Bid evaluation method and device

Abstract

The invention discloses a bid evaluation method and device. The method comprises the following steps: acquiring bid invitation data, bid evaluation data and bid data to be evaluated from a target e-commerce platform through a preset data interface; according to the bid invitation data, the bid evaluation data and the bidding data, determining an evaluation index of the bidding data; determining a target weight of the evaluation index; inputting the bidding data, the evaluation indexes and the target weights of the evaluation indexes into a first preset bidding evaluation model, and determining candidate bidding schemes in the bidding data; if the number of the candidate bidding schemes is greater than or equal to a first preset threshold value, inputting the candidate bidding schemes, the evaluation indexes and target weights of the evaluation indexes into a second preset bidding evaluation model, and determining a target bidding scheme; and if the number of the candidate bidding schemes is smaller than a first preset threshold value, determining the candidate bidding schemes as target bidding schemes. Through automation technology and process optimization, manual operation and repeated work are reduced, and the efficiency and quality of review work are improved.

Description

Technical Field

[0001] This application relates to the technical field of data processing, and in particular, to a bid evaluation method and device. Background Art

[0002] As a power supplier, the power grid company's review business covers multiple links of power engineering, and a large number of engineering plans, construction drawings, operation data, etc. need to be reviewed.

[0003] In the related art, the power grid company uses an e-commerce platform to obtain and store bid-related data, and then conducts manual review based on the data in the e-commerce platform. This not only has low work efficiency, but also cannot guarantee the fairness and accuracy of the bid evaluation results, cannot well adapt to the market demand and industry development trend, is difficult to meet the business growth and market competition needs, and affects the business development and market competitiveness of the power grid company. Summary of the Invention

[0004] In view of this, this application provides a bid evaluation method and device, which reduce manual operations and repetitive work through automation technology and process optimization, and improve the efficiency and quality of the review work.

[0005] According to one aspect of this application, a bid evaluation method is provided, including: obtaining bid data, bid evaluation data, and bid data to be evaluated from a target e-commerce platform through a preset data interface; determining evaluation indicators of the bid data according to the bid data, the bid evaluation data, and the bid data; determining the target weights of the evaluation indicators; inputting the bid data, the evaluation indicators, and the target weights of the evaluation indicators into a first preset bid evaluation model to determine candidate bid proposals in the bid data; if the number of candidate bid proposals is greater than or equal to a first preset threshold, inputting the candidate bid proposals, the evaluation indicators, and the target weights of the evaluation indicators into a second preset bid evaluation model to determine the target bid proposal; if the number of candidate bid proposals is less than the first preset threshold, determining the candidate bid proposal as the target bid proposal.

[0006] According to another aspect of the present application, there is provided an evaluation device for bid evaluation, including: an acquisition module, configured to acquire tender data, bid evaluation data, and tender data to be evaluated from a target e-commerce platform through a preset data interface; a determination module, configured to determine evaluation indicators for the tender data according to the tender data, the bid evaluation data, and the tender data; and determine target weights for the evaluation indicators; a bid evaluation module, configured to input the tender data, the evaluation indicators, and the target weights of the evaluation indicators into a first preset bid evaluation model to determine candidate tender proposals in the tender data; if the number of candidate tender proposals is greater than or equal to a first preset threshold, input the candidate tender proposals, the evaluation indicators, and the target weights of the evaluation indicators into a second preset bid evaluation model to determine a target tender proposal; if the number of candidate tender proposals is less than the first preset threshold, determine the candidate tender proposal as the target tender proposal.

[0007] According to still another aspect of the present application, there is provided a readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the above-mentioned bid evaluation method are implemented.

[0008] According to yet another aspect of the present application, there is provided a computer device, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, and when the processor executes the program, the steps of the above-mentioned bid evaluation method are implemented.

[0009] By means of the above technical solutions, the present application provides a bid evaluation method and device, which automatically collect bid-related data from the e-commerce platform used by the power grid company through a preset data interface, transfer the data in the target e-commerce platform for processing, and make up for the lack of data processing capabilities of the e-commerce platform. After transferring the data in the target e-commerce platform, deeply mine and analyze the tender data, and combine historical experience and the characteristics of the tender data to determine the target weights of the evaluation indicators, so as to obtain a more scientific and reasonable bid evaluation index weight system, reduce the deviation caused by human factors, and improve the accuracy of the bid evaluation results. Then, according to the target weights and the preset bid evaluation model, improve the fairness and transparency of the evaluation work, help decision-makers determine the optimal tender proposal, improve the bid evaluation efficiency, eliminate the possibility of human intervention in the bid evaluation process to the greatest extent, reduce the risk of abuse of scoring discretion and subjective evaluation, and enhance the objectivity of the scoring results.

[0010] The above description is only an overview of the technical solutions of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features, and advantages of the present application more obvious and understandable, the specific embodiments of the present application are specifically given below. Description of the Drawings

[0011] The accompanying drawings described herein are used to provide a further understanding of the present application, and constitute a part of the present application. The schematic embodiments of the present application and their descriptions are used to explain the present application, and do not constitute an improper limitation of the present application. In the drawings:

[0012] Figure 1 A flowchart showing the evaluation method provided by the embodiment of the present application is shown;

[0013] Figure 2 A block diagram showing the structure of the evaluation device provided by the embodiment of the present application is shown. Detailed implementation manners

[0014] The present application will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments may be combined with each other.

[0015] The embodiments of the present application are described in detail below. The examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the accompanying drawings are exemplary and are only used to explain the present application, and should not be construed as a limitation of the present application.

[0016] Those skilled in the art of the present technology can understand that, unless specifically stated otherwise, the singular forms "a", "an", "the" and "said" used herein may also include the plural forms. It should be further understood that the term "comprising" used in the specification of the present application means the presence of the described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or their groups. It should be understood that when we say that an element is "connected" or "joined" to another element, it can be directly connected or joined to other elements, or there may also be intermediate elements. In addition, the "connection" or "joining" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any unit and all combinations of one or more of the associated listed items.

[0017] Now, exemplary embodiments according to the present application will be described in more detail with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many different forms and should not be construed as being limited only to the embodiments set forth herein. It should be understood that these embodiments are provided so that the disclosure of the present application is thorough and complete, and the concept of these exemplary embodiments is fully conveyed to those of ordinary skill in the art.

[0018] In this embodiment, an evaluation method is provided, as Figure 1 shown, the method includes:

[0019] Step 101: Obtain tender data, bid evaluation data, and tender bid data to be evaluated from the target e-commerce platform through a preset data interface.

[0020] In this embodiment, through the data interface, triggered according to events such as the opening of tenders, basic information and tender proposals of tendering units, project information, supplier qualification information, tender evaluation rules, and other tender data, bid evaluation data, and tender bid data to be evaluated are automatically collected from the ECP (E-Commercial Platform, e-commerce platform) used by the power grid company, and the data in the target e-commerce platform is transferred out, avoiding the time-consuming and errors of manual collection and improving work efficiency. Thus, in subsequent steps, the transferred data is processed according to various requirements in the tendering and bidding business to give play to the value of data assets. Moreover, an automated cleaning tool is used to quickly remove duplicate, incorrect, and incomplete data to form a data set available for intelligent bid evaluation, improving the speed of data processing while ensuring the security and integrity of the data.

[0021] Exemplarily, data exchange and integration are carried out with other enterprise systems or third-party services through API (Application Programming Interface) interface technology, and new integration functions can be easily added. Different systems can achieve data sharing and collaborative work by calling each other's API interfaces. For example, one system can obtain the required data by calling the API of another system, promoting collaboration and information flow between systems. Another example is that during the project creation and setup process, data such as project information, tender conditions, and bid evaluation criteria are entered from the target e-commerce platform through the API interface. In addition, external data sources such as qualification certification data provided by third-party institutions can be introduced. To ensure data quality, strict specifications are formulated for data entry, including requirements for the format of tender documents, standards for internal data entry in the system, and processes.

[0022] Step 102: Determine the evaluation indicators for the tender bid data based on the tender data, bid evaluation data, and tender bid data.

[0023] Among them, the tender bid data includes various tender proposals collected.

[0024] In this embodiment, combined with tender data such as tender announcements and evaluation rules publicly announced in the tendering stage and bid evaluation data, a reasonable and perfect bid evaluation index system is established according to the characteristics and requirements of the project. The evaluation indicators can include price, technical ability, service, delivery time, construction period, construction quality, safety assurance measures, environmental protection measures, and enterprise qualifications, reputation, etc. The determination of evaluation indicators needs to comprehensively consider the actual situation of the project and the preferences of users to ensure the comprehensiveness and scientificity of the evaluation.

[0025] Step 103: Determine the target weights of the evaluation indicators.

[0026] In this embodiment, by combining historical experience and objective data, a more scientific and reasonable bid evaluation index weight system is obtained, reducing the deviation caused by human factors, thereby improving the accuracy and fairness of the bid evaluation results.

[0027] Further, as a refinement and extension of the specific implementation manner of the above embodiment, to fully illustrate the specific implementation process of this embodiment, step 103, that is, the step of determining the target weights of the evaluation indicators, specifically includes: receiving the subjective weights input by the user, determining the first weight of the evaluation indicators according to the subjective weights; determining the second weight of the evaluation indicators according to the bid data; constructing an objective function of the target weights with the goal of minimizing the deviation between the first weight and the second weight; and solving the objective function according to the first weight and the second weight to obtain the value of the target weights of the evaluation indicators.

[0028] In this embodiment, the subjective judgment of the user is introduced to determine the first weight of the evaluation indicators, and the second weight is determined based on the support of objective bid data. Thus, through the method of combined weighting, the deficiencies of each are compensated to obtain a more scientific and reasonable bid evaluation index weight system, reducing the deviation caused by human factors and improving the accuracy of the bid evaluation results.

[0029] Exemplarily, with the goal of minimizing the deviation between the first weight and the second weight, an objective function δ(w) of the target weights is constructed as shown in the following formula:

[0030]

[0031] where α i is the first weight of the i-th evaluation indicator, β i is the second weight of the i-th evaluation indicator, d(α i , β i ) is the distance between the first weight α i and the second weight β i , which is used to measure the degree of difference between the first weight α i and the second weight β i . w i is the target weight of the i-th evaluation indicator, m is the number of evaluation indicators, and N is the set of natural numbers.

[0032] Further, with the help of the linear programming function of the matlab software, the solution of the above objective function is obtained, thereby obtaining the value of the target weights of the comprehensive weighting of the evaluation indicators.

[0033] In this embodiment, the goal is to minimize the degree of difference between the first weight and the second weight, so that the target weight can not only reflect subjective judgment but also reflect objective data, improve the accuracy of the target weight, and thus make the final bid evaluation result more reliable.

[0034] Further, as a refinement and extension of the specific implementation manner of the above embodiment, to fully illustrate the specific implementation process of this embodiment, the steps of determining the first weight of the evaluation index according to the subjective weight specifically include: determining the relative importance degree between the evaluation indexes according to the subjective weight; determining the judgment matrix of the evaluation indexes according to the relative importance degree between the evaluation indexes; calculating the maximum eigenvalue of the judgment matrix; determining the consistency ratio of the judgment matrix according to the maximum eigenvalue; if the consistency ratio is greater than or equal to the second preset threshold, adjusting the relative importance degree until the consistency ratio is less than the second preset threshold; if the consistency ratio is less than the second preset threshold, performing a product operation on the rows of the judgment matrix to determine the product vector of the judgment matrix; performing a square root processing and a normalization processing on the product vector to determine the target vector; and determining the first weight of the evaluation index according to the target vector.

[0035] In this embodiment, the geometric mean processing is performed on the subjective weights of experts and other users by using the Analytic Hierarchy Process (AHP) to ensure that the subjective opinions can reasonably reflect the importance of the evaluation indexes, and the first weight is obtained. And a consistency test is performed on the first weight to ensure the reliability of the result, thereby forming a reasonable and accurate weight system to provide a basis for the subsequent bid evaluation process.

[0036] Exemplarily, first, a hierarchical structure is established, decomposing the bid evaluation problem into multiple levels, including: the goal level, the criterion level, and the scheme level. Among them, the goal level is to select the best bidding scheme, the criterion level includes various evaluation indicators to ensure that the main aspects of evaluating the bidding scheme are covered, and the scheme level is each bidding scheme in the bidding data, so that the bidding schemes are evaluated and compared according to the various evaluation indicators in the criterion level. Then, through pairwise comparison of the various evaluation indicators in the criterion level by experts, the importance degree of one evaluation indicator relative to another is determined, so as to obtain the relative importance degree between the evaluation indicators according to the subjective weights of the experts, and then a judgment matrix of the evaluation indicators is established according to the relative importance degree. For example, an expert may think that quality is more important than technical ability, and this relationship is reflected in the judgment matrix. Experts can use a numerical scale to quantify the importance degree between evaluation indicators. For example, the 1-9 scale method: where 1 means that the two evaluation indicators are equally important, 3 means that one evaluation indicator is slightly more important than the other, 5 means that one evaluation indicator is significantly more important than the other, 7 means that one evaluation indicator is strongly more important than the other, and 9 means that one evaluation indicator is extremely more important than the other. 2, 4, 6, and 8 represent the intermediate values of the above adjacent judgments.

[0037] Next, the sum-product method is used to calculate the maximum eigenvalue of the judgment matrix. Specifically, each column of the judgment matrix is normalized, and then the normalized judgment matrix is summed row by row to obtain the first vector of the judgment matrix. Then, the first vector is normalized to obtain the eigenvector of the judgment matrix, and thus the maximum eigenvalue of the judgment matrix is approximately calculated using the eigenvector.

[0038] Then, according to the formula CI = (λ max - n) / (n - 1), the consistency index CI of the judgment matrix is calculated, where λ max is the maximum eigenvalue of the judgment matrix, and n is the order of the judgment matrix. Thus, according to the formula CR = CI / RI, the consistency ratio CR of the judgment matrix is calculated, where RI is the random consistency index, and its value depends on the order of the judgment matrix and can be obtained by looking up the table.

[0039] Further, if the consistency ratio CR is less than 0.1, it is considered that the judgment matrix is reasonable and the judgments of the experts are consistent. Otherwise, the judgment matrix needs to be adjusted until the consistency requirement is met. When the consistency ratio CR is less than 0.1, the root method is used to calculate the target vector of the judgment matrix, and thus the first weight of the evaluation index is obtained according to the target vector. Specifically, the elements of each row of the judgment matrix are multiplied to obtain the product vector of the judgment matrix. Then, the nth root of the product vector is calculated to obtain the normalized vector. Next, the normalized vector is normalized to obtain the target vector, and thus the first weight of each evaluation index is determined according to each component in the target vector. For example, the first weight of an evaluation index is 0.4, and the first weight of another evaluation index is 0.3, thus forming a complete weight distribution. It should be noted that the target vector obtained by using the root method also needs to be subjected to subsequent processing such as consistency test to ensure the reliability of the first weight.

[0040] Further, as a refinement and extension of the specific implementation manner of the above embodiment, in order to completely illustrate the specific implementation process of this embodiment, the steps of determining the relative importance degree between evaluation indexes according to the subjective weight include: determining the index order of the evaluation indexes according to the subjective weight, and the ratio of the importance degree between adjacent evaluation indexes in the index order; calculating the initial weight of the evaluation indexes according to the index order and the ratio of the importance degree; and determining the relative importance degree between the evaluation indexes according to the initial weight of the evaluation indexes.

[0041] In this embodiment, the experts compare all the evaluation indexes to obtain the ranking of the importance degree of the evaluation indexes according to the subjective weight of the experts. And the experts assign values to the ratio of the importance degree between adjacent evaluation indexes in the index order, so as to solve and obtain the initial weights of all the evaluation indexes according to the index order and the ratio of the importance degree. The logic is clear, the calculation is simple, and the relative importance degree obtained according to the initial weight has sufficient effectiveness, further improving the accuracy of the bid evaluation result.

[0042] Exemplarily, using the G1 method, the experts assign values to the ratio of the importance degree between adjacent evaluation indexes in the index order. For example, if the adjacent evaluation indexes have the same importance, the ratio of the importance degree is 1.0. If the previous evaluation index is slightly more important than the latter evaluation index, the ratio of the importance degree is 1.2. If the previous evaluation index is significantly more important than the latter evaluation index, the ratio of the importance degree is 1.4. If the previous evaluation index is strongly more important than the latter evaluation index, the ratio of the importance degree is 1.6. If the previous evaluation index is extremely more important than the latter evaluation index, the ratio of the importance degree is 1.8.

[0043] Next, calculate the initial weight of the mth evaluation index in the index ranking according to the index order and the ratio of the importance degree where, b jis the ratio of importance levels. Then, the initial weight a of the (m - 1)-th evaluation index in the index ranking is calculated successively m-1 = b m a m , the initial weight a of the (m - 2)-th evaluation index in the index ranking m-2 = b m-1 a m-1 ,..., a2 = b3a3, a1 = b2a2. Thus, the initial weights of each evaluation index are obtained. Furthermore, based on the ratio of the initial weights, the relative importance levels among the evaluation indexes are determined.

[0044] Furthermore, as a refinement and extension of the specific implementation manner of the above embodiment, to fully illustrate the specific implementation process of this embodiment, the steps for determining the second weight of the evaluation index according to the bid data include: based on the bid data, according to the index type of the evaluation index, perform positive normalization and standardization processing on the data corresponding to the evaluation index to obtain the target data corresponding to the evaluation index; calculate the information entropy of the evaluation index according to the target data corresponding to the evaluation index; determine the second weight of the evaluation index according to the information entropy of the evaluation index.

[0045] In this embodiment, the entropy weight method is used to measure the importance of each evaluation index according to the magnitude of the information entropy, which well avoids the interference of subjective factors, makes the determination of the second weight more scientific and objective, and can objectively reflect the importance of each evaluation index in bid evaluation.

[0046] Exemplarily, different types of indexes have different properties. For example, for cost-type indexes, the smaller the value, the better, while for benefit-type indexes, the larger the value, the better. Therefore, positive normalization and standardization processing are performed on the data corresponding to each evaluation index in the bid data to eliminate the influence of different dimensions and directions, convert the index types of the evaluation indexes into the same type, and obtain the target data corresponding to the evaluation index. Then, the information entropy of each evaluation index is calculated according to the target data. Among them, the smaller the information entropy, the greater the amount of information of the evaluation index and the greater the weight.

[0047] Then, according to the information entropy of the evaluation index, the second weight β of the evaluation index is calculated i . Specifically, as shown in the formula:

[0048]

[0049] where d i is the information entropy redundancy, d i = 1 - e i , and e i is the information entropy of the evaluation index.

[0050] For example, the second weight of a certain evaluation index may be 0.25, and the second weight of another evaluation index may be 0.15.

[0051] In one embodiment, the bid evaluation method further includes: obtaining the correlation coefficients between evaluation indexes; establishing a correlation coefficient matrix of the evaluation indexes according to the correlation coefficients, and determining the eigenvalues of the correlation coefficient matrix and the eigenvectors corresponding to the eigenvalues; determining the evaluation factors with eigenvalues greater than a third preset threshold; determining the loadings of the evaluation indexes on the evaluation factors according to the eigenvectors of the evaluation factors, and establishing a loading matrix of the evaluation factors; performing a rotation process on the loading matrix; determining the target evaluation factors corresponding to the evaluation indexes according to the evaluation factors corresponding to the loadings greater than a fourth preset threshold in the rotated loading matrix; and updating the evaluation indexes corresponding to the target evaluation factors according to the target evaluation factors.

[0052] In this embodiment, using the factor analysis method and based on the idea of dimensionality reduction, multiple evaluation indexes are aggregated into a few independent target evaluation factors, which can reflect the main information of the original numerous evaluation indexes. During the bid evaluation process, it can be used to reduce the number of evaluation indexes while retaining key information. For example, through factor analysis, multiple evaluation indexes related to technical scores are aggregated into several main technical ability factors (target evaluation factors) for further comprehensive evaluation or decision support.

[0053] Specifically, first, the correlation analysis is used to identify the correlation between different evaluation indexes, such as the relationship between price and quality scores. After the correlation analysis, the correlation coefficients between different evaluation indexes can be obtained to understand which evaluation indexes are positively correlated and which are negatively correlated.

[0054] Exemplarily, calculate the mean and standard deviation of the data corresponding to the evaluation indexes in the bid data. Standardize the data corresponding to the evaluation indexes according to the mean and standard deviation. Determine the correlation coefficients between the evaluation indexes according to the mean and standard deviation after the standardization process.

[0055] Next, establish the correlation coefficient matrix between evaluation indicators, perform eigenvalue decomposition on the correlation coefficient matrix to determine the eigenvalues of the correlation coefficient matrix and the corresponding eigenvectors. Among them, a series of eigenvalues are obtained by performing eigenvalue decomposition on the correlation coefficient matrix, and each eigenvalue corresponds to a factor. A larger eigenvalue means that the factor explains more variance of the original data, that is, the factor corresponding to the larger eigenvalue plays a more important role in the information of the evaluation indicators. Therefore, important evaluation factors are selected according to the size of the eigenvalues, that is, the factors with eigenvalues greater than 1 are determined as evaluation factors. The eigenvectors determine the loadings of each evaluation indicator on the evaluation factors, thereby establishing the loading matrix of the evaluation factors. For each evaluation factor, each component in its corresponding eigenvector represents the degree of association between the corresponding evaluation indicator and the evaluation factor.

[0056] Then, perform rotation processing on the loading matrix of the evaluation factors, such as orthogonal rotation processing or oblique rotation processing, so that each evaluation indicator has a large loading on a few evaluation factors, which is convenient for explaining the main aspects represented by the evaluation indicators. Further, according to the rotated loading matrix, observe the loading sizes of each evaluation indicator on each evaluation factor. The larger the loading value, the stronger the correlation between the evaluation indicator and the corresponding evaluation factor. Therefore, multiple original evaluation indicators are aggregated into a few independent and common target evaluation factors according to the size of the loading values. Among them, a loading value greater than 0.5 is considered to have a strong correlation. Specifically, by observing which evaluation indicators have a high loading on a certain target evaluation factor, the main aspects represented by the target evaluation factor are determined and given a meaningful name. For example, if evaluation indicators such as technological advancement and solution integrity have a high loading on a target evaluation factor, it can be interpreted as the "technical ability factor"; if evaluation indicators such as enterprise qualifications and performance have a high loading on a target evaluation factor, it can be interpreted as the "enterprise strength factor"; if evaluation indicators such as after-sales service response time and training plan have a high loading on a target evaluation factor, it can be interpreted as the "service quality factor".

[0057] Step 104: Input the bid data, evaluation indicators, and the target weights of the evaluation indicators into the first preset bid evaluation model to determine the candidate bid proposals in the bid data.

[0058] Step 105: If the number of candidate bid proposals is greater than or equal to the first preset threshold, input the candidate bid proposals, evaluation indicators, and the target weights of the evaluation indicators into the second preset bid evaluation model to determine the target bid proposal;

[0059] Step 106: If the number of candidate bid proposals is less than the first preset threshold, determine the candidate bid proposal as the target bid proposal.

[0060] In this embodiment, a preset bid evaluation model is used to deeply analyze the bid data and conduct multiple evaluations, reducing the deviation caused by human factors, improving the objectivity and accuracy of the bid evaluation results, helping decision-makers determine the optimal bid proposal, improving the bid evaluation efficiency, eliminating the possibility of human intervention in the bid evaluation process to the greatest extent, reducing the risk of bid evaluation experts abusing the scoring discretion and subjective evaluation, enhancing the objectivity of the scoring results, and effectively improving the quality and efficiency of the tendering and procurement bid evaluation work.

[0061] Further, as a refinement and extension of the specific implementation manner of the above embodiment, to fully illustrate the specific implementation process of this embodiment, step 104, that is, the step of inputting the bid data, evaluation indicators, and the target weights of the evaluation indicators into the first preset bid evaluation model to determine the candidate bid proposals in the bid data, specifically includes: determining the evaluation indicators that meet the first preset condition as the input indicators of the first preset bid evaluation model; determining the evaluation indicators that meet the second preset condition as the output indicators of the first preset bid evaluation model; determining the first target quantity according to the input indicators and the target weights of the input indicators; determining the second target quantity according to the output indicators and the target weights of the output indicators; constructing a linear programming problem of the first preset bid evaluation model according to the first target quantity and the second target quantity; solving the linear programming problem according to the bid data to obtain the first evaluation result of the bid proposals in the bid data; and determining the bid proposals with the first evaluation result equal to the fifth preset threshold as the candidate bid proposals.

[0062] In this embodiment, the data envelopment analysis (DEA) method is used to evaluate the production efficiency of each bid proposal in the bid data, so as to conduct a preliminary evaluation of the bid proposals, screen out effective candidate bid proposals, and improve the bid evaluation efficiency and reliability.

[0063] Exemplarily, the input indicators (such as resource consumption indicators like cost and time) and output indicators (such as result output indicators like quality and benefit) are specified in the evaluation indicators. For each bid proposal, the following CCR (Charnes-Cooper-Rhodes) model is constructed:

[0064] Assume that there are c bid proposals in the bid data, and each bid proposal has h input indicators and s output indicators. For the y-th bid proposal in the bid data, according to the data corresponding to the input indicators and the data corresponding to the output indicators, the input vector G corresponding to each input indicator is obtained as y =(g 1y , g 2y ,..., g hy ), and the output vector Q corresponding to each output indicator is y =(q 1y , q 2y ,..., qsy )。

[0065] The goal of the CCR model is to solve the following linear programming problem:

[0066]

[0067] where u r is the target weight of the evaluation index corresponding to the output vector q ry and v f is the target weight of the evaluation index corresponding to the input vector g fy and ε is a very small positive number to avoid the target weight being 0.

[0068] Furthermore, use linear programming solution software, such as MaxDEA software, to solve the above linear programming problem, obtain the efficiency value of each bidding plan in the bidding data, and use the efficiency value as the first evaluation result of the bidding plan. If the first evaluation result of the bidding plan is equal to 1, it means that the bidding plan is effective, and this bidding plan is determined as the candidate bidding plan. If the first evaluation result is less than 1, it means that the bidding plan is relatively inefficient.

[0069] Furthermore, as a refinement and extension of the specific implementation manner of the above embodiment, in order to fully illustrate the specific implementation process of this embodiment, the steps of inputting the candidate bidding plan, evaluation index, and target weight of the evaluation index into the second preset bid evaluation model to determine the target bidding plan specifically include: constructing a decision matrix according to the candidate bidding plan and evaluation index; performing positive normalization and standardization processing on the decision matrix according to the index type of the evaluation index to obtain the target decision matrix; calculating the first distance between the candidate bidding plan and the first preset plan, and the second distance between the candidate bidding plan and the second preset plan according to the target decision matrix and the target weight of the evaluation index; determining the second evaluation result of the candidate bidding plan according to the first distance and the second distance; and screening the target bidding plan from the candidate bidding plans according to the second evaluation result.

[0070] In this embodiment, if the number of candidate bidding plans after preliminary evaluation is 2 or more than 2, then use the second preset bid evaluation model to perform a secondary evaluation on the candidate bidding plans to determine the optimal bidding plan, better select the bidding plan that best meets the requirements and has the best comprehensive advantages, thereby improving the success rate and efficiency of the bid evaluation process.

[0071] Exemplarily, the topsis model is used as the second preset bid evaluation model. First, a decision matrix is constructed. Each column of the decision matrix is an evaluation index, and each row is a bidding scheme. For different types of evaluation indexes, some evaluation indexes correspond to data that is better when it is larger, some evaluation indexes correspond to data that is better when it is closer to a certain value, and some are the best within a certain range. Therefore, the data in the decision matrix is normalized to use the same standard, simplify the analysis process, and avoid confusion caused by different directions and ranges. For example, after normalization, all data is better when it is larger. Then, the normalized decision matrix is standardized to eliminate the influence of different index dimensions and obtain the target decision matrix.

[0072] Then, experts give the optimal scheme (the first preset scheme) and the worst scheme (the second preset scheme) for the bidding data. Thus, the first distance D between the P-th candidate bidding scheme and the first preset scheme is calculated. P + :

[0073]

[0074] where Z i + is the value of the i-th evaluation index in the first preset scheme, and z iP is the value of the i-th evaluation index in the p-th candidate bidding scheme, p = 1, 2,..., L, and L is the number of candidate bidding schemes.

[0075] And the second distance D between the P-th candidate bidding scheme and the second preset scheme is calculated. P - :

[0076]

[0077] where Z i - is the value of the i-th evaluation index in the second preset scheme.

[0078] In this embodiment, the first distance and the second distance are used to evaluate the distance between the candidate bidding scheme and the optimal scheme or the worst scheme. The larger the value, the farther the distance. The relatively smaller the first distance and the relatively larger the second distance indicate that the candidate bidding scheme is better.

[0079] Furthermore, the closeness between the candidate bidding scheme and the optimal scheme is calculated based on the first distance and the second distance to determine the second evaluation result η p .

[0080]

[0081] Therefore, the candidate bidding schemes are sorted according to the second evaluation result to determine the scheme order, and the candidate bidding scheme ranked first in the scheme order is determined as the target bidding scheme.

[0082] In one embodiment, the bid evaluation method further includes: initializing the parameter combination of the backpropagation neural network in the preset genetic algorithm optimized neural network model; according to the parameter combination, using the backpropagation neural network to perform forward propagation calculation on the input data in the training sample set according to the historical weights of the historical evaluation indicators, and obtaining the predicted evaluation result of the input data, where the input data in the training sample set is determined according to the data corresponding to the historical evaluation indicators in the historical bidding scheme; according to the predicted evaluation result and the output data in the training sample set, determining the prediction error, where the output data in the training sample set is determined according to the historical evaluation results of the historical bidding scheme; according to the prediction error, determining the fitness of the parameter combination; screening the target parameter combination from the parameter combinations according to the fitness; performing genetic crossover and genetic mutation operations on the target parameter combination to update the parameter combination; according to the updated parameter combination, using the backpropagation neural network to perform forward propagation calculation on the input data in the training sample set according to the historical weights of the historical evaluation indicators until a preset stop condition is reached, and obtaining the second preset bid evaluation model.

[0083] In this embodiment, the second preset bid evaluation model is determined according to the GA-BP (Genetic Algorithm-Back Propagation) model. Specifically, experts evaluate the historical bid evaluation schemes according to the historical evaluation indicators to obtain the historical evaluation results, and thus construct a training set for the preset genetic algorithm optimized neural network model according to the historical bidding schemes, historical evaluation indicators, historical weights corresponding to the historical evaluation indicators, and historical evaluation results, and use the training set to train the preset genetic algorithm optimized neural network model, comprehensively considering various external influencing factors, discovering the bid evaluation rules and patterns through machine learning algorithms, and obtaining the second preset bid evaluation model, so as to improve the accuracy and efficiency of bid evaluation.

[0084] Exemplarily, a certain number of individuals are randomly generated, and each individual represents a parameter combination of a backpropagation neural network in a preset genetic algorithm-optimized neural network model. For each individual in the population, the input data (historical evaluation index values of historical bidding schemes) in the training sample set and the historical weights of the historical evaluation indexes are input into the backpropagation neural network for forward propagation calculation to obtain a predicted evaluation result. Then, according to the predicted difference between the predicted evaluation result and the historical evaluation result, that is, according to the prediction error of the backpropagation neural network on the training set, the fitness of each individual is calculated. According to the fitness size, excellent individuals are selected as parents. Cross and mutation operations are performed on the parents to generate new offspring. The original population is replaced with the new offspring. The backpropagation neural network is trained using the updated individuals. The above steps are repeated until the stop condition is reached. A second preset bid evaluation model is obtained, so that the second preset bid evaluation model can effectively process complex regression prediction problems under a large-scale data set, has strong generalization ability, avoids the backpropagation neural network falling into a local optimal solution, and improves the robustness and stability of the second preset bid evaluation model.

[0085] It is worth mentioning that in this embodiment, during the forward propagation calculation process of the entire backpropagation neural network, the historical weights of the historical evaluation indexes are added. For example, when calculating the sum of the weight products from the input layer to the hidden layer and from the hidden layer to the output layer, the value of each historical evaluation index is multiplied by its corresponding historical weight. This can make the network pay more attention to those evaluation indexes that are considered important in history during the calculation process.

[0086] During the actual bid evaluation process, after inputting the candidate bidding scheme, the evaluation index, and the target weight of the evaluation index into the second preset bid evaluation model obtained according to the preset genetic algorithm-optimized neural network model, the second preset bid evaluation model outputs the third evaluation result of the candidate bidding scheme, and the optimal target bidding scheme in the candidate bidding scheme is selected according to the third evaluation result.

[0087] In one embodiment, the bid evaluation method further includes: determining the data characteristics of the bid data through descriptive analysis of the bid data, such as the distribution of bidders, the central tendency of quotations, etc. After descriptive analysis, the distribution characteristics and central tendency of the data can be obtained, such as the average quotation of bidders, the median of technical scores, etc. Cluster analysis can also be performed on the bid data to group the bidding schemes according to certain characteristics (such as quotations, technical scores, etc.) to better understand the similarities and differences between the bidding schemes. After cluster analysis, the classification results of different bidding schemes can be obtained, and each category may represent a group of bidding schemes with similar characteristics.

[0088] In one embodiment, the bid evaluation method further includes: using security technologies to comply with relevant requirements and designing security control in multiple aspects. During the bid evaluation process, the data that needs to be protected includes sensitive information such as bid documents, the record of bid reading by bid evaluation experts, and bid evaluation reports. The bid proposals are decrypted during the bid opening. After the bid evaluation, data such as the record of bid reading by bid evaluation experts and bid evaluation reports will be subject to bid cleaning. In daily bid evaluation work, the information data of all bidding manufacturers must be decrypted and synchronously pushed after the bid opening, and the information data of bidding manufacturers cannot be obtained before that. Thus, the integrity, confidentiality, and availability of the data related to the bid evaluation process are protected, ensuring the fairness and transparency of the bid evaluation process.

[0089] In one embodiment, the bid evaluation method further includes: using deep learning to identify unstructured data in bid data, performing word segmentation, grammar, and semantic analysis on the text content through natural language processing, and extracting key scoring elements required for the preset bid evaluation model. Processing structured data so that the preset bid evaluation model can perform scoring, such as price, etc.

[0090] In one embodiment, the bid evaluation method further includes: through cloud storage technology, saving the data generated by online bidding and tendering work into the database system for future query and invocation. Moreover, a layered architecture design is adopted, with different functions and responsibilities divided into different layers, reducing the coupling degree between layers. For example, it is divided into a presentation layer, a business logic layer, a data access layer, a data storage layer, a security service layer, and an integration layer. When function expansion or modification is required, the layered architecture enables the changes to be confined within a specific layer and will not have too much impact on other layers. For example, to add a new business function, only the business logic layer needs to be developed and deployed, without affecting other parts such as the presentation layer and the data storage layer, improving the maintainability and scalability of the system. At the same time, the business logic is decomposed into independent services, and each service focuses on a business function, such as a tendering microservice, a bidding microservice, a bid evaluation microservice, etc. These microservices can be developed, tested, and deployed independently, improving the development efficiency. For example, the development team can develop different microservices in parallel, accelerating the overall progress of the project. In addition, the packaging and deployment of the application are realized, and the application and its dependent environment are packaged into a container, enabling the application to be quickly deployed in different environments. For example, in different server environments or cloud environments, only the container needs to be deployed without reconfiguring the environment, greatly improving the deployment efficiency. The containerized application can run in any environment that supports the container runtime, with strong portability. This enables the application to run conveniently on different platforms and devices, improving the applicability of the application.

[0091] In one embodiment, the bid evaluation method further includes: using message queue technology to transfer data from the target e-commerce platform. Specifically, in the case of high concurrency where a large number of bidders submit bid documents simultaneously, using a message queue can buffer these requests. In this way, bidders can quickly receive a submission success response without having to wait for the system to perform complex verification and storage operations on the bid documents. Then, the system processes the bid tasks one by one from the message queue according to its own processing capacity, thus effectively coping with the high-concurrency scenario to prevent the system from being overloaded or even crashing.

[0092] The bid evaluation method of this application reduces costs by reducing labor and time costs, as well as optimizing resource allocation, reducing the costs and inputs of the evaluation work, and improving the overall efficiency. By optimizing the user interface and interaction design, it improves the user operation experience and satisfaction, enhancing users' recognition and trust in the evaluation work. By adopting advanced security technologies and measures, it protects the security and reliability of the evaluation data, ensuring the stable operation of the system and data security. By introducing advanced data automation technologies and innovative thinking, it promotes the innovation and development of the evaluation business, meeting the ever-changing market demands and business challenges. Through system integration and collaborative working mechanisms, it promotes collaboration and communication among team members, improving work efficiency and cooperation effectiveness. By reasonably allocating resources and optimizing the work process, it improves the resource utilization efficiency and management level of the evaluation work. Through automation technologies and backup mechanisms, it ensures the continuity and stability of the evaluation business application, guaranteeing the reliability of business operation.

[0093] Further, as Figure 2 shown, as a specific implementation of the above bid evaluation method, an embodiment of this application provides a bid evaluation device 200, which includes: an acquisition module 201, a determination module 202, and a bid evaluation module 203.

[0094] Among them, the acquisition module 201 is used to obtain tender data, bid evaluation data, and bid data to be evaluated from the target e-commerce platform through a preset data interface;

[0095] The determination module 202 is used to determine the evaluation indicators of the bid data according to the tender data, the data, and the bid data; and determine the target weights of the evaluation indicators;

[0096] The bid evaluation module 203 is used to input the bid data, evaluation indicators, and target weights of the evaluation indicators into a first preset bid evaluation model to determine the candidate bid proposals in the bid data; if the number of candidate bid proposals is greater than or equal to a first preset threshold, then input the candidate bid proposals, evaluation indicators, and target weights of the evaluation indicators into a second preset bid evaluation model to determine the target bid proposal; if the number of candidate bid proposals is less than the first preset threshold, then determine the candidate bid proposal as the target bid proposal.

[0097] In one embodiment, the determining module 202 is specifically configured to receive the subjective weight input by the user, determine the first weight of the evaluation index according to the subjective weight; determine the second weight of the evaluation index according to the bid data; construct an objective function of the target weight with the goal of minimizing the deviation between the first weight and the second weight; and solve the objective function according to the first weight and the second weight to obtain the value of the target weight of the evaluation index.

[0098] In one embodiment, the determining module 202 is specifically configured to determine the relative importance degree between evaluation indexes according to the subjective weight; determine the judgment matrix of the evaluation indexes according to the relative importance degree between the evaluation indexes; calculate the maximum eigenvalue of the judgment matrix; determine the consistency ratio of the judgment matrix according to the maximum eigenvalue; if the consistency ratio is greater than or equal to the second preset threshold, adjust the relative importance degree until the consistency ratio is less than the second preset threshold; if the consistency ratio is less than the second preset threshold, perform a product operation on the judgment matrix by row to determine the product vector of the judgment matrix; perform a square root processing and a normalization processing on the product vector to determine the target vector; and determine the first weight of the evaluation index according to the target vector.

[0099] In one embodiment, the determining module 202 is specifically configured to determine the index order of the evaluation indexes according to the subjective weight and the ratio of the importance degrees of adjacent evaluation indexes in the index order; calculate the initial weight of the evaluation index according to the index order and the ratio of the importance degrees; and determine the relative importance degree between the evaluation indexes according to the initial weight of the evaluation index.

[0100] In one embodiment, the determining module 202 is specifically configured to, based on the bid data, perform a positive processing and a standardization processing on the data corresponding to the evaluation index according to the index type of the evaluation index to obtain the target data corresponding to the evaluation index; calculate the information entropy of the evaluation index according to the target data corresponding to the evaluation index; and determine the second weight of the evaluation index according to the information entropy of the evaluation index.

[0101] In one embodiment, the bid evaluation device 200 further includes:

[0102] An updating module, configured to obtain the correlation coefficient between evaluation indexes; determine the eigenvalue of the correlation coefficient matrix and the eigenvector corresponding to the eigenvalue according to the correlation coefficient; determine the evaluation factor by using the eigenvalue greater than the third preset threshold; determine the load of the evaluation index on the evaluation factor according to the eigenvector of the evaluation factor, and establish a load matrix of the evaluation factor; perform a rotation processing on the load matrix; determine the target evaluation factor corresponding to the evaluation index according to the evaluation factor corresponding to the load greater than the fourth preset threshold in the rotated load matrix; and update the evaluation index corresponding to the target evaluation factor according to the target evaluation factor.

[0103] In one embodiment, the bid evaluation module 203 is specifically configured to determine evaluation indicators that meet the first preset condition as the input indicators of the first preset bid evaluation model; determine evaluation indicators that meet the second preset condition as the output indicators of the first preset bid evaluation model; determine a first target quantity according to the input indicators and the target weights of the input indicators; determine a second target quantity according to the output indicators and the target weights of the output indicators; construct a linear programming problem of the first preset bid evaluation model according to the first target quantity and the second target quantity; solve the linear programming problem according to the bid data to obtain a first evaluation result of the bid proposal in the bid data; and determine a bid proposal with the first evaluation result equal to the fifth preset threshold as a candidate bid proposal.

[0104] In one embodiment, the bid evaluation module 203 is specifically configured to construct a decision matrix according to the candidate bid proposal and the evaluation indicators; perform positive normalization processing and standardization processing on the decision matrix according to the indicator types of the evaluation indicators to obtain a target decision matrix; calculate a first distance between the candidate bid proposal and the first preset proposal, and a second distance between the candidate bid proposal and the second preset proposal respectively according to the target decision matrix and the target weights of the evaluation indicators; determine a second evaluation result of the candidate bid proposal according to the first distance and the second distance; and screen a target bid proposal from the candidate bid proposals according to the second evaluation result.

[0105] In one embodiment, the bid evaluation device 200 further includes:

[0106] A training module, configured to initialize a parameter combination of a backpropagation neural network in a preset genetic algorithm optimized neural network model; according to the parameter combination, perform forward propagation calculation on the input data in the training sample set according to the historical weights of the historical evaluation indicators to obtain a predicted evaluation result of the input data, where the input data in the training sample set is determined according to the data corresponding to the historical evaluation indicators in the historical bid proposal; determine a prediction error according to the predicted evaluation result and the output data in the training sample set, where the output data in the training sample set is determined according to the historical evaluation results of the historical bid proposal; determine the fitness of the parameter combination according to the prediction error; screen a target parameter combination from the parameter combinations according to the fitness; perform genetic crossover and genetic mutation operations on the target parameter combination to update the parameter combination; and according to the updated parameter combination, perform forward propagation calculation on the input data in the training sample set according to the historical weights of the historical evaluation indicators until a preset stop condition is reached to obtain a second preset bid evaluation model.

[0107] For the specific limitations of the bid evaluation device, reference may be made to the limitations of the bid evaluation method in the foregoing text, which will not be elaborated here. Each module in the above bid evaluation device can be implemented in whole or in part by software, hardware, and their combinations. Each of the above modules can be embedded in the processor of the computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each of the above modules.

[0108] Based on the above such as Figure 1 shown method, correspondingly, an embodiment of the present application also provides a readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the above such as Figure 1 shown bid evaluation method is implemented.

[0109] Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product, and the software product can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various implementation scenarios of the present application.

[0110] Based on the above such as Figure 1 shown method, and Figure 2 shown virtual device embodiment, in order to achieve the above object, an embodiment of the present application also provides a computer device, specifically a personal computer, a server, a network device, etc., and the computer device includes a storage medium and a processor; the storage medium is used for storing a computer program; the processor is used for executing the computer program to implement the above such as Figure 1 shown bid evaluation method.

[0111] Optionally, the computer device may further include a user interface, a network interface, a camera, a radio frequency (RF) circuit, sensors, an audio circuit, a WI-FI module, and so on. The user interface may include a display screen (Display), an input unit such as a keyboard (Keyboard), etc., and optionally the user interface may further include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (such as a Bluetooth interface, a WI-FI interface), etc.

[0112] Those skilled in the art can understand that the structure of a computer device provided in this embodiment does not constitute a limitation on the computer device, and it may include more or fewer components, or combine some components, or have different component arrangements.

[0113] The storage medium may further include an operating system and a network communication module. The operating system is a program for managing and storing the hardware and software resources of a computer device, and supports the operation of information processing programs and other software and / or programs. The network communication module is used to implement communication between components inside the storage medium, as well as communication with other hardware and software in the entity device.

[0114] Through the description of the above embodiments, those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform, or the embodiments of the present application can be implemented by hardware.

[0115] Those skilled in the art can understand that the drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the drawings are not necessarily essential for implementing the present application. Those skilled in the art can understand that the modules in the device in the implementation scenario can be distributed in the device in the implementation scenario according to the description of the implementation scenario, or can be correspondingly changed and located in one or more devices different from the present implementation scenario. The modules in the above implementation scenario can be combined into one module, or further split into multiple sub-modules.

[0116] The above serial numbers of the present application are only for description and do not represent the advantages or disadvantages of the implementation scenarios. The above disclosure only shows several specific implementation scenarios of the present application. However, the present application is not limited thereto, and any changes that can be conceived by those skilled in the art should fall within the protection scope of the present application.

Claims

1. A bid evaluation method, characterized in that, The method includes: Obtaining tender data, bid evaluation data, and tender bid data to be evaluated from a target e-commerce platform through a preset data interface; Determining evaluation indicators for the tender bid data according to the tender data, the bid evaluation data, and the tender bid data; Determining the target weights of the evaluation indicators; Inputting the tender bid data, the evaluation indicators, and the target weights of the evaluation indicators into a first preset bid evaluation model to determine candidate tender bid proposals in the tender bid data; If the number of candidate tender bid proposals is greater than or equal to a first preset threshold, inputting the candidate tender bid proposals, the evaluation indicators, and the target weights of the evaluation indicators into a second preset bid evaluation model to determine the target tender bid proposal; If the number of candidate tender bid proposals is less than the first preset threshold, determining the candidate tender bid proposals as the target tender bid proposal.

2. The bid evaluation method according to claim 1, wherein The determining the target weights of the evaluation indicators includes: Receiving subjective weights input by a user, and determining a first weight of the evaluation indicators according to the subjective weights; Determining a second weight of the evaluation indicators according to the tender bid data; Taking the minimum deviation between the first weight and the second weight as the target, constructing an objective function for the target weights; Solving the objective function according to the first weight and the second weight to obtain the values of the target weights of the evaluation indicators.

3. The bid evaluation method according to claim 2, characterized in that, The determining the first weight of the evaluation indicators according to the subjective weights includes: Determining the relative importance degree between the evaluation indicators according to the subjective weights; Determining a judgment matrix of the evaluation indicators according to the relative importance degree between the evaluation indicators; Calculating the maximum eigenvalue of the judgment matrix; Determining the consistency ratio of the judgment matrix according to the maximum eigenvalue; If the consistency ratio is greater than or equal to a second preset threshold, adjusting the relative importance degree until the consistency ratio is less than the second preset threshold; If the consistency ratio is less than the second preset threshold, performing a product operation on the rows of the judgment matrix to determine the product vector of the judgment matrix; Performing a square root processing and a normalization processing on the product vector to determine a target vector; Determining the first weight of the evaluation indicators according to the target vector.

4. The bid evaluation method according to claim 3, characterized in that The step of determining the relative importance degree between the evaluation indicators according to the subjective weights includes: Determining the index order of the evaluation indicators according to the subjective weights, and the ratio of the importance degree between adjacent evaluation indicators in the index order; Calculating the initial weights of the evaluation indicators according to the index order and the ratio of the importance degree; Determining the relative importance degree between the evaluation indicators according to the initial weights of the evaluation indicators.

5. The bid evaluation method according to claim 2, wherein The determining the second weight of the evaluation indicators according to the tender bid data includes: Based on the tender bid data, performing a positive processing and a standardization processing on the data corresponding to the evaluation indicators according to the index types of the evaluation indicators to obtain the target data corresponding to the evaluation indicators; Calculating the information entropy of the evaluation indicators according to the target data corresponding to the evaluation indicators; Determining the second weight of the evaluation indicators according to the information entropy of the evaluation indicators.

6. The bid evaluation method according to claim 1, wherein The method further includes: Obtaining the correlation coefficients between the evaluation indicators; Based on the correlation coefficients, establishing a correlation coefficient matrix of the evaluation indicators, and determining the eigenvalues of the correlation coefficient matrix and the eigenvectors corresponding to the eigenvalues; Determining the eigenvalues greater than a third preset threshold as evaluation factors; Based on the eigenvectors of the evaluation factors, determining the loadings of the evaluation indicators on the evaluation factors, and establishing a loading matrix of the evaluation factors; Performing a rotation process on the loading matrix; Based on the evaluation factors corresponding to the loadings greater than a fourth preset threshold in the rotated loading matrix, determining the target evaluation factors corresponding to the evaluation indicators; Updating the evaluation indicators corresponding to the target evaluation factors according to the target evaluation factors.

7. The bid evaluation method according to claim 1, characterized in that, The step of inputting the bid data, the evaluation indicators, and the target weights of the evaluation indicators into a first preset bid evaluation model to determine the candidate bid proposals in the bid data includes: Determining the evaluation indicators meeting a first preset condition as the input indicators of the first preset bid evaluation model; Determining the evaluation indicators meeting a second preset condition as the output indicators of the first preset bid evaluation model; Determining a first target quantity according to the input indicators and the target weights of the input indicators; Determining a second target quantity according to the output indicators and the target weights of the output indicators; Constructing a linear programming problem of the first preset bid evaluation model according to the first target quantity and the second target quantity; Solving the linear programming problem according to the bid data to obtain a first evaluation result of the bid proposals in the bid data; Determining the bid proposals with the first evaluation result equal to a fifth preset threshold as the candidate bid proposals.

8. The bid evaluation method according to claim 1, characterized in that The step of inputting the candidate bid proposals, the evaluation indicators, and the target weights of the evaluation indicators into the second preset bid evaluation model to determine the target bid proposal includes: Constructing a decision matrix according to the candidate bid proposals and the evaluation indicators; Performing a positive normalization process and a standardization process on the decision matrix according to the index types of the evaluation indicators to obtain a target decision matrix; Calculating a first distance between the candidate bid proposal and a first preset proposal, and a second distance between the candidate bid proposal and a second preset proposal respectively according to the target decision matrix and the target weights of the evaluation indicators; Determining a second evaluation result of the candidate bid proposal according to the first distance and the second distance; Selecting the target bid proposal from the candidate bid proposals according to the second evaluation result.

9. The bid evaluation method according to claim 1, wherein The method further includes: Initializing the parameter combination of the backpropagation neural network in a preset genetic algorithm optimized neural network model; According to the parameter combination, using the backpropagation neural network to perform forward propagation calculation on the input data in the training sample set according to the historical weights of the historical evaluation indicators, to obtain a predicted evaluation result of the input data, where the input data in the training sample set is determined according to the data corresponding to the historical evaluation indicators in the historical bid proposals; Determine a prediction error according to the predicted evaluation result and the output data in the training sample set, where the output data in the training sample set is determined according to the historical evaluation result of the historical bidding scheme; Determine the fitness of the parameter combination according to the prediction error; Screen a target parameter combination from the parameter combinations according to the fitness; Perform genetic crossover and genetic mutation operations on the target parameter combination to update the parameter combination; According to the updated parameter combination, use the backpropagation neural network to perform forward propagation calculation on the input data in the training sample set according to the historical weights of the historical evaluation indicators until a preset stop condition is reached, and obtain the second preset bid evaluation model.

10. An evaluation device for bid, characterized in that, The device includes: An acquisition module, configured to acquire bidding data, bid evaluation data, and bid data to be evaluated from a target e-commerce platform through a preset data interface; A determination module, configured to determine evaluation indicators of the bid data according to the bidding data, the bid evaluation data, and the bid data; and Determine the target weight of the evaluation indicators; A bid evaluation module, configured to input the bid data, the evaluation indicators, and the target weights of the evaluation indicators into a first preset bid evaluation model to determine candidate bidding schemes in the bid data; If the number of candidate bidding schemes is greater than or equal to a first preset threshold, input the candidate bidding schemes, the evaluation indicators, and the target weights of the evaluation indicators into a second preset bid evaluation model to determine a target bidding scheme; If the number of candidate bidding schemes is less than the first preset threshold, determine the candidate bidding scheme as the target bidding scheme.

Images