Innovation service policy intelligent matching decision-making system and method based on big data

Abstract

The invention discloses an innovation service policy intelligent matching decision-making system and method based on big data, and belongs to the technical field of policy matching services, and the system comprises an information collection module, a policy matching module and a matching analysis module. The information collection module is used for collecting policy information and enterprise information and establishing a corresponding policy information base and an enterprise information base; the policy matching module is used for performing policy matching, identifying policy information in the policy information base and enterprise information in the enterprise information base in real time, and obtaining policy labels and enterprise feature labels corresponding to the policy information and the enterprise information; combining the policy labels and the enterprise feature labels in pairs to obtain matching evaluation combinations; evaluating each matching evaluation combination, and determining a matching enterprise corresponding to each policy information; setting a corresponding enterprise matching list according to each matching enterprise corresponding to the policy information; and the matching analysis module is used for performing decision analysis on the enterprise matching list corresponding to the policy information.

Description

Technical Field

[0001] The present invention belongs to the technical field of policy matching services, and specifically relates to an innovative service policy intelligent matching decision system and method based on big data. Background Art

[0002] With the development of the economy, there are a series of subsidies and tax incentives to support the development of enterprises. However, these policies usually contain complex conditions and requirements, and enterprises must meet specific standards to enjoy them. Due to the large differences in the actual conditions of different enterprises, business owners often find it difficult to quickly and accurately judge whether they meet the policy requirements, thus missing out on policy benefits; in particular, many small and medium-sized enterprises are not aware of the corresponding policies; the above phenomenon leads to the failure to give full play to the original intention of the corresponding policies; based on this, in order to solve the above problems, the present invention provides an innovative service policy intelligent matching decision system and method based on big data. Summary of the invention

[0003] In order to solve the problems existing in the above-mentioned solutions, the present invention provides an innovative service policy intelligent matching decision system and method based on big data.

[0004] The purpose of the present invention can be achieved through the following technical solutions:

[0005] An innovative service policy intelligent matching decision-making system based on big data, including information collection module, policy matching module and matching analysis module;

[0006] The information collection module is used to collect policy information and enterprise information, and to establish corresponding policy information database and enterprise information database; the policy information database is used to store various policy information for enterprises; and the enterprise information database is used to store enterprise information of various enterprises.

[0007] Furthermore, the method for obtaining policy information and enterprise information is as follows:

[0008] Determine the service scope, and determine each policy source according to the service scope; connect with each policy source, and obtain the release information of each policy source in real time; filter the release information and determine each policy information;

[0009] Extract features of policy information, obtain policy features corresponding to each policy information, and set policy labels for corresponding policy information according to each policy feature;

[0010] Collect enterprise information according to the service scope, process the enterprise information through the preset enterprise characteristic template, and obtain the enterprise characteristic label corresponding to each enterprise information.

[0011] Furthermore, the method for establishing the policy information database includes:

[0012] Establish a database, send the collected policy information to the database for storage, and mark the current database as a policy information database;

[0013] Identify the stored policy information and corresponding policy labels in real time, make storage judgments on the policy information according to the policy labels, and filter the stored policy information according to the storage judgment results;

[0014] Identify the policy information that needs to be stored and the corresponding policy tags, perform storage evaluation based on the identified policy information and policy tags, and obtain corresponding storage evaluation results, wherein the storage evaluation results include allowed storage and existing storage; and perform corresponding processing on the policy information that needs to be stored based on the storage evaluation results.

[0015] Furthermore, the method for storing and judging the policy information according to the policy tag includes:

[0016] A storage judgment model is established. The expression of the storage judgment model is:

[0017]

[0018] Where: s is the input data, the input data is the policy label; the output data stores the judgment value CP(s);

[0019] The storage judgment model is used to perform storage judgment on each stored policy information, and the storage judgment value corresponding to each policy information is obtained, and the policy information with a storage judgment value of 1 is eliminated.

[0020] The policy matching module is used to perform policy matching, identify the policy information in the policy information database and the enterprise information in the enterprise information database in real time, obtain the policy labels and enterprise feature labels corresponding to each policy information and enterprise information; combine each policy label and enterprise feature label in pairs to obtain each matching evaluation combination; evaluate each matching evaluation combination to determine the matching enterprise corresponding to each policy information;

[0021] A corresponding enterprise matching list is set according to each matching enterprise corresponding to the policy information, and the enterprise matching list includes a conflict column and a matching column.

[0022] Furthermore, the method for evaluating each matching evaluation combination includes:

[0023] Establish a matching model, the expression of the matching model is:

[0024]

[0025] Where: q is the input data, the input data is the matching evaluation combination; the output data is the single matching value PE(q) corresponding to the corresponding policy feature;

[0026] The matching evaluation combination is evaluated through the matching model to obtain the individual matching value corresponding to each policy feature; the enterprise information with each individual matching value of 1 is marked as a matching enterprise.

[0027] Furthermore, the enterprise matching list also includes a cultivation column, which is used to count the cultivation enterprises that meet the corresponding policy tags after cultivation.

[0028] Furthermore, the methods for determining the incubation enterprises include:

[0029] Determine each screening enterprise according to the conflict column and the matching column, obtain each single matching value corresponding to the screening enterprise, and determine the influencing characteristics according to each single matching value; set the difference overcoming data according to the influencing characteristics; identify the cultivation time in real time;

[0030] Establish a cultivation judgment model, analyze the enterprise information, cultivation time and difference overcoming data of the screening enterprises through the cultivation judgment model, and obtain the single cultivation value corresponding to each difference overcoming data; mark the obtained single cultivation value as PYi, i represents the difference overcoming data corresponding to the screening enterprise, i=1, 2, ..., n, n is a positive integer;

[0031] According to the formula Calculate the corresponding enterprise cultivation value; where: PK is the enterprise cultivation value;

[0032] Generate a corresponding enterprise cultivation curve according to the enterprise cultivation value corresponding to each time, with the horizontal axis being time and the vertical axis being the enterprise cultivation value; set a cultivation positioning line in the enterprise cultivation curve with the enterprise cultivation value being 1;

[0033] According to the current enterprise cultivation curve and cultivation positioning line, dynamically evaluate whether the corresponding screened enterprises are cultivation enterprises.

[0034] The matching analysis module is used to perform decision analysis on the enterprise matching list corresponding to each policy information, identify each matching enterprise according to the enterprise matching list, and determine the matching registered enterprise according to whether the matching enterprise is registered;

[0035] Obtain the enterprise update information corresponding to the matching registered enterprises, adjust and update the enterprise matching list according to the enterprise update information, and provide policy assistance and guidance based on the updated enterprise matching list.

[0036] Furthermore, when the enterprise matching list includes a cultivation column, each cultivation enterprise is identified, policy cultivation information is sent to each cultivation enterprise, and cultivation processing is performed according to the feedback results of the cultivation enterprise.

[0037] An innovative service policy intelligent matching decision-making method based on big data, including:

[0038] Collect policy information and enterprise information, and establish corresponding policy information database and enterprise information database;

[0039] Obtain policy labels and enterprise characteristic labels corresponding to each policy information and enterprise information; combine each policy label and enterprise characteristic label in pairs to obtain each matching evaluation combination; evaluate each matching evaluation combination to determine the matching enterprise corresponding to each policy information; set the corresponding enterprise matching list according to each matching enterprise corresponding to the policy information;

[0040] Determine the matching registered enterprises based on whether the matching enterprises are registered; obtain the enterprise update information corresponding to the matching registered enterprises, adjust and update the enterprise matching list based on the enterprise update information, and provide policy assistance guidance based on the updated enterprise matching list.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] Through the mutual cooperation between the information collection module, policy matching module and matching analysis module, intelligent policy matching is achieved, which solves the problem that enterprises find it difficult to quickly and accurately judge whether they meet policy requirements and thus miss out on policy benefits; by setting up an enterprise cultivation curve and dynamically updating it, the cultivation enterprises can be dynamically determined according to changes in enterprise information and the advancement of time, thereby improving work efficiency, avoiding the cultivation of screened enterprises that are no longer cultivation enterprises, and avoiding ineffective work. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 This is a principle block diagram of the present invention. DETAILED DESCRIPTION

[0045] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] like Figure 1 As shown, the intelligent matching decision-making system for innovative service policies based on big data includes an information collection module, a policy matching module and a matching analysis module;

[0047] The information collection module is used to collect policy information and enterprise information, and to establish corresponding policy information database and enterprise information database; the policy information database is used to store policy information for enterprises; and the enterprise information database is used to store enterprise information of each enterprise.

[0048] Among them, the method of obtaining policy information and enterprise information is as follows:

[0049] Determine the service scope, that is, for which areas the platform provides policy matching services; determine the policy sources according to the service scope, and the policy sources are the sources of the release of each policy; connect with each policy source to obtain the corresponding release information from each policy source in real time; screen the collected release information and determine each policy information, that is, determine whether it is policy information for enterprises that meets the preset requirements of the platform based on the content of the release information. The preset requirements are set according to the actual situation of the enterprises that the platform needs to serve, such as subsidy scale, enterprise nature, etc.; extract features from the policy information to obtain the policy features corresponding to each policy information, and the policy features are the requirements and time features of the policy for the enterprise; the existing data recognition technology can be used to realize the screening and feature extraction of the release information; set the policy tags corresponding to the corresponding policy information according to each policy feature, and identify the policy features corresponding to the policy information through the policy tags;

[0050] According to the service scope, each enterprise can be identified based on the existing information platforms, and the enterprise information of each enterprise can be collected. The characteristics of each enterprise can be extracted. The extraction of enterprise characteristics is set based on a preset enterprise characteristic template. The enterprise characteristic template determines the enterprise characteristics that may be needed according to the policy characteristics that each policy information may have, and then combines and sets the enterprise characteristic template; sets the corresponding enterprise characteristic label according to the characteristics of each enterprise; and makes corresponding updates and adjustments according to the changes in enterprise information.

[0051] The methods for establishing a policy information database include:

[0052] Establish a database, send the collected policy information to the database for storage, and mark the current database as a policy information database;

[0053] Identify the policy labels corresponding to each stored policy information in real time, and make storage judgments on the policy information based on the policy labels. Mainly judge whether the policy information is invalid based on the characteristics such as the restriction time of the policy information. Specifically, a corresponding storage judgment model can be established. The expression of the storage judgment model is: Where: s is the input data, the input data is the policy label; the output data stores the judgment value CP(s);

[0054] Perform storage judgment on each stored policy information through the storage judgment model, obtain the storage judgment value corresponding to each policy information, and filter the stored policy information according to the storage judgment value; that is, eliminate the policy information with a storage judgment value of 1;

[0055] Identify the policy information that needs to be stored and the corresponding policy tags, perform storage evaluation based on the identified policy information and policy tags, analyze whether it has been stored, and obtain the corresponding storage evaluation results, which include allowed storage and existing storage; perform corresponding processing on the policy information that needs to be stored based on the storage evaluation results.

[0056] The policy matching module is used to perform policy matching, identify the policy information in the policy information database and the enterprise information in the enterprise information database in real time, obtain the policy labels and enterprise feature labels corresponding to each policy information and enterprise information; combine each policy label and enterprise feature label in pairs to obtain each matching evaluation combination, that is, a policy label and an enterprise feature label form a matching evaluation combination; evaluate each matching evaluation combination to determine whether the enterprise feature label conforms to the policy label; obtain the single matching value corresponding to each policy feature in the policy label, and the single matching value includes 1 or 0; specifically, a matching model can be established, and the expression of the matching model is In the formula: q is the input data, the input data is the matching evaluation combination; the output data is the single matching value PE(q) corresponding to the corresponding policy feature.

[0057] The enterprise information whose single matching values ​​corresponding to the policy label are all 1 is marked as a matching enterprise; the corresponding enterprise matching list is set according to the matching enterprises corresponding to the policy information; and the matching list of each enterprise is updated according to the mutual exclusivity of each policy information, that is, the policy information that cannot be enjoyed at the same time is retained for the greatest help to the enterprise, and other conflicts are marked as conflicts in the corresponding enterprise matching list. The policy information with conflict marks does not need to be pushed to the enterprise, or if the enterprise does not apply for the policy information that helps the most, it will be adjusted synchronously. Then, the enterprise matching list includes the conflict column corresponding to the conflicting part and the matching column that meets the matching requirements; the enterprise corresponding to the matching column is the matching enterprise, and the enterprise in the conflict column is the conflicting enterprise.

[0058] In one embodiment, because in actual applications, many enterprises do not comply with the policy under the current circumstances, but will comply with the corresponding policy after a certain amount of development and cultivation, in order to develop the enterprises, a cultivation column is added to the enterprise matching list. The cultivation column is used to count the information of each enterprise that can meet the policy label requirements after cultivation, and mark them as cultivated enterprises.

[0059] Methods for determining the incubation enterprises include:

[0060] Identify the information of each enterprise that fails to match, mark it as a screening enterprise, obtain the individual matching values ​​corresponding to the screening enterprise, determine the policy characteristics of the unqualified match based on the individual matching values, and mark them as influencing characteristics; determine the data that needs to be improved or changed for the screening enterprise to overcome the problem based on the influencing characteristics, and mark them as difference overcoming data, such as turnover, tax returns, number of employees, etc.; determine the remaining time based on the time limit in the policy label, and mark it as cultivation time;

[0061] A corresponding cultivation judgment model is established based on a neural network such as a CNN network or a DNN network. The cultivation judgment model is used to analyze the difference overcoming data according to the enterprise information and the cultivation time, and to judge whether the screened enterprise can overcome the problem under the existing conditions; specifically, a training set is established manually for training, and the training set includes input data and output data. The input data is the enterprise information, the cultivation time, and the difference overcoming data; the output data is a single cultivation value of 1 or 0; in the actual application process, a large amount of enterprise development and change data can be obtained based on the platform's data accumulation and big data technology, and then the data of enterprises meeting the corresponding policy characteristics under different circumstances can be summarized and integrated into the training set; the single cultivation value corresponding to each difference overcoming data is obtained through analysis by the cultivation judgment model after successful training; the obtained single cultivation value is marked as PYi, i represents the difference overcoming data corresponding to the screened enterprise, i=1, 2,..., n, and n is a positive integer;

[0062] According to the formula Calculate the corresponding enterprise cultivation value; where: PK is the enterprise cultivation value;

[0063] Generate a corresponding enterprise cultivation curve according to the enterprise cultivation value corresponding to each time, where the horizontal axis is time, i.e. the time for evaluating the cultivation value; the vertical axis is the enterprise cultivation value; a cultivation positioning line is set in the enterprise cultivation curve with the enterprise cultivation value of 1;

[0064] Identify the enterprise cultivation curve of each screened enterprise in real time. When the enterprise cultivation curve intersects with the cultivation positioning line, that is, the enterprise cultivation value is 1, and the screened enterprise is marked as a cultivation enterprise; when the enterprise cultivation curve does not intersect with the cultivation positioning line, that is, the enterprise cultivation value is less than 1, the corresponding cultivation enterprise is cancelled. If it is not a cultivation enterprise, no corresponding operation is performed.

[0065] It is possible to dynamically determine each cultivation enterprise according to changes in enterprise information and the advancement of time, improve work efficiency, avoid cultivating selected enterprises that are no longer cultivation enterprises, and avoid ineffective work.

[0066] The matching analysis module is used to perform decision analysis on the enterprise matching list corresponding to each policy information, identify each matching enterprise according to the enterprise matching list, and identify whether each matching enterprise is registered, that is, whether it is registered in this system, so as to determine its current real enterprise information; send registration push information to unregistered matching enterprises, including corresponding policy information, registration benefits, etc., and the specific information that the registration push information should include is set by the platform; matching enterprises register according to the registration push information; mark the registered matching enterprises as matching registered enterprises, and obtain the enterprise update information corresponding to the matching registered enterprises. The enterprise update information is the real enterprise information uploaded by the enterprise itself; adjust and update the enterprise matching list according to the obtained enterprise update information, that is, re-analyze and confirm according to the enterprise update information, and make corresponding adjustments according to the analysis results; and provide policy auxiliary guidance according to the updated enterprise matching list.

[0067] In one embodiment, when the enterprise matching list includes a cultivation column, each cultivation enterprise is identified and policy cultivation information is sent to each cultivation enterprise. The policy cultivation information includes various relevant information for describing the corresponding cultivation enterprise, and cultivation processing is performed based on the feedback results of the cultivation enterprise.

[0068] An innovative service policy intelligent matching decision-making method based on big data, including:

[0069] Collect policy information and enterprise information, and establish corresponding policy information database and enterprise information database;

[0070] Obtain policy labels and enterprise characteristic labels corresponding to each policy information and enterprise information; combine each policy label and enterprise characteristic label in pairs to obtain each matching evaluation combination; evaluate each matching evaluation combination to determine the matching enterprise corresponding to each policy information; set the corresponding enterprise matching list according to each matching enterprise corresponding to the policy information;

[0071] Determine the matching registered enterprises based on whether the matching enterprises are registered; obtain the enterprise update information corresponding to the matching registered enterprises, adjust and update the enterprise matching list based on the enterprise update information, and provide policy assistance guidance based on the updated enterprise matching list.

[0072] The above formulas are all calculated by removing dimensions and taking their numerical values. The formula is a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained by simulating a large amount of data.

[0073] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An innovative service policy intelligent matching decision-making system based on big data, characterized by: It includes information collection module, policy matching module and matching analysis module; The information collection module is used to collect policy information and enterprise information, and establish corresponding policy information database and enterprise information database; the policy information database is used to store various policy information for enterprises; The enterprise information database is used to store the enterprise information of each enterprise; The policy matching module is used to perform policy matching, identify the policy information in the policy information database and the enterprise information in the enterprise information database in real time, and obtain the policy tags and enterprise feature tags corresponding to each policy information and enterprise information; Combine each policy label and enterprise characteristic label in pairs to obtain each matching evaluation combination; evaluate each matching evaluation combination to determine the matching enterprise corresponding to each policy information; According to each matching enterprise corresponding to the policy information, a corresponding enterprise matching list is set, and the enterprise matching list includes a conflict column and a matching column; The matching analysis module is used to perform decision analysis on the enterprise matching list corresponding to each policy information, identify each matching enterprise according to the enterprise matching list, and determine the matching registered enterprise according to whether the matching enterprise is registered; Obtain the enterprise update information corresponding to the matching registered enterprises, adjust and update the enterprise matching list according to the enterprise update information, and provide policy assistance and guidance based on the updated enterprise matching list.

2. According to the big data-based innovative service policy intelligent matching decision system of claim 1, it is characterized by: The methods for obtaining policy information and enterprise information are as follows: Determine the service scope, and determine each policy source according to the service scope; connect with each policy source, and obtain the release information of each policy source in real time; filter the release information and determine each policy information; Extract features of policy information, obtain policy features corresponding to each policy information, and set policy labels for corresponding policy information according to each policy feature; Collect enterprise information according to the service scope, process the enterprise information through the preset enterprise characteristic template, and obtain the enterprise characteristic label corresponding to each enterprise information.

3. According to the big data-based innovative service policy intelligent matching decision system of claim 1, it is characterized by: The methods for establishing a policy information database include: Establish a database, send the collected policy information to the database for storage, and mark the current database as a policy information database; Identify the stored policy information and corresponding policy labels in real time, make storage judgments on the policy information according to the policy labels, and filter the stored policy information according to the storage judgment results; Identify the policy information that needs to be stored and the corresponding policy tags, perform storage evaluation based on the identified policy information and policy tags, and obtain corresponding storage evaluation results, wherein the storage evaluation results include allowed storage and existing storage; and perform corresponding processing on the policy information that needs to be stored based on the storage evaluation results.

4. According to claim 3, the innovative service policy intelligent matching decision system based on big data is characterized in that: Methods for storing and judging policy information based on policy tags include: A storage judgment model is established. The expression of the storage judgment model is: Where: s is the input data, the input data is the policy label; the output data stores the judgment value CP(s); The storage judgment model is used to perform storage judgment on each stored policy information, and the storage judgment value corresponding to each policy information is obtained, and the policy information with a storage judgment value of 1 is eliminated.

5. According to the big data-based innovative service policy intelligent matching decision system of claim 1, it is characterized by: Methods for evaluating each matching evaluation combination include: Establish a matching model, the expression of the matching model is: Where: q is the input data, the input data is the matching evaluation combination; the output data is the single matching value PE(q) corresponding to the corresponding policy feature; The matching evaluation combination is evaluated through the matching model to obtain the individual matching value corresponding to each policy feature; the enterprise information with each individual matching value of 1 is marked as a matching enterprise.

6. The innovative service policy intelligent matching decision system based on big data according to claim 1 is characterized in that: The enterprise matching list also includes a cultivation column, which is used to count the cultivation enterprises that meet the corresponding policy tags after cultivation.

7. The innovative service policy intelligent matching decision system based on big data according to claim 6 is characterized in that: Methods for determining the incubation enterprises include: Determine each screening enterprise according to the conflict column and the matching column, obtain each single matching value corresponding to the screening enterprise, and determine the influencing characteristics according to each single matching value; set the difference overcoming data according to the influencing characteristics; identify the cultivation time in real time; Establish a cultivation judgment model, analyze the enterprise information, cultivation time and difference overcoming data of the screening enterprises through the cultivation judgment model, and obtain the single cultivation value corresponding to each difference overcoming data; mark the obtained single cultivation value as PYi, i represents the difference overcoming data corresponding to the screening enterprise, i=1, 2, ..., n, n is a positive integer; According to the formula Calculate the corresponding enterprise cultivation value; where: PK is the enterprise cultivation value; Generate a corresponding enterprise cultivation curve according to the enterprise cultivation value corresponding to each time, with the horizontal axis being time and the vertical axis being the enterprise cultivation value; set a cultivation positioning line in the enterprise cultivation curve with the enterprise cultivation value being 1; According to the current enterprise cultivation curve and cultivation positioning line, dynamically evaluate whether the corresponding screened enterprises are cultivation enterprises.

8. The innovative service policy intelligent matching decision system based on big data according to claim 1 is characterized in that: When the enterprise matching list includes a cultivation column, each cultivation enterprise is identified, policy cultivation information is sent to each cultivation enterprise, and cultivation processing is performed based on the feedback results of the cultivation enterprise.

9. An innovative service policy intelligent matching decision method based on big data, characterized in that: Applied to the intelligent matching decision system for innovative service policies based on big data as described in any one of claims 1 to 8, the method comprises: Collect policy information and enterprise information, and establish corresponding policy information database and enterprise information database; Obtain policy labels and enterprise characteristic labels corresponding to each policy information and enterprise information; combine each policy label and enterprise characteristic label in pairs to obtain each matching evaluation combination; evaluate each matching evaluation combination to determine the matching enterprise corresponding to each policy information; set the corresponding enterprise matching list according to each matching enterprise corresponding to the policy information; Determine the matching registered enterprises based on whether the matching enterprises are registered; obtain the enterprise update information corresponding to the matching registered enterprises, adjust and update the enterprise matching list based on the enterprise update information, and provide policy assistance guidance based on the updated enterprise matching list.

Images