Enterprise policy matching method, equipment and medium

Abstract

The invention discloses an enterprise policy matching method and device and a medium, and the method comprises the steps: crawling policy data of a government website, and extracting policy multi-dimensional label information of each policy in the policy data; extracting enterprise label information from the enterprise data of the to-be-matched enterprise; according to a principal component analysis model, calculating the policy multi-dimensional label information and the enterprise label information to obtain a matching degree between the enterprise and each policy; and pushing the target policy when the matching degree is higher than a preset matching threshold value to the enterprise. Based on massive and complicated policy data, accurate policy pushing is provided for enterprises.

Description

Technical Field

[0001] This application relates to the field of big data analysis technology, and particularly to an enterprise policy matching method, device, and medium. Background Art

[0002] Usually, government departments at all levels formulate a large number of industrial support and enterprise-benefiting policies every year. Due to the lack of a bridge for efficient information exchange, many policies are formulated but not known to people. For the policies that enterprises are concerned about, they need to search through channels such as the websites of industry departments, WeChat official accounts, and various WeChat groups, etc., and it is easy to miss or overlook them. There is often a disconnect between the government and enterprises in policy services.

[0003] Currently, it is difficult to obtain policy data: Policy information is usually scattered in various websites and channels. Enterprises need to spend a lot of time and energy to search and filter out the relevant policies they want, and the difficulty of information acquisition is very high. At the same time, policies have a high timeliness, and it is easy to miss important policies. In addition, many enterprises wait for others to send subsidy notices in the enterprise group. The information acquisition channels are relatively single and passive, and it is easy to miss the application opportunity. Moreover, policy data cannot be utilized: The document information is complex and diverse. When enterprises face a vast amount of policy information, many policies are not relevant to themselves. Enterprises need policies that highly match themselves and filter out irrelevant policy information. Additionally, policy services are difficult: Due to the lack of a bridge for efficient information exchange, there is often a disconnect between the government and enterprises in policy services. The government cannot accurately deliver policies, resulting in the dilemma that although the policies are good, they cannot be reached or utilized. Therefore, it is impossible to accurately match corresponding policies for enterprises, and the matching efficiency is low. Summary of the Invention

[0004] Embodiments of this application provide an enterprise policy matching method, device, and medium, which are used to solve the problem that it is impossible to accurately match corresponding policies for enterprises and the matching efficiency is low.

[0005] Embodiments of this application adopt the following technical solutions:

[0006] On the one hand, embodiments of this application provide an enterprise policy matching method, which includes: crawling policy data from government websites, and extracting policy multi-dimensional label information of each policy in the policy data; extracting enterprise label information from the enterprise data of the enterprise to be matched; calculating the policy multi-dimensional label information and the enterprise label information according to the principal component analysis model to obtain the matching degree between the enterprise and each policy; and pushing the target policy with a matching degree higher than a preset matching threshold to the enterprise.

[0007] In one example, extracting the policy multi-dimensional label information of each policy in the policy data specifically includes: performing natural language processing on the policy data to obtain the policy semantic text of each policy; vectorizing the policy semantic text to obtain a policy text vector; and labeling the policy text vector according to a preset policy multi-dimensional label to obtain the policy multi-dimensional label information of each policy.

[0008] In one example, calculating the matching degree between the enterprise and each policy according to the principal component analysis model for the policy multi-dimensional label information and the enterprise label information specifically includes: performing data standardization processing on the policy multi-dimensional label information and the enterprise label information to obtain standardized policy label values and standardized enterprise label values; obtaining a standardized data matrix according to the standardized policy label values and the standardized enterprise label values; obtaining a characteristic matrix corresponding to the covariance matrix according to the standardized data matrix, and selecting the principal components with the cumulative contribution rate exceeding a preset threshold from the characteristic matrix; calculating the load of each label on the principal component to determine the weight of each label on the principal component; and obtaining the matching degree between the enterprise and each policy by calculating the distances between the enterprise and each policy in the principal component space respectively.

[0009] In one example, obtaining the characteristic matrix corresponding to the covariance matrix according to the standardized data matrix specifically includes: obtaining the covariance matrix according to the covariance matrix formula and the standardized data matrix; calculating the eigenvalues and eigenvectors of the covariance matrix; sorting the multiple eigenvalues from largest to smallest, selecting the eigenvectors corresponding to the first preset number of eigenvalues, and selecting the principal components with the cumulative contribution rate reaching a preset contribution threshold to obtain the characteristic matrix.

[0010] In one example, pushing the target policy with a matching degree higher than a preset matching threshold to the enterprise specifically includes: determining initial push label information from the policy multi-dimensional label information of the target policy and generating the original link of the target policy; and fusing the push label information, the original link, and the matching degree to generate the push label information of the target policy to push the target policy to the enterprise.

[0011] In one example, the method further includes: obtaining the policy keywords uploaded by the client; retrieving the policy keywords from the policy database according to multi-condition combinations to obtain a target policy combination, and feeding back the policy target combination and the original link of each policy in the policy target combination to the client.

[0012] In one example, extracting enterprise label information from the enterprise data of the enterprise to be matched specifically includes: matching the industrial and commercial information of the enterprise according to the registration information of the enterprise to be matched; the registration information includes the enterprise name; and extracting the enterprise label information of the enterprise from the industrial and commercial information according to the preset enterprise label.

[0013] In one example, before extracting the policy multi-dimensional label information of each policy in the policy data, the method further includes: removing duplicates of the same policy repeatedly published on different government websites in the policy data; filling in missing values in the policy data; and processing outliers in the policy data.

[0014] On the other hand, an embodiment of the present application provides an enterprise policy matching device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute an enterprise policy matching method as described in any one of the above.

[0015] On the other hand, an embodiment of the present application provides a non-volatile computer storage medium for enterprise policy matching, storing computer-executable instructions that can execute an enterprise policy matching method as described in any one of the above.

[0016] The above at least one technical solution adopted in the embodiments of the present application can achieve the following beneficial effects:

[0017] Through big data methods, various policy document data are collected in real time with high frequency, helping users to timely grasp the policy trends and directions. In addition, the collection, processing, and labeling of massive and multi-source policy data support users to perform fuzzy queries or multi-condition custom queries, improving work efficiency. In addition, a policy matching model is established to realize dual portraits of policies and enterprises, and more accurately push suitable policies to enterprises, reducing their screening effort and cost. Thus, based on massive and complex policy data, through methods such as text processing, intelligent labeling, and intelligent matching, policy precise push is provided for enterprises. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the present application, some embodiments of the present application will be described in detail below with reference to the drawings, in which:

[0019] Figure 1 is a flowchart of an enterprise policy matching method provided by an embodiment of the present application;

[0020] Figure 2 is a framework diagram of an enterprise policy matching system provided by an embodiment of the present application;

[0021] Figure 3 This is a schematic structural diagram of an enterprise policy matching device provided by an embodiment of the present application. Detailed implementation manners

[0022] To make the objectives, technical solutions, and advantages of the present application clearer, the technical solutions of the present application will be clearly and completely described below in conjunction with specific embodiments and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0023] The following will refer to the drawings to detail some embodiments of the present application.

[0024] Figure 1 This is a schematic flow diagram of an enterprise policy matching method provided by an embodiment of the present application. This method can be applied to different business fields, such as the Internet finance business field, the e-commerce business field, the instant messaging business field, the game business field, the official business field, etc. Some input parameters or intermediate results in this process allow manual intervention and adjustment to help improve accuracy.

[0025] The implementation of the analysis method involved in the embodiments of the present application can be a terminal device or a server, and the present application does not make special restrictions on this. For the convenience of understanding and description, the following embodiments will be described in detail taking the server as an example.

[0026] It should be noted that this server can be a single device or a system composed of multiple devices, that is, a distributed server, and the present application does not make specific limitations on this.

[0027] Figure 1 The process in [[ ]] includes the following steps:

[0028] S101: Crawl the policy data of government websites and extract the policy multi-dimensional label information of each policy in the policy data.

[0029] It should be noted that the policy data is government public data, sourced from the policy information publicly available on the websites of government departments at all levels. The main fields collected include website name, policy title, issuing agency, document number, release date, written date, text content, attachment link, and other information.

[0030] In some embodiments of the present application, it is necessary to pre-construct multi-dimensional policy tags. Among them, since policies are formulated by government departments at all levels and have typical regional and industrial attributes, some policies have specific application condition requirements. Therefore, it is necessary to tag the collected policy data. For example, the multi-dimensional policy tags include tags such as document classification, release time, release agency's affiliated system, policy level, region, industrial category, application status, and application conditions.

[0031] It should be noted that document classification refers to differentiating the types of policies, such as laws and regulations, application notices, public announcements, and policy documents.

[0032] Release time refers to forming tags such as the release date, release month, release year of the policy based on the release date, as well as derivative tags such as the last week, last month, last three months, and last half year.

[0033] Release agency's affiliated system refers to forming tags of the affiliated system based on the policy release agency, such as science and technology, development and reform, industry and information technology, etc.

[0034] Policy level refers to forming tags such as national level, provincial level, municipal level, district and county level, town and street level, etc. according to the level of the policy release agency.

[0035] Region refers to differentiating regions such as provinces, cities, districts and counties, towns and streets according to the location of the policy release agency.

[0036] Industrial category refers to forming industry tags according to the direction of the policy content, including national economic industry categories, national economic industry major categories, etc., as well as other industrial classification standards.

[0037] Application status refers to forming tags such as in application, application deadline, not yet started, etc. according to the start date and end date of the policy application.

[0038] Application conditions refer to tags such as application region requirements, establishment time requirements, registered capital requirements, industry requirements, business data requirements, financial data requirements, intellectual property requirements, legal and compliance situation requirements, etc.

[0039] In addition, in order to improve the accuracy of policy data, the policy data will be merged and data cleaning will be carried out. For example, in the policy data, duplicate policies published on different government websites will be deduplicated. Missing values in the policy data will be filled. Outliers in the policy data will be processed.

[0040] Based on this, the process of extracting the multi-dimensional policy tag information of each policy in the policy data is as follows:

[0041] First, perform natural language processing on the policy data to obtain the policy semantic text for each policy. Then, vectorize the policy semantic text to obtain the policy text vector. Finally, label the policy text vector according to the preset policy multi-dimensional labels to obtain the policy multi-dimensional label information for each policy.

[0042] S102: Extract enterprise label information from the enterprise data of the enterprise to be matched.

[0043] It should be noted that since policies have typical regional and industrial attributes, that is, an enterprise is more concerned about national, provincial, municipal, and district / county-level policies related to the current main business of the enterprise, and at the same time, some policies have specific application condition requirements. Therefore, an enterprise label system consistent with the policies is established, for example, including region, industrial category, establishment time, registered capital, operating industry, operating data, financial data, intellectual property data, legal compliance status, etc.

[0044] Based on this, the process of extracting enterprise label information is as follows:

[0045] First, according to the registration information of the enterprise to be matched, match the industrial and commercial information of the enterprise. Among them, the registration information includes the enterprise name. Finally, according to the preset enterprise labels, extract the enterprise label information of the enterprise from the industrial and commercial information.

[0046] It should be noted that the user simply registers through the platform, fills in the unit name, and the system automatically matches its unified social credit code, location, registered capital, establishment time, and other industrial and commercial information. The user can confirm or modify the matched information.

[0047] S103: Calculate the matching degree between the enterprise and each policy according to the principal component analysis model for the policy multi-dimensional label information and the enterprise label information.

[0048] Among them, by constructing a principal component analysis (PCA) model, through a dimensionality reduction method, multiple labels are converted into a few principal components, and the weight of each principal component reflects its importance.

[0049] In some embodiments of the present application, the process of calculating the policy multi-dimensional label information and the enterprise label information is as follows:

[0050] First, perform data standardization processing on the policy multi-dimensional label information and the enterprise label information to obtain the standardized policy label value and the standardized enterprise label value.

[0051] It should be noted that, calculate the mean and standard deviation of the multi-dimensional label values of the policies, obtain the difference between each multi-dimensional label value of the policy and the mean, and divide each difference by the standard deviation to obtain the standardized value of each policy multi-dimensional label. Similarly, calculate the mean and standard deviation of the enterprise label values, obtain the difference between each enterprise label value and the mean, and divide each difference by the standard deviation to obtain the standardized value of each enterprise label.

[0052] Then, based on the standardized policy label values and standardized enterprise label values, obtain the standardized data matrix.

[0053] Then, based on the standardized data matrix, obtain the eigenmatrix corresponding to the covariance matrix, and from the eigenmatrix, select the principal components whose cumulative contribution rate exceeds the preset threshold.

[0054] It should be noted that the process of obtaining the eigenmatrix corresponding to the covariance matrix is as follows:

[0055] First, based on the covariance matrix formula and the standardized data matrix, obtain the covariance matrix. Then, calculate the eigenvalues and eigenvectors of the covariance matrix. Then, sort the multiple eigenvalues from largest to smallest, select the eigenvectors corresponding to the top preset number of eigenvalues, select the principal components whose cumulative contribution rate reaches the preset contribution threshold, and obtain the eigenmatrix.

[0056] Among them, the covariance matrix formula is as follows:

[0057]

[0058] C is the covariance matrix, X is the standardized data matrix, and n is the number of samples, that is, the number of rows of the standardized data matrix.

[0059] Then, calculate the load of each label on the principal component to determine the weight of each label on the principal component.

[0060] For example, convert multiple labels into principal components such as region, national economic industry, general industrial classification, enterprise establishment time, registered capital, etc., to form different weights.

[0061] Finally, by calculating the distance between the enterprise and each policy in the principal component space respectively, obtain the matching degree between the enterprise and each policy.

[0062] It should be noted that in the principal component space, we regard the weight value of the enterprise on the principal component as the coordinates of a point, and the weight value of a single policy on the principal component as the coordinates of another point, and calculate the distance between them through the calculation formula of the Euclidean distance.

[0063] S104: Push the target policy with a matching degree higher than the preset matching threshold to the said enterprise.

[0064] In some embodiments of the present application, the pushing process is as follows:

[0065] From the policy multi-dimensional label information of the target policy, determine the initial push label information and generate the original text link of the target policy. Integrate the push label information, the original text link, and the matching degree to generate the push label information of the target policy for pushing the target policy to the enterprise.

[0066] For example, the push fields include information such as policy name, policy release time, policy release agency, matching degree, application status, etc., and support filtering by document category, by time, by region, by level, by centralized management system, by matching degree, by application status. Automatically push policies with a matching degree higher than 50% to the enterprise. The user can click on a certain policy to view details, including detailed information such as the title of the policy, the industry label of the policy, the application status, the application time, the policy text, the original text link, etc.

[0067] In some embodiments of the present application, the user can also perform independent queries, and the process is as follows:

[0068] First, obtain the policy keywords uploaded by the client. Then, according to the multi-condition combination, retrieve the policy keywords from the policy database to obtain the target policy combination. Finally, feedback the policy target combination and the original text link of each policy in the policy target combination to the client.

[0069] For example, provide fuzzy query retrieval by keyword and custom combination retrieval according to the release agency, region, centralized management system, and industry category. The user can click on a certain policy to view details.

[0070] Through Figure 1 the method, realize the high-frequency real-time collection of various policy document data in a big data manner, helping users to timely grasp the policy dynamics and directions. In addition, the collection, processing, and labeling of massive and multi-source policy data support users to perform fuzzy queries or multi-condition custom queries, improving work efficiency. In addition, a policy matching model is established to realize the dual portraits of policies and enterprises, and more accurately push suitable policies to enterprises, reducing their screening efforts and costs.

[0071] It should be noted that although the embodiments of the present application are described with reference to Figure 1 to introduce and explain steps S101 to S104 in sequence, this does not mean that steps S101 to S104 must be executed in a strict order. The reason why the embodiments of the present application are in accordance with Figure 1The order shown in [description] is used to introduce and explain steps S101 to S104 in sequence for the convenience of those skilled in the art to understand the technical solution of the embodiment of the present application. In other words, in the embodiment of the present application, the sequence among steps S101 to S104 can be appropriately adjusted according to actual needs.

[0072] More intuitively, Figure 2 is a schematic framework diagram of an enterprise policy matching system provided for an embodiment of the present application.

[0073] In Figure 2 it includes a data collection module, a data preliminary processing module, a policy label system establishment module, an enterprise label system establishment module, a policy matching model establishment module, a visualization interface and interaction module.

[0074] Among them, the data collection module: comes from government departments at all levels. The collection fields include policy title, release date, release agency, policy text, policy attachments, policy original link, website name, and document number.

[0075] The data preliminary processing module: mainly performs data cleaning, including removing duplicate values, filling or deleting missing values, and handling outliers.

[0076] The policy label system establishment module: performs text preprocessing and text vectorization on policy data, and establishes labels including text classification, release time, centralized system, policy level, region, industrial category, application status, application conditions, etc.

[0077] The enterprise label system establishment module: includes labels such as registered capital, business industry, establishment time, business data, financial data, region, industrial category, intellectual property data, and legal compliance.

[0078] The policy matching model establishment module: through principal component analysis, including processes such as standardization processing, covariance matrix, eigenvalue decomposition, constructing a feature matrix, determining load weights, and forming a matching degree result.

[0079] Regarding the visualization interface and interaction module: includes functions such as enterprise information registration, automatic matching, matching result screening, and automatic push.

[0080] Through Figure 2 the system aims to provide comprehensive and timely policy data services for the massive, scattered, and complex policy document information by collecting policy information publicly disclosed by government departments at all levels and using methods such as text processing, intelligent label algorithms, and intelligent matching models, helping enterprises master policy dynamics, seize policy directions, and promote digital transformation and upgrading.

[0081] Based on the same idea, some embodiments of the present application also provide devices and non-volatile computer storage media corresponding to the above methods.

[0082] Figure 3 The following is a schematic structural diagram of an enterprise policy matching device provided by an embodiment of the present application, including:

[0083] At least one processor; and,

[0084] A memory communicatively connected to the at least one processor; wherein,

[0085] The memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor is enabled to execute an enterprise policy matching method according to any one of the above.

[0086] A non-volatile computer storage medium for enterprise policy matching provided by some embodiments of the present application stores computer-executable instructions that can execute an enterprise policy matching method according to any one of the above.

[0087] The embodiments in the present application are all described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the device and medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and for the relevant parts, reference can be made to the partial description of the method embodiments.

[0088] The devices and media provided by the embodiments of the present application correspond one-to-one with the methods. Therefore, the devices and media also have beneficial technical effects similar to those of the corresponding methods. Since the beneficial technical effects of the methods have been described in detail above, the beneficial technical effects of the devices and media will not be elaborated here.

[0089] Those skilled in the art should understand that the embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0090] The present invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each flow and / or block in the flowchart illustrations and / or block diagrams, and combinations of flows and / or blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions executed by the processor of the computer or other programmable data processing apparatus create means for implementing the functions specified in the flowchart Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0091] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means that implement the functions specified in the flowchart Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0092] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0093] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0094] The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM), and / or non-volatile memory such as read-only memory (ROM) or flash RAM. The memory is an example of computer-readable media.

[0095] Computer-readable media include both permanent and non-permanent, removable and non-removable media and can store information by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transitory medium that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0096] It should also be noted that the term "comprising", "including" or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0097] The above description is only for the embodiments of the present application and is not intended to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the technical principles of the present application shall fall within the protection scope of the present application.

Claims

1. An enterprise policy matching method, characterized in that: The method comprises: Crawl policy data from government websites and extract policy multi-dimensional label information for each policy in the policy data; Extracting enterprise tag information from enterprise data of the enterprise to be matched; According to the principal component analysis model, the policy multi-dimensional label information and the enterprise label information are calculated to obtain the matching degree between the enterprise and each policy; When the matching degree is higher than the preset matching threshold, the target policy is pushed to the enterprise.

2. The method according to claim 1, characterized in that The extracting of the policy multi-dimensional label information of each policy in the policy data specifically includes: Performing natural language processing on the policy data to obtain a policy semantic text of each policy; Vectorizing the policy semantic text to obtain a policy text vector; According to the preset policy multi-dimensional labels, the policy text vector is labeled to obtain the policy multi-dimensional label information of each policy.

3. The method according to claim 2, characterized in that The calculation of the policy multi-dimensional label information and the enterprise label information based on the principal component analysis model to obtain the matching degree between the enterprise and each policy specifically includes: Performing data standardization processing on the policy multi-dimensional label information and the enterprise label information to obtain a standardized policy label value and a standardized enterprise label value; Obtaining a standardized data matrix according to the standardized policy tag value and the standardized enterprise tag value; According to the standardized data matrix, a characteristic matrix corresponding to the covariance matrix is ​​obtained, and from the characteristic matrix, a principal component whose cumulative contribution rate exceeds a preset threshold is selected; Calculate the load of each label on the principal component and determine the weight of each label on the principal component; By respectively calculating the distance between the enterprise and each policy in the principal component space, the matching degree between the enterprise and each policy is obtained.

4. The method according to claim 3, characterized in that The step of obtaining a characteristic matrix corresponding to the covariance matrix according to the standardized data matrix specifically includes: According to the covariance matrix formula and the standardized data matrix, a covariance matrix is ​​obtained; Calculating the eigenvalues ​​and eigenvectors of the covariance matrix; Sort multiple eigenvalues ​​from large to small, select the eigenvectors corresponding to the first preset number of eigenvalues, select the principal component whose cumulative contribution rate reaches the preset contribution threshold, and obtain the characteristic matrix.

5. The method according to claim 1, characterized in that The pushing of the target policy to the enterprise when the matching degree is higher than the preset matching threshold specifically includes: Determine initial push tag information from the policy multi-dimensional tag information of the target policy, and generate an original text link of the target policy; The push tag information, the original text link and the matching degree are integrated to generate push tag information of the target policy, so as to push the target policy to the enterprise.

6. The method according to claim 1, characterized in that The method further comprises: Get the policy keywords uploaded by the client; According to the combination of multiple conditions, the policy keywords are retrieved from the policy database to obtain the target policy combination, and the policy target combination and the original link of each policy in the policy target combination are fed back to the client.

7. The method according to claim 1, characterized in that The step of extracting enterprise tag information from the enterprise data of the enterprise to be matched specifically includes: Match the business information of the enterprise according to the registration information of the enterprise to be matched; the registration information includes the name of the enterprise; According to the preset enterprise tag, the enterprise tag information of the enterprise is extracted from the industrial and commercial information.

8. The method according to claim 1, characterized in that: Before extracting the policy multi-dimensional label information of each policy in the policy data, the method further includes: In the policy data, duplicate policies published on different government websites are removed; Filling missing values ​​in the policy data; The policy data is processed for outliers.

9. An enterprise policy matching device, characterized in that: include: at least one processor; as well as, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the enterprise policy matching method described in any one of claims 1 to 8.

10. An enterprise policy matching non-volatile computer storage medium storing computer executable instructions, characterized in that: The computer executable instructions can execute an enterprise policy matching method as described in any one of claims 1-8.

Images