Bidding and tendering data screening recommendation method

Abstract

The invention discloses a bidding and tendering data screening recommendation method, and relates to the technical field of data governance and large model application. Comprising the steps of (1) collecting bidding and tendering data, (2) preprocessing the bidding and tendering data, (3) carrying out keyword extraction on the preprocessed bidding and tendering data, (4) constructing a business opportunity evaluation model, and (5) recommending bidding and tendering projects with higher scores to enterprises and providing project analysis reports according to scoring results of the business opportunity evaluation model. And 6, monitoring the found bidding and tendering project in real time, obtaining the latest progress and change information of the bidding and tendering project in time, and dynamically adjusting the evaluation result.

Description

Technical Field

[0001] The invention discloses a method for screening and recommending bidding data, and relates to the technical field of data governance and large model application. Background Art

[0002] Bidding is one of the important ways for enterprises to obtain projects and businesses. However, existing bidding data discovery methods usually rely on manual search and analysis, which is inefficient and easy to miss important information. With the development of information technology, although some bidding information platforms have emerged, most of these platforms simply provide information listing and lack effective data mining and analysis functions. Summary of the invention

[0003] In view of the problems of the prior art, the present invention provides a bidding data screening and recommendation method, which extracts keywords from a large amount of non-standardized announcement text information through data collection and data governance processes, thereby evaluating and inferring the bidding data, and can help enterprises quickly and accurately discover potential bidding data and improve their market competitiveness.

[0004] The specific scheme proposed by the present invention is:

[0005] The present invention provides a bidding data screening and recommendation method, comprising:

[0006] Step 1: Collect bidding data, including project name, bidding unit, project content, budget amount, and bid opening time.

[0007] Step 2: Preprocess bidding data,

[0008] Step 3: Extract keywords from the preprocessed bidding data: Use the general big model to extract key data from the bidding data. The key data includes industry keywords, product keywords, and regional keywords. The existing open source general big model is used as the basic model to extract key data from the bidding projects.

[0009] Step 4: Build a business opportunity assessment model:

[0010] Utilize key data and establish a business opportunity evaluation model based on the company's own business characteristics. Use the business opportunity evaluation model to evaluate and score each bidding project based on project budget, bidding unit reputation, and matching degree with the company's business.

[0011] Step 5: Based on the scoring results of the business opportunity evaluation model, recommend the bidding projects with higher scores to the enterprises and provide project analysis reports.

[0012] Step 6: Conduct real-time monitoring of discovered bidding projects, obtain the latest progress and change information of bidding projects in a timely manner, and dynamically adjust the evaluation results.

[0013] Furthermore, in step 1 of the bidding data screening and recommendation method, web crawler technology is used to collect bidding data from bidding websites, public resource trading platforms and related data sources.

[0014] Furthermore, in step 2 of the bidding data screening and recommendation method, the bidding data is preprocessed, including: cleaning and organizing the collected original bidding data, removing duplicate, invalid and erroneous data, and converting the data into a unified format.

[0015] Furthermore, in step 3 of the bidding data screening and recommendation method, a small amount of samples with labeled information are constructed, the basic model is fine-tuned by LORA, and the generated LORA fine-tuning is merged into the original basic model.

[0016] The present invention also provides a bidding data screening and recommendation device, comprising a collection module, a preprocessing module, a large model extraction module, a model management module, a recommendation module and a monitoring and adjustment module.

[0017] The collection module collects bidding data, which includes project name, bidding unit, project content, budget amount, and bid opening time.

[0018] The preprocessing module preprocesses the bidding data.

[0019] The big model extraction module extracts keywords from the preprocessed bidding data: it uses the general big model to extract key data from the bidding data. The key data includes industry keywords, product keywords, and regional keywords. The existing open source general big model is used as the basic model to extract key data from the bidding projects.

[0020] The model management module builds a business opportunity assessment model:

[0021] Utilize key data and establish a business opportunity evaluation model based on the company's own business characteristics. Use the business opportunity evaluation model to evaluate and score each bidding project based on project budget, bidding unit reputation, and matching degree with the company's business.

[0022] The recommendation module recommends high-scoring bidding projects to enterprises and provides project analysis reports based on the scoring results of the business opportunity evaluation model.

[0023] The monitoring and adjustment module monitors the discovered bidding projects in real time, promptly obtains the latest progress and change information of the bidding projects, and dynamically adjusts the evaluation results.

[0024] Furthermore, the collection module of the bidding data screening and recommendation device uses web crawler technology to collect bidding data from bidding websites, public resource trading platforms and related data sources.

[0025] Furthermore, the preprocessing module of the bidding data screening and recommendation device preprocesses the bidding data, including: cleaning and sorting the collected original bidding data, removing duplicate, invalid and erroneous data, and converting the data into a unified format.

[0026] Furthermore, the large model extraction module of the bidding data screening and recommendation device constructs a small number of samples with labeled information, performs LORA fine-tuning on the basic model, and merges the generated LORA fine-tuning into the original basic model.

[0027] The benefits of the method of the present invention are:

[0028] (1) It can help enterprises quickly and accurately identify potential bidding opportunities.

[0029] (2) It can effectively reduce manual participation and save labor.

[0030] (3) It greatly improves processing efficiency and can automatically mine valuable business opportunity information from large amounts of bidding data.

[0031] (4) Enable enterprises to improve their market competitiveness and obtain greater profits. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a schematic flow chart of the method of the present invention. DETAILED DESCRIPTION

[0033] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0034] Example 1

[0035] The present invention provides a bidding data screening and recommendation method, comprising:

[0036] Step 1: Collect bidding data, which includes project name, bidding unit, project content, budget amount, and bid opening time.

[0037] Web crawler technology can be used to collect bidding data from bidding websites, public resource trading platforms and related data sources.

[0038] Step 2: Preprocessing the bidding data. This may specifically include: cleaning and sorting the collected raw bidding data, removing duplicate, invalid and erroneous data, and converting the data into a unified format. For example, standardizing inconsistent project names on different websites.

[0039] Step 3: Extract keywords from the preprocessed bidding data: Use the general large model to extract key data from the bidding data. The key data includes industry keywords, product keywords, and regional keywords. The existing open source general large model is used as the basic model. The basic model is used to extract key data from the bidding project, such as qwen, chatglm, etc. According to resource allocation, you can choose 1.5b, 7b, 14b and other parameter scales. Use the language understanding and code understanding capabilities of the open source general large model to extract keywords from the bidding data information. You can also construct a small number of samples with labeled information, perform LORA fine-tuning on the basic model, and merge the generated LORA fine-tuning into the original basic model.

[0040] Step 4: Build a business opportunity assessment model:

[0041] Utilize key data and establish a business opportunity evaluation model based on the company's own business characteristics. Use the business opportunity evaluation model to evaluate and score each bidding project based on project budget, bidding unit reputation, and the degree of match with the company's business. For example, higher scores will be given to projects with higher budgets, good bidding unit reputation, and a high degree of match with the company's business.

[0042] Step 5: Based on the scoring results of the business opportunity evaluation model, recommend the bidding projects with higher scores to the enterprises and provide a project analysis report, which may include project background, competition situation, potential risks, etc.

[0043] Step 6: Conduct real-time monitoring of discovered bidding projects, obtain the latest progress and change information of bidding projects in a timely manner, and dynamically adjust the evaluation results.

[0044] The method embodiment of the present invention can automatically mine valuable business opportunity information from a large amount of bidding data efficiently and accurately, and recommend it to enterprises.

[0045] Example 2

[0046] The present invention also provides a bidding data screening and recommendation device, comprising a collection module, a preprocessing module, a large model extraction module, a model management module, a recommendation module and a monitoring and adjustment module.

[0047] The collection module collects bidding data, which includes project name, bidding unit, project content, budget amount, and bid opening time.

[0048] The preprocessing module preprocesses the bidding data.

[0049] The big model extraction module extracts keywords from the preprocessed bidding data: it uses the general big model to extract key data from the bidding data. The key data includes industry keywords, product keywords, and regional keywords. The existing open source general big model is used as the basic model to extract key data from the bidding projects.

[0050] The model management module builds a business opportunity assessment model:

[0051] Utilize key data and establish a business opportunity evaluation model based on the company's own business characteristics. Use the business opportunity evaluation model to evaluate and score each bidding project based on project budget, bidding unit reputation, and matching degree with the company's business.

[0052] The recommendation module recommends high-scoring bidding projects to enterprises and provides project analysis reports based on the scoring results of the business opportunity evaluation model.

[0053] The monitoring and adjustment module monitors the discovered bidding projects in real time, promptly obtains the latest progress and change information of the bidding projects, and dynamically adjusts the evaluation results.

[0054] As the information interaction and execution process between the modules in the above-mentioned device are based on the same concept as the embodiment of the method of the present invention, the specific contents can be found in the description of the embodiment of the method of the present invention and will not be repeated here.

[0055] Likewise, the benefits of the device of the present invention are:

[0056] (1) It can help enterprises quickly and accurately identify potential bidding opportunities.

[0057] (2) It can effectively reduce manual participation and save labor.

[0058] (3) It greatly improves processing efficiency and can automatically mine valuable business opportunity information from large amounts of bidding data.

[0059] (4) Enable enterprises to improve their market competitiveness and obtain greater profits.

[0060] It should be noted that not all steps and modules in the above-mentioned processes and device structures are necessary, and some steps or modules can be ignored according to actual needs. The execution order of each step is not fixed and can be adjusted as needed. The system structure described in the above-mentioned embodiments can be a physical structure or a logical structure, that is, some modules may be implemented by the same physical entity, or some modules may be implemented by multiple physical entities, or some components in multiple independent devices may be implemented together.

[0061] The above-described embodiments are only preferred embodiments for fully illustrating the present invention, and the protection scope of the present invention is not limited thereto. Equivalent substitutions or changes made by those skilled in the art based on the present invention are within the protection scope of the present invention. The protection scope of the present invention shall be subject to the claims.

Claims

1. A method for screening and recommending bidding data, characterized in that include: Step 1: Collect bidding data, including project name, bidding unit, project content, budget amount, and bid opening time. Step 2: Preprocess bidding data, Step 3: Extract keywords from the preprocessed bidding data: Use the general big model to extract key data from the bidding data. The key data includes industry keywords, product keywords, and regional keywords. The existing open source general big model is used as the basic model to extract key data from the bidding projects. Step 4: Build a business opportunity assessment model: Utilize key data and establish a business opportunity evaluation model based on the company's own business characteristics. Use the business opportunity evaluation model to evaluate and score each bidding project based on project budget, bidding unit reputation, and matching degree with the company's business. Step 5: Based on the scoring results of the business opportunity evaluation model, recommend the bidding projects with higher scores to the enterprises and provide project analysis reports. Step 6: Conduct real-time monitoring of discovered bidding projects, obtain the latest progress and change information of bidding projects in a timely manner, and dynamically adjust the evaluation results.

2. The method for screening and recommending bidding data according to claim 1, characterized in that In step 1, web crawler technology is used to collect bidding data from bidding websites, public resource trading platforms and related data sources.

3. The method for screening and recommending bidding data according to claim 1, characterized in that The bidding data is preprocessed in step 2, including: cleaning and organizing the collected original bidding data, removing duplicate, invalid and erroneous data, and converting the data into a unified format.

4. The method for screening and recommending bidding data according to claim 1, characterized in that In step 3, a small amount of samples with labeled information are constructed, the basic model is fine-tuned by LoRa, and the generated LoRa fine-tuning is merged into the original basic model.

5. A bidding data screening and recommendation device, characterized in that It includes acquisition module, preprocessing module, large model extraction module, model management module, recommendation module and monitoring and adjustment module. The collection module collects bidding data, which includes project name, bidding unit, project content, budget amount, and bid opening time. The preprocessing module preprocesses the bidding data. The big model extraction module extracts keywords from the preprocessed bidding data: it uses the general big model to extract key data from the bidding data. The key data includes industry keywords, product keywords, and regional keywords. The existing open source general big model is used as the basic model to extract key data from the bidding projects. The model management module builds a business opportunity assessment model: Utilize key data and establish a business opportunity evaluation model based on the company's own business characteristics. Use the business opportunity evaluation model to evaluate and score each bidding project based on project budget, bidding unit reputation, and matching degree with the company's business. The recommendation module recommends high-scoring bidding projects to enterprises and provides project analysis reports based on the scoring results of the business opportunity evaluation model. The monitoring and adjustment module monitors the discovered bidding projects in real time, promptly obtains the latest progress and change information of the bidding projects, and dynamically adjusts the evaluation results.

6. The device for screening and recommending bidding data according to claim 5, characterized in that The collection module uses web crawler technology to collect bidding data from bidding websites, public resource trading platforms and related data sources.

7. The device for screening and recommending bidding data according to claim 5, characterized in that The preprocessing module preprocesses the bidding data, including: cleaning and organizing the collected original bidding data, removing duplicate, invalid and erroneous data, and converting the data into a unified format.

8. The device for screening and recommending bidding data according to claim 5, characterized in that The large model extraction module constructs a small number of samples with labeled information, performs LoRa fine-tuning on the basic model, and merges the generated LoRa fine-tuning into the original basic model.

Images