Enterprise policy automatic pushing method based on big data

Abstract

The present application relates to the field of information communication technology, disclose a kind of enterprise policy automatic push method based on big data, comprising the following steps: S1: set the policy label of this push, push level and tracking level by policy setting module;S2: policy label, push level, policy text content and enterprise information are analyzed by analysis module, obtain policy matching index;S3: policy matching index is analyzed by analysis module, and the enterprise of this push is filtered out and is pushed;S4: the tracking information data of each enterprise after this push is obtained by tracking module;S5: enterprise matching level, tracking level and tracking information data are analyzed by analysis module, and whether this push is up to standard according to the analysis result is judged, the push effect after evaluating policy information release is realized;And the policy of push can be accurately sent to relevant enterprise.

Description

Technical Field

[0001] This invention relates to the field of information and communication technology, and specifically to a method for automatically pushing enterprise policies based on big data. Background Technology

[0002] With the gradual maturation of mobile internet technology and mobile devices, electronic office work has gradually replaced paper-based office work, and government departments are also releasing policy information through government websites.

[0003] Currently, government departments release policy information on government websites, which businesses cannot directly access. They must visit the government website to view the relevant information. Due to the large amount of information on government websites and limited homepage space, the homepage usually only displays a few recently updated pieces of information. Therefore, businesses can only find relevant information when they actively search for it.

[0004] However, businesses have many operations and it is difficult for them to log in to government websites to keep up with policy information at all times. As a result, businesses may miss a lot of policy information that is relevant to their business. At the same time, after government departments release policy information, staff members are unable to assess the effectiveness of the push notifications. Summary of the Invention

[0005] The purpose of this invention is to provide a method for automatically pushing enterprise policies based on big data, and to solve the following technical problems:

[0006] How to evaluate the effectiveness of policy information dissemination after its release.

[0007] The objective of this invention can be achieved through the following technical solutions:

[0008] A method for automatically pushing enterprise policies based on big data, the method comprising the following steps:

[0009] S1: Configure the policy tag, push level, and tracking level for this push notification through the policy settings module;

[0010] S2: The analysis module analyzes policy tags, push levels, policy text content, and enterprise information to obtain a policy matching index;

[0011] S3: Analyze the policy matching index through the analysis module, select the enterprises to be pushed in this round, and push them to them;

[0012] S4: Obtain tracking information data for each enterprise after this push through the tracking module;

[0013] S5: The analysis module analyzes the enterprise matching level, tracking level, and tracking information data, and determines whether the push notification meets the standards based on the analysis results.

[0014] As a further solution of the present invention: The policy matching index is calculated by the formula:

[0015]

[0016] Calculate the comprehensive policy matching index T of the i-th enterprise and the policy pushed this time i ;

[0017] where, i is the enterprise number; P i is the actual label matching degree of the i-th enterprise and the policy pushed this time; P0 is the preset label matching degree; Q i is the keyword matching degree of the i-th enterprise and the policy pushed this time; P0 is the preset keyword matching degree; T0 is the basic policy matching index; γ1 is the first weight coefficient; γ2 is the second weight coefficient; C1 is the first preset constant; C2 is the second preset constant; u n is the push level coefficient of the n-th level.

[0018] As a further solution of the present invention: The push levels include N levels, N≥2; n∈[1, N]; The push level coefficient u of the n-th level n satisfies 0 = u1P0 < u2P0…u n P0 < u N P0 ≤ 1, and at the same time satisfies 0 = u (1) Q0 < u (2) Q0…u n Q0 < u N Q0 ≤ 1.

[0019] As a further solution of the present invention: In step S3, the process of screening enterprises for this push is as follows:

[0020] Compare the comprehensive policy matching index T of the i-th enterprise and the policy pushed this time i with the preset value R1;

[0021] When T i < T0, the matching level of the i-th enterprise is level one and no push is made;

[0022] When T0 ≤ T i < R1, the matching level of the i-th enterprise is level two and a push is made;

[0023] When T i ≤ R1, the matching level of the i-th enterprise is level three and a push is made.

[0024] As a further solution of the present invention: The tracking levels include multiple levels; In step S4, the tracking information data includes the reading status and the download status; The reading status includes read and unread; The download status includes downloaded and not downloaded.

[0025] As a further aspect of the present invention: In step S5, the process of determining whether the current push meets the criteria includes using the following formula:

[0026]

[0027] Calculate the judgment index C.

[0028] As a further aspect of the present invention, the process for determining whether the current push meets the criteria further includes: when C≥1, the current push meets the criteria; otherwise, the current push does not meet the criteria.

[0029] As a further aspect of the present invention: the actual tag matching degree P between the i-th enterprise and the current push policy i Through formula Calculated;

[0030] The actual keyword matching degree Q between the i-th enterprise and the policy pushed in this notification is... i Through formula Calculated;

[0031] Among them, X A Xs represents the total number of policy tags; Xs represents the number of tag matches between the i-th enterprise and the policy tags pushed in this policy push; Y represents the total number of policy tags. A Xs represents the total number of keywords in this push notification; Xs represents the number of keyword matches between the i-th enterprise and the keywords in this push notification.

[0032] The beneficial effects of this invention are:

[0033] This invention sets the policy tags, push level, and tracking level for this push notification through a policy setting module; analyzes the policy tags, push level, policy text content, and enterprise information through an analysis module to obtain a policy matching index; analyzes the policy matching index through the analysis module to select enterprises for this push notification and push them accordingly; obtains tracking information data for each enterprise after this push notification through a tracking module; and analyzes the enterprise matching level, tracking level, and tracking information data through the analysis module to determine whether the push notification meets the standards based on the analysis results. This allows for the evaluation of the push effect after the policy information is released and ensures that the pushed policies are accurately delivered to relevant enterprises. Attached Figure Description

[0034] The invention will now be further described with reference to the accompanying drawings.

[0035] Figure 1 This is a flowchart of a method according to an embodiment of the present invention. Detailed Implementation

[0036] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0037] Please see Figure 1 As shown, in one embodiment, a method for automatically pushing enterprise policies based on big data is provided, the method comprising the following steps:

[0038] S1: Configure the policy tag, push level, and tracking level for this push notification through the policy settings module;

[0039] S2: The analysis module analyzes policy tags, push levels, policy text content, and enterprise information to obtain a policy matching index;

[0040] S3: Analyze the policy matching index through the analysis module, select the enterprises to be pushed in this round, and push them to them;

[0041] S4: Obtain tracking information data for each enterprise after this push through the tracking module;

[0042] S5: The analysis module analyzes the enterprise matching level, tracking level, and tracking information data, and determines whether the push notification meets the standards based on the analysis results;

[0043] Through the above technical solution, this embodiment sets the policy tag, push level, and tracking level for this push through the policy setting module; analyzes the policy tag, push level, policy text content, and enterprise information through the analysis module to obtain a policy matching index; analyzes the policy matching index through the analysis module to select enterprises for this push and push them; obtains the tracking information data of each enterprise after this push through the tracking module; analyzes the enterprise matching level, tracking level, and tracking information data through the analysis module, and judges whether this push meets the standards based on the analysis results; thus, it achieves the evaluation of the push effect after the release of policy information; and ensures that the pushed policies can be accurately sent to relevant enterprises.

[0044] As one embodiment of the present invention, the policy matching index is obtained by formula:

[0045]

[0046] Calculate the comprehensive policy matching index T between the i-th enterprise and the policy pushed this time. i ;

[0047] Where i is the company number; Pi Pi represents the actual tag matching degree between the i-th enterprise and the policy pushed in this notification; P0 represents the preset tag matching degree; Qi represents the actual tag matching degree between the i-th enterprise and the policy pushed in this notification ... P0 represents the preset tag matching degree; Qi represents the actual tag matching i The keyword matching degree between the i-th enterprise and the policy pushed in this notification; P0 is the preset keyword matching degree; T0 is the basic policy matching index; γ1 is the first weight coefficient; γ2 is the second weight coefficient; C1 is the first preset constant; C2 is the second preset constant; u n This represents the push notification level coefficient for the nth level.

[0048] Through the above technical solution, this embodiment u n P0 represents the preset tag matching degree of the policy text type pushed this time; P i -u n P0 is the difference between the actual tag matching degree of the i-th enterprise and the preset tag matching degree of this push policy; when P i -u n When P0>0, it indicates that the actual tag matching degree is higher than the preset tag matching degree. Therefore, the tag matching degree difference P between the actual tag matching degree and the preset tag matching degree of the i-th enterprise and the current push policy is calculated. i -u n The larger P0 is, the higher the comprehensive policy matching index T between the i-th enterprise and the policy pushed this time. i The larger P is; i -u n When P0 < 0, it indicates that the actual tag matching degree is lower than the preset tag matching degree for this push. Therefore, the tag matching degree difference P between the actual tag matching degree and the preset tag matching degree for the i-th enterprise and the current push policy is calculated. i -u n The larger the absolute value of P0, the higher the comprehensive policy matching index T between the i-th enterprise and the policy pushed this time. i The smaller; u n Q0 represents the preset keyword matching score for the policy text type pushed this time; Q i -u n Q0 represents the difference between the actual keyword matching degree of the i-th enterprise and the keyword matching degree of this push policy, and the preset keyword matching degree; when Q i -u n When Q0>0, it indicates that the actual keyword matching degree is higher than the preset keyword matching degree. Therefore, the difference Q between the actual keyword matching degree and the preset keyword matching degree between the i-th enterprise and the current policy is the keyword matching degree. i -u n The larger Q0 is, the higher the comprehensive policy matching index T between the i-th enterprise and the policy pushed this time. i The larger Q is; i -u nWhen Q0 < 0, it indicates that the actual keyword matching degree is lower than the preset keyword matching degree for this time. Therefore, the keyword matching degree difference Q between the actual keyword matching degree of the i-th enterprise and the preset keyword matching degree of this pushed policy i -u n The larger the absolute value of Q0, the smaller the comprehensive policy matching index T of the i-th enterprise and this pushed policy i is;

[0049] It should be noted that the preset label matching degree P0, the preset keyword matching degree P0, the basic policy matching index T0, the first weight coefficient γ1, the second weight coefficient γ2, the first preset constant C1, and the second preset constant C2 are preset values and are obtained based on experience, which will not be elaborated here.

[0050] As an implementation manner of the present invention, the push level includes N levels, N≥2; n ∈ [1, N]; the push level coefficient u of the n-th level n satisfies 0 = u1P0 < u2P0…u n P0 < u N P0 ≤ 1, and at the same time satisfies 0 = u1Q0 < u2Q0…u n Q0 < u N Q0 ≤ 1;

[0051] Through the above technical solution, in this embodiment, by setting the push level, the staff can set the push accuracy by setting the push level, thereby adjusting the push range; the first-level push level is less than the N-level push level; the smaller the push level, the lower the push accuracy and the larger the push range; the larger the push level, the higher the push accuracy and the smaller the push range;

[0052] It should be noted that the push level coefficient u of the n-th level n is a preset value and is obtained based on experience, which will not be elaborated here.

[0053] As an implementation manner of the present invention, in step S3, the process of screening enterprises for this push is as follows:

[0054] Compare the comprehensive policy matching index T of the i-th enterprise and this pushed policy i with the preset value R1;

[0055] When T i < T0, the matching level of the i-th enterprise is the first level and no push is performed;

[0056] When T0 ≤ T i < R1, the matching level of the i-th enterprise is the second level and a push is performed;

[0057] When T iWhen it is ≤ R1, the matching level of the i-th enterprise is level three, and a push is made.

[0058] Through the above technical solution, the comprehensive policy matching index T of the i-th enterprise in this embodiment and the push policy this time i is compared with the preset value R1; T0 < R1; when T i < T0, the matching level of the i-th enterprise is level one, and the matching degree is low, so no push is made; when T0 ≤ T i < R1, the matching level of the i-th enterprise is level two, and the matching degree is high, so a push is made; when T i ≤ R1, the matching level of the i-th enterprise is level three, and the matching degree is very high, so a push is made.

[0059] It should be noted that the preset value R1 is obtained based on experience and will not be elaborated here.

[0060] As an implementation manner of the present invention, the tracking level includes multiple levels; in step S4, the tracking information data includes a reading status and a download status; the reading status includes read and unread; the download status includes downloaded and not downloaded;

[0061] Through the above technical solution, in this embodiment, by setting multiple tracking levels, each tracking level corresponds to different secondary preset parameters and tertiary preset parameters; the secondary preset parameters are used to evaluate the tracking information data of enterprises with a matching level of level two; the tertiary preset parameters are used to evaluate the tracking information data of enterprises with a matching level of level three; the secondary preset parameters include a secondary preset reading rate and a secondary preset download rate; the tertiary preset parameters include a tertiary preset reading rate and a tertiary preset download rate; by analyzing the tracking information data of the enterprises to which the push is made and the secondary preset parameters and tertiary preset parameters selected by the staff, the push effect this time is judged.

[0062] It should be noted that the secondary preset reading rate, secondary preset download rate, tertiary preset reading rate, and tertiary preset download rate of each tracking level are preset values, obtained based on experience, and will not be elaborated here.

[0063] As an implementation manner of the present invention, in step S5, the process of judging whether the push this time meets the standard includes using the formula:

[0064]

[0065] to calculate the judgment index C;

[0066] where, RD2 is the number of enterprises with a matching level of level two and a reading status of read; PP2 is the total number of enterprises with a matching level of level two; ρ RD2 is the secondary preset reading rate of the tracking level set by the staff; ρ DL2The preset download rate for the second level of tracking set for staff; RD3 is the number of enterprises with a reading status of "read" at enterprise matching level three; PP3 is the total number of enterprises with an enterprise matching level of three; ρ RD3 The three preset reading rates for tracking levels set for staff; ρ DL3 The preset download rate for the three tracking levels set for staff; DL2 is the number of enterprises with a matching level of three that have downloaded; θ1 is the first weight coefficient; θ2 is the second weight coefficient; μ2 is the second weight coefficient; μ3 is the third weight coefficient;

[0067] If C≥1, this push notification meets the criteria; otherwise, this push notification does not meet the criteria.

[0068] Through the above technical solution, this embodiment Match the actual readership rate of enterprises at level two; The difference between the actual reading rate at level two for the enterprise and the preset reading rate at level two for the tracking level set by the staff, when... This indicates that the actual read rate of an enterprise with a matching level of 2 is higher than or equal to the preset read rate of 2, and the difference between the actual read rate of an enterprise with a matching level of 2 and the preset read rate of 2 for the tracking level set by the staff is the second-level read rate. The larger the value, the better the push effect; the larger the judgment index C, the better. This indicates that the actual reading rate for a company with a matching level of 2 is less than the preset reading rate for level 2. The difference between the actual reading rate for a company with a matching level of 2 and the preset reading rate for level 2 set by the staff is the second-level reading rate. The larger the absolute value, the worse the push effect, and the smaller the judgment index C. Match the actual download rate to level 2 for enterprises; The difference between the actual download rate at level 2 for enterprises and the preset download rate at level 2 for tracking levels set by staff, when... A value of ≥0 indicates that the actual download rate for a company with a matching level of 2 is higher than or equal to the preset download rate for level 2. The difference between the actual download rate for a company with a matching level of 2 and the preset download rate for level 2 set by the staff is the download rate for level 2. The larger the value, the better the push effect; the larger the judgment index C, the better. This indicates that the actual download rate for a company with a matching level of 2 is less than the preset download rate for level 2. The difference between the actual download rate for a company with a matching level of 2 and the preset download rate for level 2 set by the staff is the download rate difference. The larger the absolute value, the worse the push effect, and the smaller the judgment index C. Match the actual readership rate of enterprises at level three; The difference between the actual reading rate at level three for the enterprise and the preset reading rate at level three set by the staff for tracking is calculated. When this occurs, it indicates that the actual read rate for a company with a matching level of three is higher than or equal to the preset read rate for level three. The difference between the actual read rate for a company with a matching level of three and the preset read rate for level three set by the staff is considered a level three read rate. The larger the value, the better the push effect; the larger the judgment index C, the better. This indicates that the actual reading rate for a company with a matching level of three is less than the preset reading rate for level three. The difference between the actual reading rate for a company with a matching level of three and the preset reading rate for level three set by the staff is the third-level reading rate. The larger the absolute value, the worse the push effect, and the smaller the judgment index C. Match the actual download rate of enterprises at level three; The difference between the actual download rate at level three for enterprises and the preset download rate at level three set by staff for tracking is calculated. When this occurs, it indicates that the actual download rate for a company with a matching level of three is higher than or equal to the preset download rate for level three. The difference between the actual download rate for a company with a matching level of three and the preset download rate for level three (set by the staff) is the download rate for the tracking level. The larger the value, the better the push effect; the larger the judgment index C, the better. This indicates that the actual download rate for a company with a matching level of three is less than the preset download rate for level three. The difference between the actual download rate for a company with a matching level of three and the preset download rate for level three (set by the staff) is the download rate difference. The larger the absolute value, the worse the push effect, and the smaller the judgment index C. When C≥1, the push is successful; otherwise, the push is unsuccessful, and the push can be pushed again to users who have not read or downloaded it to improve the push effect until the push is successful.

[0069] It should be noted that the first weight coefficient θ1, the second weight coefficient θ2, the second weight coefficient μ2, and the third weight coefficient μ3 are preset values ​​obtained based on experience, and satisfy θ1+θ2=1; μ2+μ3=1; details will not be elaborated.

[0070] As one embodiment of the present invention, the actual tag matching degree P between the i-th enterprise and the current push policy i Through formula Calculated;

[0071] The actual keyword matching degree Q between the i-th enterprise and the policy pushed in this notification is... i Through formula Calculated;

[0072] Among them, X A Xs represents the total number of policy tags; Xs represents the number of tag matches between the i-th enterprise and the policy tags pushed in this policy push; Y represents the total number of policy tags. A Xs represents the total number of keywords in this push notification; Xs represents the number of keyword matches between the i-th enterprise and the keywords in this push notification.

[0073] Through the above technical solution, the policy tags in this embodiment include enterprise type, industry, enterprise size, etc.; the number of matches between the i-th enterprise and the policy tags of this push policy can be obtained by matching the enterprise tag of the i-th enterprise with the policy tags set in this push, and the method of obtaining the tag is existing technology and will not be described in detail here; the push keywords are automatically generated based on the policy text content and title of this push, and the generation method is existing technology and will not be described in detail here; the number of keyword matches between the i-th enterprise and the push keywords is obtained by analyzing the business scope or business content of the i-th enterprise with the push keywords, and the analysis process is existing technology and will not be described in detail here.

[0074] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A method for automatically pushing enterprise policies based on big data, characterized in that, The push method includes the following steps: S1: Configure the policy tag, push level, and tracking level for this push notification through the policy settings module; S2: The analysis module analyzes policy tags, push levels, policy text content, and enterprise information to obtain a policy matching index; S3: Analyze the policy matching index through the analysis module, select the enterprises to be pushed in this round, and push them to them; S4: Obtain tracking information data for each enterprise after this push through the tracking module; S5: The analysis module analyzes the enterprise matching level, tracking level, and tracking information data, and determines whether the push notification meets the standards based on the analysis results; The process for determining whether this push notification meets the criteria includes using the following formula: In the formula, Match the number of enterprises with a reading status of "read" at level 2; The total number of enterprises matched at level 2; The second-level preset reading rate for the tracking level set for staff; The preset download rate for the second level of tracking set for staff; Match the number of enterprises with a reading status of "read" at level three; The total number of enterprises matched with a level 3 rating; The three preset reading rates are set for the tracking levels of staff; The three preset download rates for the tracking levels set for staff; Match the number of downloaded enterprises at level three; This is the first weighting coefficient; This is the second weighting coefficient; These are the secondary weighting coefficients; These are the three-level weighting coefficients; Calculate the judgment index ; The process for determining whether this push notification meets the criteria also includes: when If the push notification meets the criteria, the push notification will be considered successful; otherwise, the push notification will be considered unsuccessful.

2. The method for automatically pushing enterprise policies based on big data according to claim 1, characterized in that, The policy matching index is calculated using the formula: Calculate the first The overall policy matching index of each enterprise with the policies pushed out this time. ; in, Assign a company number; For the first The degree of actual tag matching between each enterprise and the policy pushed out in this notification; Preset tag matching degree; For the first The degree of keyword matching between each enterprise and the policy pushed out in this notification; Preset keyword matching degree; As a basic policy matching index; The first weighting coefficient; This is the second weighting coefficient; This is the first preset constant; This is the second preset constant; For the first Each level of push notification level coefficient.

3. The method for automatically pushing enterprise policies based on big data according to claim 2, characterized in that, The push level includes Each level ; The first Each level of push notification coefficient satisfy And simultaneously satisfy .

4. The method for automatically pushing enterprise policies based on big data according to claim 3, characterized in that, In step S3, the process of selecting companies for this push notification is as follows: The first The overall policy matching index of each enterprise with the policies pushed out this time. Compared with preset value Compare; when At that time, the first The matching level for each enterprise is Level 1, and no push notifications will be sent. when At that time, the first The enterprise is matched at level two and a push notification is sent. when At that time, the first The enterprise is matched at level three and a push notification is sent.

5. The method for automatically pushing enterprise policies based on big data according to claim 4, characterized in that, The tracking level includes multiple levels; in step S4, the tracking information data includes reading status and download status; the reading status includes read and unread; the download status includes downloaded and not downloaded.

6. The method for automatically pushing enterprise policies based on big data according to claim 1, characterized in that, The first The actual tag matching degree between the individual enterprise and the policy pushed out in this notification Through formula Calculated; The first The actual keyword matching degree between each enterprise and the policy pushed out in this message. Through formula Calculated; in, The total number of policy labels; For the first The number of tags that match the policy tags of each enterprise in this push notification; This refers to the total number of keywords in this push notification. For the first The number of keywords that match the keywords in this push notification for each enterprise.

Images