Method, device, computer device and storage medium for proposal scoring

Abstract

The application relates to a proposal scoring method, device, computer equipment and storage medium. The method comprises the following steps: establishing a comprehensive database, wherein the comprehensive database comprises a historical file library, a legal file library and a historical proposal library; obtaining a proposal to be scored; obtaining a first scoring dimension according to the similarity between the title of the proposal to be scored and the historical file library; obtaining a second scoring dimension according to the reference relationship between the text of the proposal to be scored and the legal file library; obtaining a third scoring dimension according to the similarity between the text of the proposal to be scored and the historical proposal library; establishing a scoring standard library; the scoring standard library comprises a scoring selection area and a scoring setting area; the scoring standard library is displayed in a visual manner, and a proposal scoring model is adaptively output according to the selected scoring dimension in the scoring selection area, so that the score and the rating of the proposal to be scored are obtained. The method improves the accuracy and efficiency of proposal scoring.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to methods, apparatus, computer devices, and storage media for proposal scoring. Background Technology

[0002] With the development of computer technology, traditional manual tasks are gradually being replaced by computers. During meetings, participants typically submit a large number of proposals. These proposals represent the participants' voices and reflect their opinions on hot topics and institutional systems in a direct way. After the proposals are submitted, the reviewers will select the outstanding ones and assign them to key personnel for dedicated handling.

[0003] In traditional methods, the reviewers rely on their extensive experience to determine whether a proposal is one that needs to be prioritized, thereby selecting the best proposals.

[0004] However, because there are no scoring rules, the selection of excellent proposals is highly subjective and the accuracy rate is not high. Moreover, manual screening is time-consuming, labor-intensive, and inefficient. Summary of the Invention

[0005] Therefore, it is necessary to provide a method, apparatus, computer equipment, and storage medium for proposal scoring to address the aforementioned technical problems, thereby improving the accuracy and efficiency of selecting excellent proposals.

[0006] A method for evaluating proposals, the method comprising:

[0007] A comprehensive database is established, comprising: a historical document database, a legal document database, and a historical proposal database. The historical document database includes historical documents extracted from the organization's server, including work priorities and hot topics. The legal document database includes legal texts. The historical proposal database includes proposals from previous years.

[0008] Obtain the proposal to be scored, and obtain a first analysis result based on the similarity between the title of the historical document library and the title of the proposal to be scored. Then, obtain a first scoring dimension based on the first analysis result.

[0009] Based on the citation relationship between the legal document database and the text of the proposal to be scored, a second analysis result is obtained, and a second scoring dimension is obtained based on the second analysis result;

[0010] Based on the similarity between the text of the historical proposal database and the proposal to be scored, a third analysis result is obtained, and a third scoring dimension is obtained based on the third analysis result;

[0011] A rating standard library is established; the rating standard library includes a rating selection area and a rating setting area; the data in the rating selection area includes a first rating dimension, a second rating dimension, and a third rating dimension; the rating setting area has a preset format dimension filling form; the dimension filling form is used to receive input rating dimension data, and after receiving a confirmation instruction, the content in the dimension filling form is standardized and input into the rating selection area;

[0012] The scoring criteria library is displayed in a visual manner, and a proposal scoring model for proposal scoring is adaptively output based on the selected scoring dimensions in the scoring selection area.

[0013] The score and rating of the proposal to be scored are obtained based on the proposed scoring model.

[0014] In one embodiment, the adaptive output of a proposal scoring model for proposal scoring, based on the selected scoring dimension in the scoring selection region, includes:

[0015] Obtain the importance ratio corresponding to the selected rating dimension through visualization, and determine whether the selected rating dimension belongs to the label dimension. Based on the importance ratio and whether the selected rating dimension belongs to the label dimension, output the proposal scoring model for proposal scoring; the proposal scoring model is used to output the proposal score and rating.

[0016] When the selected rating dimension belongs to the indicator dimension, and the rating of the selected rating dimension reaches the set value, the corresponding rating is directly output.

[0017] When the selected rating dimension is not a marker dimension, or when the selected rating dimension is a marker dimension and the rating of the selected rating dimension is less than the set value, the rating is calculated based on the importance ratio of the selected rating dimension, and the corresponding rating is determined and output based on the rating.

[0018] In one embodiment, the first analysis result obtained based on the similarity between the title of the historical document library and the proposal to be rated includes:

[0019] Based on the cosine similarity algorithm, a similarity analysis is performed on each of the historical files to obtain an optimal library;

[0020] Based on the K-means clustering algorithm, each historical file in the preferred library is clustered to obtain clusters;

[0021] Calculate the number of historical files for each cluster class to obtain the hotspot clusters;

[0022] Keywords are extracted from each historical file of the hotspot cluster, the frequency of keywords in each historical file is calculated, and each keyword is sorted according to the frequency of keywords to obtain the topic words of the hotspot cluster;

[0023] By combining the aforementioned keywords, popular topics can be obtained;

[0024] Based on a pre-set word segmentation library, the title of the proposal to be scored is segmented to obtain a set of title words;

[0025] Based on the TF-IDF algorithm, the weight of each word in the title word set is calculated to obtain the optimal set;

[0026] Based on the cosine similarity algorithm, the preferred set is analyzed for similarity with the popular topics to obtain the first analysis result.

[0027] In one embodiment, the first analysis result obtained based on the similarity between the title of the historical document library and the proposal to be rated includes:

[0028] Extract keywords from the historical documents and calculate the weight of each keyword. Sort each keyword in descending order of weight and select the keywords with the highest weight as hot words.

[0029] An index is created for each of the hot words, and the index value is used based on the number of times the hot word appears in the historical files;

[0030] Obtain the expected value of the index value, and determine popular topics based on the indexes where the index value reaches the expected value;

[0031] Based on a pre-set word segmentation library, the title of the proposal to be scored is segmented to obtain a set of title words;

[0032] Based on the TF-IDF algorithm, the weight of each word in the title word set is calculated to obtain the optimal set;

[0033] Based on the cosine similarity algorithm, the preferred set is analyzed for similarity with popular topics to obtain the first analysis result.

[0034] In one embodiment, based on the citation relationship between the legal document database and the text of the proposal to be scored, the second analysis result includes:

[0035] Based on the legal texts in the legal document database, keyword matching templates and token matching templates are constructed respectively; both keyword matching templates and token matching templates include content to be filled in.

[0036] The keyword matching template and the tag matching template are concatenated to obtain the citation relationship matching template;

[0037] Based on the reference relationship matching template, the proposal to be analyzed is matched to obtain the corresponding text content in the content to be filled;

[0038] Based on the legal document database, the text content is matched to obtain the cited words and citation frequency; based on the historical document database, the proposal to be scored is matched to obtain the number of citations.

[0039] The highest score is awarded based on the cited words and their frequency.

[0040] Based on the number of citations, a heat value analysis is performed on the proposals to be scored and the historical document library to obtain a second score;

[0041] Based on the first and second scores, the second analysis result is obtained.

[0042] In one embodiment, the third analysis result obtained based on the similarity between the text of the historical proposal database and the proposal to be scored includes:

[0043] Based on a pre-set word segmentation database, each proposal in the historical proposal database is segmented into words to obtain a word set for the proposal database.

[0044] Based on the TF-IDF algorithm, the weight of each word in the proposal database word set is calculated to obtain the key set;

[0045] Based on a pre-set word segmentation library, the proposals to be scored are segmented to obtain a set of proposal words;

[0046] Based on the TF-IDF algorithm, the weight of each word in the proposal word set is calculated to obtain the key set;

[0047] Based on the cosine similarity algorithm, the similarity analysis is performed between the key set and the focus set to obtain the third analysis result.

[0048] In one embodiment, the rating selection area further includes: a fourth rating dimension;

[0049] A sufficiency analysis of arguments and evidence is performed on the proposals to be scored to obtain a fourth analysis result, and a fourth scoring dimension is obtained based on the fourth analysis result.

[0050] An analysis of the sufficiency of arguments and evidence in the proposals to be evaluated yielded the following fourth analysis results:

[0051] Based on a pre-set set of feature words, feature word identification is performed on the proposals to be scored to obtain the word count of each paragraph in which the feature words appear; the feature words are logical words that can characterize the arguments and evidence of the proposals to be extracted.

[0052] The fourth analysis result was obtained based on the word count of the relevant paragraphs.

[0053] An apparatus for a proposal scoring model, the apparatus comprising:

[0054] The preparation module is used to establish a comprehensive database, which includes: a historical document database, a legal document database, and a historical proposal database. The historical document database includes historical documents extracted from the organization's server, including work priorities and hot topics. The legal document database includes legal texts. The historical proposal database includes proposals from previous years.

[0055] The first analysis module is used to obtain the proposal to be scored, obtain the first analysis result based on the similarity between the title of the historical document library and the title of the proposal to be scored, and obtain the first scoring dimension based on the first analysis result.

[0056] The second analysis module is used to obtain a second analysis result based on the citation relationship between the legal document library and the text of the proposal to be scored, and to obtain a second scoring dimension based on the second analysis result;

[0057] The third analysis module is used to obtain a third analysis result based on the similarity between the text of the historical proposal database and the proposal to be scored, and to obtain a third scoring dimension based on the third analysis result.

[0058] A rating criteria library creation module is used to create a rating criteria library; the rating criteria library includes: a rating selection area and a rating setting area; the data in the rating selection area includes: a first rating dimension, a second rating dimension, and a third rating dimension; the rating setting area sets a dimension filling form with a preset format; the dimension filling form is used to receive input rating dimension data, and after receiving a confirmation instruction, the content in the dimension filling form is standardized and input into the rating selection area;

[0059] The proposal scoring model building module is used to display the scoring standard library in a visual manner, and adaptively output a proposal scoring model for proposal scoring based on the selected scoring dimensions in the scoring selection area.

[0060] The output module is used to obtain the score and rating of the proposal to be scored based on the proposal scoring model.

[0061] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program performing the following steps:

[0062] A comprehensive database is established, comprising: a historical document database, a legal document database, and a historical proposal database. The historical document database includes historical documents extracted from the organization's server, including work priorities and hot topics. The legal document database includes legal texts. The historical proposal database includes proposals from previous years.

[0063] Obtain the proposal to be scored, and obtain a first analysis result based on the similarity between the title of the historical document library and the title of the proposal to be scored. Then, obtain a first scoring dimension based on the first analysis result.

[0064] Based on the citation relationship between the legal document database and the text of the proposal to be scored, a second analysis result is obtained, and a second scoring dimension is obtained based on the second analysis result;

[0065] Based on the similarity between the text of the historical proposal database and the proposal to be scored, a third analysis result is obtained, and a third scoring dimension is obtained based on the third analysis result;

[0066] A rating standard library is established; the rating standard library includes a rating selection area and a rating setting area; the data in the rating selection area includes a first rating dimension, a second rating dimension, and a third rating dimension; the rating setting area has a preset format dimension filling form; the dimension filling form is used to receive input rating dimension data, and after receiving a confirmation instruction, the content in the dimension filling form is standardized and input into the rating selection area;

[0067] The scoring criteria library is displayed in a visual manner, and a proposal scoring model for proposal scoring is adaptively output based on the selected scoring dimensions in the scoring selection area.

[0068] The score and rating of the proposal to be scored are obtained based on the proposed scoring model.

[0069] A computer-readable storage medium having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0070] A comprehensive database is established, comprising: a historical document database, a legal document database, and a historical proposal database. The historical document database includes historical documents extracted from the organization's server, including work priorities and hot topics. The legal document database includes legal texts. The historical proposal database includes proposals from previous years.

[0071] Obtain the proposal to be scored, and obtain a first analysis result based on the similarity between the title of the historical document library and the title of the proposal to be scored. Then, obtain a first scoring dimension based on the first analysis result.

[0072] Based on the citation relationship between the legal document database and the text of the proposal to be scored, a second analysis result is obtained, and a second scoring dimension is obtained based on the second analysis result;

[0073] Based on the similarity between the text of the historical proposal database and the proposal to be scored, a third analysis result is obtained, and a third scoring dimension is obtained based on the third analysis result;

[0074] A rating standard library is established; the rating standard library includes a rating selection area and a rating setting area; the data in the rating selection area includes a first rating dimension, a second rating dimension, and a third rating dimension; the rating setting area has a preset format dimension filling form; the dimension filling form is used to receive input rating dimension data, and after receiving a confirmation instruction, the content in the dimension filling form is standardized and input into the rating selection area;

[0075] The scoring criteria library is displayed in a visual manner, and a proposal scoring model for proposal scoring is adaptively output based on the selected scoring dimensions in the scoring selection area.

[0076] The score and rating of the proposal to be scored are obtained based on the proposed scoring model.

[0077] The aforementioned proposal scoring methods, devices, computer equipment, and storage media establish three independent databases: a historical document database, a legal document database, and a historical proposal database. Data mining is performed from multiple dimensions within the proposal domain to ensure the comprehensiveness and objectivity of the data sources used for matching. For the historical document database, the titles that best reflect the content of the proposal text are extracted and analyzed to obtain the first scoring dimension. For the legal document database, the legal citations that best reflect the legal inclination of the proposal text are extracted and analyzed to obtain the second scoring dimension. For the historical proposal database, the main body of the proposal text that best reflects the events depicted in the text is extracted and analyzed to obtain the third scoring dimension. By establishing different document databases, on the one hand, different levels of features in the proposal text are extracted for matching, improving the objectivity and accuracy of the scoring; on the other hand, matching through the document databases improves the efficiency of proposal scoring. This application also establishes a scoring standard library, which includes a static scoring selection area and a dynamic scoring setting area. The static scoring selection area includes the three scoring dimensions mentioned above, supporting user selection and maintenance. The dynamic scoring setting area supports user customization, solving the problem that the original scoring dimensions become unsuitable when the requirements for scoring dimensions change in different time periods or regions. Furthermore, the scoring standard library is visual and can adaptively output the corresponding proposal scoring model after user operation. By establishing the scoring standard library and proposal scoring model, on the one hand, it is simple, intuitive, and easy to operate, improving the efficiency of proposal scoring; on the other hand, it provides space to change the scoring dimensions as needed, making the scoring more targeted and improving the accuracy of the scoring. Attached Figure Description

[0078] Figure 1 This is a diagram illustrating an application scenario of the proposal scoring method in one embodiment.

[0079] Figure 2 This is a flowchart illustrating a proposal scoring method in one embodiment;

[0080] Figure 3 A structural block diagram of a proposal scoring device in one embodiment;

[0081] Figure 4 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0082] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0083] The method for establishing the proposal scoring model provided in this application can be applied to, for example, Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices, while server 104 can be a server corresponding to various portal websites or work system backends.

[0084] In one embodiment, such as Figure 2 As shown, a method for proposal scoring is provided, which is then applied to... Figure 1 Taking the terminal in the example, the explanation includes:

[0085] Step 202: Establish a comprehensive database.

[0086] The comprehensive database includes: a historical document database, a legal document database, and a historical proposal database. The historical document database contains historical documents extracted from the organization's server, including work priorities and hot topics. The legal document database contains legal texts. The historical proposal database contains proposals from previous years.

[0087] Step 204: Obtain the proposal to be scored. Based on the similarity between the title of the historical document library and the proposal to be scored, obtain the first analysis result. Based on the first analysis result, obtain the first scoring dimension.

[0088] The first analysis result is a similarity mapping value obtained through similarity analysis between the historical document database and the titles of the proposals to be scored. For example, if the similarity is 60%, the mapping value is 60.

[0089] The first rating dimension can be: using the mapped value as the rating for the first rating dimension.

[0090] The first scoring dimension can also be:

[0091] The range of the mapped value is >50, and the score is 35.

[0092] The range of the mapped values ​​is [20, 50), with a score of 30;

[0093] The range of the mapped values ​​is [10, 20), with a score of 20;

[0094] The range of the mapped value is [5, 10), and the score is 10.

[0095] The range of the mapped value is <5, and the score is 0.

[0096] Alternatively, other suitable primary rating dimensions can be selected depending on the specific circumstances.

[0097] Step 206: Based on the citation relationship between the legal document database and the main text of the proposal to be scored, the second analysis result is obtained, and the second scoring dimension is obtained based on the second analysis result.

[0098] The second analytical result is the number of legal document citations obtained by analyzing the citation relationships between the legal document database and the text of the proposal to be evaluated.

[0099] The second scoring dimension could be the number of legal documents cited as the scoring dimension.

[0100] Alternatively, other suitable second scoring dimensions can be selected depending on the specific circumstances.

[0101] Step 208: Based on the similarity between the text of the historical proposal database and the proposal to be scored, the third analysis result is obtained, and the third scoring dimension is obtained based on the third analysis result.

[0102] The third analysis result is the number of similar proposals obtained through similarity analysis between the historical proposal database and the text of the proposals to be scored.

[0103] The third scoring dimension could be the number of similar proposals.

[0104] The third rating dimension can also be:

[0105] The number of similar proposals is 0, and the score is 10.

[0106] The number of similar proposals ranges from [1, 5], with a score of 5;

[0107] The number of similar proposals is greater than 5, and the score is 0.

[0108] Alternatively, other suitable third rating dimensions can be selected depending on the specific circumstances.

[0109] Step 210: Establish a scoring criteria library.

[0110] The rating criteria library includes: a rating selection area and a rating setting area; the data in the rating selection area includes: a first rating dimension, a second rating dimension, and a third rating dimension; the rating setting area has a dimension entry form with a preset format; the dimension entry form is used to receive the input rating dimension data, and after receiving a confirmation instruction, it standardizes the content in the dimension entry form and inputs it into the rating selection area.

[0111] The rating criteria library is a collection of standards for rating proposals, including a rating selection area and a rating setting area.

[0112] The scoring selection area is a static scoring criteria area, including a first scoring dimension, a second scoring dimension, and a third scoring dimension. The first, second, and third scoring dimensions respectively represent the performance of the proposal to be scored in terms of title popularity, legal support, and topic novelty. Based on these dimensions, each proposal to be scored can be scored individually, resulting in a total score and a rating of "excellent," "qualified," or "unqualified," thus completing the selection of excellent proposals.

[0113] The rating settings area is a dynamic rating criteria area where users can enter the desired rating dimension data in the dimension fill-in form.

[0114] The dimension entry form is a structured form containing items such as dimension name, dimension definition, scoring method, and corresponding spaces. Users can fill in the relevant information in the spaces. The dimension entry form uses the same standards as the first, second, and third scoring dimensions, so it can be converted into the scoring dimensions in the scoring selection area after receiving confirmation.

[0115] The confirmation instruction refers to the user's confirmation operation. By clicking the confirmation button on the visualization page, the information entered by the user is converted into the rating dimensions in the rating selection area.

[0116] Step 212: Display the scoring criteria library in a visual manner, and adaptively output the proposal scoring model for proposal scoring based on the selected scoring dimensions in the scoring selection area.

[0117] This method can be applied to Figure 1 The terminal in this context can be a computer desktop. Different users may choose different scoring dimensions, or the same user may choose different scoring dimensions based on different needs. That is, when the input is different, different proposal scoring models can be output accordingly.

[0118] Step 214: Obtain the score and rating of the proposal to be scored based on the proposal scoring model.

[0119] Once the user selects the desired rating dimensions, the proposal rating model is complete. Inputting the proposal to be rated will then yield the corresponding score and rating. The rating includes the score for each rating dimension and the total score.

[0120] The aforementioned proposal scoring methods, devices, computer equipment, and storage media establish three independent databases: a historical document database, a legal document database, and a historical proposal database. Data mining is performed from multiple dimensions within the proposal domain to ensure the comprehensiveness and objectivity of the data sources used for matching. For the historical document database, the titles that best reflect the content of the proposal text are extracted and analyzed to obtain the first scoring dimension. For the legal document database, the legal citations that best reflect the legal inclination of the proposal text are extracted and analyzed to obtain the second scoring dimension. For the historical proposal database, the main body of the proposal text that best reflects the events depicted in the text is extracted and analyzed to obtain the third scoring dimension. By establishing different document databases, on the one hand, different levels of features in the proposal text are extracted for matching, improving the objectivity and accuracy of the scoring; on the other hand, matching through the document databases improves the efficiency of proposal scoring. This application also establishes a scoring standard library, which includes a static scoring selection area and a dynamic scoring setting area. The static scoring selection area includes the three scoring dimensions mentioned above, supporting user selection and maintenance. The dynamic scoring setting area supports user customization, solving the problem that the original scoring dimensions become unsuitable when the requirements for scoring dimensions change in different time periods or regions. Furthermore, the scoring standard library is visual and can adaptively output the corresponding proposal scoring model after user operation. By establishing the scoring standard library and proposal scoring model, on the one hand, it is simple, intuitive, and easy to operate, improving the efficiency of proposal scoring; on the other hand, it provides space to change the scoring dimensions as needed, making the scoring more targeted and improving the accuracy of the scoring.

[0121] It should be understood that, although Figure 2 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 2 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0122] In one embodiment, adaptively outputting a proposal scoring model for proposal scoring based on the selected scoring dimension in the scoring selection area includes: obtaining the importance ratio corresponding to the selected scoring dimension through visualization, determining whether the selected scoring dimension belongs to the marker dimension, and outputting the proposal scoring model for proposal scoring based on the importance ratio and whether the selected scoring dimension belongs to the marker dimension; the proposal scoring model is used to output the proposal score and rating; when the selected scoring dimension belongs to the marker dimension and the score of the selected scoring dimension reaches a set value, the corresponding rating is directly output; when the selected scoring dimension does not belong to the marker dimension, or the selected scoring dimension belongs to the marker dimension and the score of the selected scoring dimension is less than the set value, the score is calculated based on the importance ratio of the selected scoring dimension, and the corresponding rating is determined and output based on the score.

[0123] Users can set the importance ratio for selected scoring dimensions according to their needs. For example, when the region focuses more on the hot topics of the year, the first scoring dimension can be set with the highest importance ratio; when the focus is more on the popularization of law in that year, the second scoring dimension can be set with the highest importance ratio. Importance ratios can also be set based on the degree of importance, which can be determined through discussions with proposal reviewers. Specifically: Title popularity > Legal support > Novelty of the topic.

[0124] Users can determine whether a selected rating dimension belongs to a flag dimension, and when a rating dimension belongs to a flag dimension, a set value can be assigned. For example, if the first rating dimension belongs to a flag dimension and the set value is 80, then when rating a proposal, if the score for the first rating dimension is 80, the proposal can be directly rated as excellent without having to rate other rating dimensions and obtain the total score. There can be one or more flag dimensions. The set values ​​for different flag dimensions can be the same or different.

[0125] If the selected scoring dimension, importance ratio, and whether the selected scoring dimension belongs to any of the label dimensions are not set, the proposal scoring model for proposal scoring can be output according to the initial settings.

[0126] Initial settings include:

[0127] The selected rating dimensions include all rating dimensions in the rating selection area; assuming there are n rating dimensions, the importance ratios are respectively... None of the selected rating dimensions are characteristic dimensions.

[0128] For example, if the rating selection area includes: the first rating dimension, the second rating dimension, and the third rating dimension, then the initial setting is: the selected rating dimensions are the first rating dimension, the second rating dimension, and the third rating dimension; their corresponding importance ratios are 1, 2 / 3, and 1 / 3, respectively; none of them belong to the label dimension.

[0129] Initial settings can be modified and saved.

[0130] In this embodiment, the total score and rating are obtained according to the initially set proposal scoring model as follows:

[0131]

[0132] In the formula, f ( x () is the total score. x i This represents the rating for the i-th rating dimension, where i = 1, 2, ..., n, where n is the number of rating dimensions and k is the importance ratio corresponding to the rating dimension. .

[0133] Right now:

[0134]

[0135] Original settings: A total score > 80 is rated as excellent, a total score between 50 and 80 (inclusive) is rated as qualified, and a total score < 50 is rated as unqualified.

[0136] The original settings can be modified and saved.

[0137] In one embodiment, obtaining the first analysis result based on the similarity between the historical document library and the title of the proposal to be rated includes: performing similarity analysis on each historical document using the cosine similarity algorithm to obtain an optimal library; clustering each historical document in the optimal library using the K-means clustering algorithm to obtain clusters; calculating the number of historical documents in each cluster to obtain hotspot clusters; extracting keywords from each historical document in the hotspot clusters, calculating the keyword frequency of each historical document, and sorting each keyword by keyword frequency to obtain the topic words of the hotspot clusters; combining the topic words to obtain popular topics; segmenting the title of the proposal to be rated using a pre-set word segmentation library to obtain a title word set; calculating the weight of each word in the title word set using the TF-IDF algorithm to obtain an optimal set; and performing similarity analysis between the optimal set and popular topics using the cosine similarity algorithm to obtain the first analysis result.

[0138] Cosine similarity, also known as cosine similarity, assesses the similarity between two vectors by calculating the cosine of the angle between them. Cosine similarity plots the vectors in a vector space, such as the most common two-dimensional space, based on their coordinate values.

[0139] Each historical file is treated as an individual. The preferred database is created by sorting each individual file according to its similarity to other files in descending order, removing those with very low similarity. For example, the top 20 or 50 historical files can be selected as the preferred database; the specific settings can be tailored to specific needs and will not be elaborated further here. The preferred database is used for individual screening before clustering. Traditional clustering algorithms consider all data, which can lead to some individuals with significant differences from other samples, greatly interfering with the cluster centroids. Such individuals can be completely discarded during our analysis to avoid affecting the process. Therefore, before clustering, the similarity between each historical file is calculated, and historical files with very low similarity to other historical files are directly excluded.

[0140] K-means clustering is an iterative clustering algorithm. Its steps are as follows: First, the data is pre-divided into K groups. Then, K objects are randomly selected as initial cluster centers. Next, the distance between each object and each seed cluster center is calculated, and each object is assigned to the nearest cluster center. Each cluster center and the objects assigned to it represent a cluster. Each time a sample is assigned, the cluster centers are recalculated based on the existing objects in the cluster. This process is repeated until a termination condition is met. The termination condition may be that no (or a minimum number) objects are reassigned to different clusters, no (or a minimum number) cluster centers change, or the sum of squared errors reaches a local minimum.

[0141] Hotspot clusters are those clusters whose historical files are ranked from largest to smallest, and whose top-ranking clusters are those whose keywords are ranked from largest to smallest, based on the number of historical files in each cluster. Combinations refer to the random combination of keywords from hotspot clusters; popular topics are combinations of keywords from hotspot clusters. For example, if the keywords for hotspot cluster 1 are A, C, E, G, and I, and the keywords for hotspot cluster 2 are A, D, E, F, and H, then popular topics would be ACE, ADE, and ACDE, etc. Popular topics are obtained by filtering the historical file database; they represent the categorization and refinement of historical files.

[0142] The word segmentation library is a set of words that provides the basis for Chinese word segmentation. Based on this set, any document can be segmented into Chinese. The title word set is obtained by segmenting the title of the proposal to be evaluated; it is a series of words related to the main content of the proposal. The TF-IDF algorithm is a statistical method used to evaluate the importance of a word to a document within a document set or corpus. The importance of a word increases proportionally to its frequency in the document, but decreases inversely proportionally to its frequency in the corpus. The optimal set is the top few words selected by sorting the weights of each word in the title word set from largest to smallest. For example, the top 3 words or the top 5 words can be selected as the optimal set; the specific settings can be configured according to actual needs and will not be elaborated further here.

[0143] The first analysis result indicates that when the similarity between each word in the preferred set and a popular topic is above 70%, the number of keywords of that popular topic is used as the heat value. The higher the heat value, the more popular the title is.

[0144] In one embodiment, obtaining the first analysis result based on the similarity between the historical document library and the title of the proposal to be rated includes: extracting keywords from the historical documents and calculating the weight of each keyword; sorting each keyword in descending order of weight and selecting the keywords with the highest weight as hot words; creating an index for each hot word and using the number of times the hot word appears in the historical documents as the index value; obtaining the expected value of the index value and determining popular topics based on the index value reaching the expected value; segmenting the title of the proposal to be rated into words according to a pre-set word segmentation library to obtain a title word set; calculating the weight of each word in the title word set according to the TF-IDF algorithm to obtain an optimal set; and performing similarity analysis between the optimal set and popular topics according to the cosine similarity algorithm to obtain the first analysis result.

[0145] For example, if there are 4 historical files, the top 6 keywords by weight will result in the following hot keywords: school, education, nutrition, teacher, student, and sports.

[0146] If we use doc to represent historical files, then:

[0147] doc1: School, Education, Nutrition, Teacher

[0148] doc2: School, Education, Sports

[0149] doc3: School, Education, Teacher

[0150] doc4: Education, Sports, Students, Teachers

[0151] After creating the index, the following structure is obtained:

[0152] School: doc1, doc2, doc3

[0153] Education: doc1, doc2, doc3, doc4

[0154] Nutrition: doc1

[0155] Teacher: doc1, doc3, doc4

[0156] Student: doc4

[0157] Sports: doc3, doc4

[0158] Correspondingly, the index values ​​are:

[0159] School: 3

[0160] Education: 4

[0161] Nutrition: 1

[0162] Teacher: 3

[0163] Student: 1

[0164] Sports: 2

[0165] If the expected value of the index is 3, then the popular topics will be a combination of schools, education, and teachers.

[0166] In one embodiment, obtaining the second analysis result based on the citation relationship between the legal document library and the text of the proposal to be scored includes: constructing a keyword matching template and a token matching template based on the legal text in the legal document library; both the keyword matching template and the token matching template include content to be filled; concatenating the keyword matching template and the token matching template to obtain a citation relationship matching template; matching the proposal to be analyzed with the citation relationship matching template to obtain the corresponding text content in the content to be filled; matching the text content with the legal document library to obtain the cited words and citation frequency; matching the proposal to be scored with the historical document library to obtain the number of citations; obtaining a first score based on the cited words and citation frequency; performing a heat value analysis on the proposal to be scored and the historical document library based on the number of citations to obtain a second score; and obtaining the second analysis result based on the first score and the second score.

[0167] Keyword matching templates refer to a series of directional words in legal texts, such as "Article 1" or "Paragraph 1"; marker matching templates refer to symbols in legal texts that may indicate clauses, such as book titles or quotation marks; citation matching templates are matching templates obtained by concatenating keyword matching templates and marker matching templates; the text content to be filled refers to the laws that may be cited in the proposal to be scored, obtained after matching with citation matching templates.

[0168] Cited words and citation frequency refer to the legal documents cited in the proposal to be extracted and the number of times they are cited; citation count refers to the number of times the legal documents cited in the proposal to be extracted appear in the historical document database. If they appear more often in the historical document database, it means that the legal document has been reprinted, reported, and interpreted more often, and is therefore considered to be popular.

[0169] The first score refers to the frequency of legal document citations. This citation frequency can represent the degree of legal support; the higher the citation frequency, the greater the legal support, indicating that the proposal to be extracted has legal support, aligns with the organization's development expectations, and is likely to be an excellent proposal. Analyzing it can add a criterion for judging excellent proposals. Specifically:

[0170] The frequency of legal document citations ranges from [2, 5], with a first score of 20;

[0171] The legal document was cited 1 time, and the first score was 15.

[0172] The legal document was cited 0 times, resulting in a score of 0.

[0173] The first score can also be set in other ways.

[0174] The second score refers to the number of times legal documents are cited. This number of citations reflects the popularity of the law; the more citations, the more popular the law, indicating that the proposal to be extracted closely follows the institution's trends and is likely to be an excellent proposal. Analyzing it can add a criterion for judging excellent proposals. Specifically:

[0175] The range for legal document citations is >20, with a second score of 25.

[0176] The range of legal document citations is [10, 20), with a second score of 20;

[0177] The range of legal document citations is [5, 10), with a second score of 15;

[0178] The range of legal document citations is [1, 5), with a second score of 10.

[0179] The second score can also be set in other ways.

[0180] The second analysis result refers to the sum of the frequency of legal document citations and the number of times legal documents are cited. The second score is only required when the first score is based on citations; otherwise, the second score is 0.

[0181] In one embodiment, obtaining the third analysis result based on the similarity between the text of the historical proposal database and the proposal to be rated includes: segmenting each proposal in the historical proposal database using a pre-set word segmentation database to obtain a proposal database word set; calculating the weight of each word in the proposal database word set using the TF-IDF algorithm to obtain a key set; segmenting the proposal to be rated using a pre-set word segmentation database to obtain a proposal word set; calculating the weight of each word in the proposal word set using the TF-IDF algorithm to obtain a key set; and performing similarity analysis between the key set and the key set using the cosine similarity algorithm to obtain the third analysis result.

[0182] The proposal database word set is a series of words obtained by segmenting each proposal in the database. The key set is the top few words in the proposal database word set, sorted by the weight of each word in descending order; for example, the top 20 words are taken as the key set. The proposal word set is a series of words obtained by segmenting the content of the proposals to be evaluated. The emphasis set is the top few words in the proposal word set, sorted by the weight of each word in descending order. The third analysis result refers to the number of proposals with a weight of 0.05 or higher in the emphasis set and a similarity of more than 40% with the key set.

[0183] In one embodiment, the scoring selection area further includes: a fourth scoring dimension; performing an argument sufficiency analysis on the proposal to be scored to obtain a fourth analysis result, and obtaining a fourth scoring dimension based on the fourth analysis result; performing an argument sufficiency analysis on the proposal to be scored to obtain a fourth analysis result includes: identifying feature words in the proposal to be scored based on a pre-set feature word set, and obtaining the word count of each paragraph in which the feature words appear; the feature words are logical words that can characterize the arguments and evidence of the proposal to be extracted; and obtaining the fourth analysis result based on the word count of the relevant paragraphs.

[0184] The sufficiency of arguments and evidence refers to whether the arguments in the proposed proposal offer solutions and whether the evidence is sufficient. If the arguments and evidence are sufficient, it indicates that the proposed proposal has feasible value and may be an excellent proposal. Analyzing the sufficiency of arguments and evidence can add a criterion for judging excellent proposals. Characteristic words can be: "first," "second," "firstly," "then," etc.

[0185] The fourth analysis result refers to the sum of the number of words in each paragraph containing the characteristic words.

[0186] The fourth rating dimension could be:

[0187] The word count ranges from [500 to 1000], and the score is 10.

[0188] The word count ranges from 200 to 500, and the score is 5.

[0189] The word count ranges from [0, 200), and the score is 0.

[0190] Alternatively, other suitable fourth rating dimensions can be selected depending on the specific circumstances.

[0191] In one embodiment, such as Figure 3 As shown, a proposal scoring device is provided, comprising: a preparation module 302, a first analysis module 304, a second analysis module 306, a third analysis module 308, a scoring standard library establishment module 310, a proposal scoring model establishment module 312, and an output module 314, wherein:

[0192] Preparation module 302 is used to establish a comprehensive database, which includes: a historical document database, a legal document database, and a historical proposal database; the historical document database includes historical documents extracted from the organization's server, including work priorities and hot topics; the legal document database includes legal texts; and the historical proposal database includes proposals from previous years.

[0193] The first analysis module 304 is used to obtain the proposal to be scored, obtain a first analysis result based on the similarity between the title of the historical document library and the title of the proposal to be scored, and obtain a first scoring dimension based on the first analysis result.

[0194] The second analysis module 306 is used to obtain a second analysis result based on the citation relationship between the legal document library and the text of the proposal to be scored, and to obtain a second scoring dimension based on the second analysis result.

[0195] The third analysis module 308 is used to obtain a third analysis result based on the similarity between the text of the historical proposal database and the proposal to be scored, and to obtain a third scoring dimension based on the third analysis result.

[0196] The scoring criteria library establishment module 310 is used to establish a scoring criteria library; the scoring criteria library includes: a scoring selection area and a scoring setting area; the data in the scoring selection area includes: a first scoring dimension, a second scoring dimension, and a third scoring dimension; the scoring setting area has a preset format dimension filling form; the dimension filling form is used to receive input scoring dimension data, and after receiving a confirmation instruction, the content in the dimension filling form is standardized and input into the scoring selection area.

[0197] The proposal scoring model building module 312 is used to display the scoring standard library in a visual manner and adaptively output a proposal scoring model for proposal scoring based on the scoring dimensions selected in the scoring selection area.

[0198] Output module 314 is used to obtain the score and rating of the proposal to be scored according to the proposal scoring model.

[0199] In one embodiment, the proposal scoring model building module 312 is further configured to adaptively output a proposal scoring model for proposal scoring based on the selected scoring dimensions in the scoring selection area, including: obtaining the importance ratio corresponding to the selected scoring dimension through visualization, and determining whether the selected scoring dimension belongs to the marker dimension; outputting the proposal scoring model for proposal scoring based on the importance ratio and whether the selected scoring dimension belongs to the marker dimension; the proposal scoring model is used to output the proposal score and rating; when the selected scoring dimension belongs to the marker dimension and the score of the selected scoring dimension reaches a set value, the corresponding rating is directly output; when the selected scoring dimension does not belong to the marker dimension, or the selected scoring dimension belongs to the marker dimension and the score of the selected scoring dimension is less than the set value, the score is calculated based on the importance ratio of the selected scoring dimension, and the corresponding rating is determined and output based on the score.

[0200] In one embodiment, the first analysis module 304 is further configured to obtain a first analysis result based on the similarity between the historical file library and the title of the proposal to be scored, including: performing similarity analysis on each historical file according to the cosine similarity algorithm to obtain an optimal library; clustering each historical file in the optimal library according to the K-means clustering algorithm to obtain clusters; calculating the number of historical files in each cluster to obtain hotspot clusters; extracting keywords from each historical file in the hotspot clusters, calculating the keyword frequency of each historical file, and sorting each keyword according to the keyword frequency to obtain the topic words of the hotspot clusters; combining the topic words to obtain popular topics; segmenting the title of the proposal to be scored according to a pre-set word segmentation library to obtain a title word set; calculating the weight of each word in the title word set according to the TF-IDF algorithm to obtain an optimal set; and performing similarity analysis between the optimal set and the popular topics according to the cosine similarity algorithm to obtain the first analysis result.

[0201] In one embodiment, the first analysis module 304 is further configured to obtain a first analysis result based on the similarity between the historical file library and the title of the proposal to be scored, including: extracting keywords from the historical files and calculating the weight of each keyword; sorting each keyword in descending order of weight and taking the keywords with higher weights as hot words; establishing an index for each hot word and using the number of times the hot word appears in the historical files as the index value of the index; obtaining the expected value of the index value and determining popular topics based on the index value reaching the expected value; segmenting the title of the proposal to be scored according to a pre-set word segmentation library to obtain a title word set; calculating the weight of each word in the title word set according to the TF-IDF algorithm to obtain an optimal set; and performing similarity analysis between the optimal set and popular topics according to the cosine similarity algorithm to obtain the first analysis result.

[0202] In one embodiment, the second analysis module 306 is further configured to obtain a second analysis result based on the citation relationship between the legal document library and the text of the proposal to be scored, including: constructing a keyword matching template and a marker matching template based on the legal text in the legal document library; both the keyword matching template and the marker matching template include content to be filled; concatenating the keyword matching template and the marker matching template to obtain a citation relationship matching template; matching the proposal to be analyzed based on the citation relationship matching template to obtain the corresponding text content in the content to be filled; matching the text content based on the legal document library to obtain the cited words and citation frequency; matching the proposal to be scored based on the historical document library to obtain the number of citations; obtaining a first score based on the cited words and citation frequency; performing heat value analysis on the proposal to be scored and the historical document library based on the number of citations to obtain a second score; and obtaining a second analysis result based on the first score and the second score.

[0203] In one embodiment, the third analysis module 308 is further configured to obtain a third analysis result based on the similarity between the historical proposal database and the body text of the proposal to be scored, including: segmenting each proposal in the historical proposal database according to a pre-set word segmentation database to obtain a proposal database word set; calculating the weight of each word in the proposal database word set according to the TF-IDF algorithm to obtain a key set; segmenting the proposal to be scored according to a pre-set word segmentation database to obtain a proposal word set; calculating the weight of each word in the proposal word set according to the TF-IDF algorithm to obtain a key set; and performing similarity analysis between the key set and the key set according to the cosine similarity algorithm to obtain the third analysis result.

[0204] In one embodiment, the scoring criteria library establishment module 310 is further configured to include the following in the scoring selection area: a fourth scoring dimension; performing an argument sufficiency analysis on the proposal to be scored to obtain a fourth analysis result, and obtaining a fourth scoring dimension based on the fourth analysis result; performing an argument sufficiency analysis on the proposal to be scored to obtain a fourth analysis result includes: identifying feature words in the proposal to be scored based on a pre-set feature word set, and obtaining the number of words in each paragraph where the feature words appear; the feature words are logical words that can characterize the arguments and evidence of the proposal to be extracted; and obtaining the fourth analysis result based on the number of words in the relevant paragraphs.

[0205] Specific limitations regarding the device for evaluating proposals can be found in the limitations regarding the method for evaluating proposals mentioned above, and will not be repeated here. Each module in the aforementioned device for evaluating proposals can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0206] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 4 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a proposal scoring method. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0207] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0208] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the method described above.

[0209] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0210] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0211] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0212] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for evaluating proposals, characterized in that, The method includes: A comprehensive database is established, comprising: a historical document database, a legal document database, and a historical proposal database. The historical document database includes historical documents extracted from the organization's server, including work priorities and hot topics. The legal document database includes legal texts. The historical proposal database includes proposals from previous years. Obtain the proposal to be scored, extract the title that best reflects the content of the proposal text for analysis, obtain the first analysis result based on the similarity between the title of the historical file library and the title of the proposal to be scored, and obtain the first scoring dimension based on the first analysis result; The legal citation relationships that best reflect the legal orientation of the proposal text are extracted and analyzed. Based on the citation relationship between the legal document database and the main text of the proposal to be scored, a second analysis result is obtained, and a second scoring dimension is obtained based on the second analysis result. The text that best reflects the event in the proposal is extracted and analyzed. Based on the similarity between the text of the historical proposal database and the text of the proposal to be scored, a third analysis result is obtained. Based on the third analysis result, a third scoring dimension is obtained. A rating standard library is established; the rating standard library includes a rating selection area and a rating setting area; the data in the rating selection area includes a first rating dimension, a second rating dimension, and a third rating dimension; the rating setting area has a preset format dimension filling form; the dimension filling form is used to receive input rating dimension data, and after receiving a confirmation instruction, the content in the dimension filling form is standardized and input into the rating selection area; The scoring criteria library is displayed in a visual manner, and a proposal scoring model for proposal scoring is adaptively output based on the selected scoring dimensions in the scoring selection area. The proposed scoring model is used to obtain the scores and ratings of the proposals to be scored, thereby improving the accuracy and efficiency of screening excellent proposals. Based on the selected rating dimensions in the rating selection area, the adaptive output of the proposal rating model for proposal rating includes: Obtain the importance ratio corresponding to the selected rating dimension through visualization, or set the importance ratio corresponding to the selected rating dimension according to the requirements, and determine whether the selected rating dimension belongs to the label dimension. Based on the importance ratio and whether the selected rating dimension belongs to the label dimension, output the proposal scoring model for proposal scoring; the proposal scoring model is used to output the proposal score and rating. When the selected rating dimension belongs to the indicator dimension, and the rating of the selected rating dimension reaches the set value, the corresponding rating is directly output. When the selected rating dimension is not a marker dimension, or when the selected rating dimension is a marker dimension and the rating of the selected rating dimension is less than the set value, the rating is calculated based on the importance ratio of the selected rating dimension, and the corresponding rating is determined and output based on the rating.

2. The method according to claim 1, characterized in that, Based on the similarity between the titles of the historical document database and the proposals to be evaluated, the first analysis results include: Based on the cosine similarity algorithm, a similarity analysis is performed on each of the historical files to obtain an optimal library; Based on the K-means clustering algorithm, each historical file in the preferred library is clustered to obtain clusters; Calculate the number of historical files for each cluster class to obtain the hotspot clusters; Keywords are extracted from each historical file of the hotspot cluster, the frequency of keywords in each historical file is calculated, and each keyword is sorted according to the frequency of keywords to obtain the topic words of the hotspot cluster; By combining the aforementioned keywords, popular topics can be obtained; Based on a pre-set word segmentation library, the title of the proposal to be scored is segmented to obtain a set of title words; Based on the TF-IDF algorithm, the weight of each word in the title word set is calculated to obtain the optimal set; Based on the cosine similarity algorithm, the preferred set is analyzed for similarity with the popular topics to obtain the first analysis result.

3. The method according to claim 1, characterized in that, Based on the similarity between the titles of the historical document database and the proposals to be evaluated, the first analysis results include: Extract keywords from the historical documents and calculate the weight of each keyword. Sort each keyword in descending order of weight and select the keywords with the highest weight as hot words. An index is created for each of the hot words, and the index value is used based on the number of times the hot word appears in the historical files; Obtain the expected value of the index value, and determine popular topics based on the indexes where the index value reaches the expected value; Based on a pre-set word segmentation library, the title of the proposal to be scored is segmented to obtain a set of title words; Based on the TF-IDF algorithm, the weight of each word in the title word set is calculated to obtain the optimal set; Based on the cosine similarity algorithm, the preferred set is analyzed for similarity with popular topics to obtain the first analysis result.

4. The method according to claim 2 or 3, characterized in that, Based on the citation relationship between the legal document database and the text of the proposal to be scored, the second analysis result includes: Based on the legal texts in the legal document database, keyword matching templates and token matching templates are constructed respectively; both keyword matching templates and token matching templates include content to be filled in. The keyword matching template and the tag matching template are concatenated to obtain the citation relationship matching template; Based on the reference relationship matching template, the proposal to be analyzed is matched to obtain the corresponding text content in the content to be filled; Based on the legal document database, the text content is matched to obtain the cited words and citation frequency; based on the historical document database, the proposal to be scored is matched to obtain the number of citations. The highest score is awarded based on the cited words and their frequency. Based on the number of citations, a heat value analysis is performed on the proposals to be scored and the historical proposal database to obtain a second score; Based on the first and second scores, the second analysis result is obtained.

5. The method according to claim 4, characterized in that, Based on the similarity between the text of the historical proposal database and the proposal to be scored, the third analysis result includes: Based on a pre-set word segmentation database, each proposal in the historical proposal database is segmented into words to obtain a word set for the proposal database. Based on the TF-IDF algorithm, the weight of each word in the proposal database word set is calculated to obtain the key set; Based on a pre-set word segmentation library, the proposals to be scored are segmented to obtain a set of proposal words; Based on the TF-IDF algorithm, the weight of each word in the proposal word set is calculated to obtain the key set; Based on the cosine similarity algorithm, the similarity analysis is performed between the key set and the focus set to obtain the third analysis result.

6. The method according to claim 5, characterized in that, The rating selection area also includes: a fourth rating dimension; A sufficiency analysis of arguments and evidence is performed on the proposals to be scored to obtain a fourth analysis result, and a fourth scoring dimension is obtained based on the fourth analysis result. An analysis of the sufficiency of arguments and evidence in the proposals to be evaluated yielded the following fourth analysis results: Based on a pre-set set of feature words, feature word identification is performed on the proposals to be scored to obtain the word count of each paragraph in which the feature words appear; the feature words are logical words that characterize the arguments and evidence of the proposals to be extracted. The fourth analysis result was obtained based on the word count of the relevant paragraphs.

7. A device for scoring proposals, characterized in that, The device includes: The preparation module is used to establish a comprehensive database, which includes: a historical document database, a legal document database, and a historical proposal database. The historical document database includes historical documents extracted from the organization's server, including work priorities and hot topics. The legal document database includes legal texts. The historical proposal database includes proposals from previous years. The first analysis module is used to obtain the proposal to be scored, obtain the first analysis result based on the similarity between the title of the historical document library and the title of the proposal to be scored, and obtain the first scoring dimension based on the first analysis result. The second analysis module is used to obtain a second analysis result based on the citation relationship between the legal document library and the text of the proposal to be scored, and to obtain a second scoring dimension based on the second analysis result; The third analysis module is used to obtain a third analysis result based on the similarity between the text of the historical proposal database and the proposal to be scored, and to obtain a third scoring dimension based on the third analysis result. A rating criteria library creation module is used to create a rating criteria library; the rating criteria library includes: a rating selection area and a rating setting area; the data in the rating selection area includes: a first rating dimension, a second rating dimension, and a third rating dimension; the rating setting area sets a dimension filling form with a preset format; the dimension filling form is used to receive input rating dimension data, and after receiving a confirmation instruction, the content in the dimension filling form is standardized and input into the rating selection area; The proposal scoring model building module is used to display the scoring standard library in a visual manner, and adaptively output a proposal scoring model for proposal scoring based on the selected scoring dimensions in the scoring selection area. The output module is used to obtain the score and rating of the proposal to be scored based on the proposal scoring model. The proposal scoring model building module is also used to adaptively output a proposal scoring model for proposal scoring based on the selected scoring dimensions in the scoring selection area, including: Obtain the importance ratio corresponding to the selected rating dimension through visualization, or set the importance ratio corresponding to the selected rating dimension according to the requirements, and determine whether the selected rating dimension belongs to the label dimension. Based on the importance ratio and whether the selected rating dimension belongs to the label dimension, output the proposal scoring model for proposal scoring; the proposal scoring model is used to output the proposal score and rating. When the selected rating dimension belongs to the indicator dimension, and the rating of the selected rating dimension reaches the set value, the corresponding rating is directly output. When the selected rating dimension is not a marker dimension, or when the selected rating dimension is a marker dimension and the rating of the selected rating dimension is less than the set value, the rating is calculated based on the importance ratio of the selected rating dimension, and the corresponding rating is determined and output based on the rating.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images