, At least one memory and at least one processor; The at least one memory is used to store machine-readable programs; The at least one processor is used to call the machine-readable programs to execute the method of any one of claims 1 to 7. ​

Images