Technical information issuing algorithm and system based on data processing platform

Abstract

The invention relates to the technical field of information publishing, and particularly discloses a technical information publishing algorithm and system based on a data processing platform, and the algorithm comprises the following steps: carrying out the verification and scoring of the text content of to-be-published technical information files of different users through an AI evaluation module in combination with an AI algorithm, the text content of the technical information file is checked and scored by combining the AI evaluation module, the quality of the technical information file to be published can be preliminarily judged, and the writing quality of the user can be reflected by the technical information file interaction data historically written by the user, so that the user experience is improved. Then, the quality of the technical information file is analyzed in combination with diversified data, so that the reliability of an analysis result can be improved, the optimization of a technical information publishing method is realized, the situation that the technical information is stolen and carried to other channels after being published in a single publishing channel is avoided, and the user experience is improved. Therefore, the right of the original user is prevented from being damaged.

Description

Technical Field

[0001] The present invention relates to the technical field of information publishing, and particularly to a technical information publishing algorithm and system based on a data processing platform. Background Art

[0002] Technical information is the information carried by technology. Technical information mainly includes two aspects: expressed technical information and unexpressed technical information. Among them, the expressed technical information can be expressed by means of words, graphics, symbols, etc.

[0003] In the process of technology development and application, in order to promote knowledge sharing and drive technological innovation, it is very necessary to actively publish and disseminate technical information. When traditional technical information is published, generally, the technical information is first published on the corresponding website through a technical service platform. Then, the data processing module in the technical service platform evaluates the quality of the technical information according to the interactive data such as the real-time view volume, click volume, and conversion rate of the technical information, and decides whether to increase the publishing channels of the technical information, so as to ensure that high-quality technical information can be known to more people and improve the professionalism of the technical service platform itself.

[0004] In the prior art for the publication of technical information, generally, the technical information is published on the corresponding website through a technical service platform. Then, the data processing module decides whether to increase the publishing channels of the technical information according to the real-time interactive data of the technical information. Since this method makes decisions based on the real-time data after the publication of the technical information, it takes a certain amount of time to analyze and judge the quality of the technical information. During this period, the high-quality content in the technical information may be stolen and moved to other channels, resulting in the original user being unable to publish the technical information on other channels, causing the rights of the original user to be violated and resulting in economic losses to the original creator, and affecting the professionalism of the technical service platform itself. Summary of the Invention

[0005] The purpose of the present invention is to provide a technical information publishing algorithm and system based on a data processing platform, and solve the following technical problems:

[0006] How to optimize the technical information publishing method to avoid the infringement of the rights of the original user.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A technical information publishing algorithm and system based on a data processing platform, the algorithm comprising the following steps:

[0009] S1: The text content of the technical information files to be released by different users is verified and scored through the AI evaluation module in combination with the AI algorithm, and this data is recorded;

[0010] S2: The interaction data of the technical information files released by the users who wrote the technical information is collected through the data collection module in the information platform;

[0011] S3: The quality of the technical information files to be released by the user is analyzed through the data evaluation module in combination with the data recorded in the AI evaluation module;

[0012] S4: The user profile of the user and the quality of the technical information files to be released are analyzed through the data processing platform in combination with the analysis results of the data evaluation module and the data recorded by the data collection module;

[0013] S5: The release channels of the technical information files to be released by the user are adjusted through the adjustment module in combination with the user profile of the user and the analysis results of the quality of the technical information files to be released.

[0014] Further, the analysis process in S3 includes:

[0015] Through the formula Calculate the passing index of the technical information file to be released by the a-th user

[0016] where a is any user, cc a is the duplicate rate of the technical information file to be released by the a-th user, cc y is the preset duplicate rate, cw a is the number of text misspelled words in the technical information file to be released by the a-th user, px a is the number of text spelling mistakes in the technical information file to be released by the a-th user, bd a is the number of text punctuation mistakes in the technical information file to be released by the a-th user, cw y is the preset number of text mistakes, x1, x2 and x3 are the first weight coefficients, set by fitting according to empirical data, i is any AI algorithm, n is the total number of AI algorithms for scoring, fs ai is the practicality score of the i-th AI algorithm for the technical information file to be released by the a-th user, wz ai is the innovation score of the i-th AI algorithm for the technical information file to be released by the a-th user, γ i is the empirical weight coefficient of the i-th AI algorithm, set by empirical fitting.

[0017] Further, the analysis process in S3 also includes;

[0018] By comparing the qualified index of the technical information file to be released by the a-th user with the preset threshold of the file qualified index for comparison;

[0019] If it is determined that the technical information file to be released by the a-th user is unqualified, and the release of this file is refused;

[0020] If it is determined that the technical information file to be released by the a-th user is qualified, the file can be released.

[0021] Furthermore, the analysis process in S3 also includes:

[0022] By assigning a value to the qualified index of the technical information file to be released by the a-th user to generate a quality weight value of the technical information file to be released by the a-th user, which is between 1 and 3 and increases as the qualified index of the technical information file to be released by the a-th user increases;

[0023] Among them, the quality weight value of the technical information file to be released by the a-th user corresponding to the qualified index of the technical information file to be released by the a-th user is set as

[0024] Furthermore, the analysis process in S4 includes:

[0025] By the formula calculate the readability index μ of the technical information files written by the a-th user in history a ;

[0026] Among them, t is any one of the technical information files written in history, z is the total number of historical writings of technical information files, djl at is the click-through rate of the t-th technical information file written by the a-th user in history, djl b is the standard value of djl at , fw at is the total viewing duration of the t-th technical information file written by the a-th user in history, fw b is the standard value of fw at , fsj at is the number of days since the release of the t-th technical information file written by the a-th user in history, ly at is the number of sharing times of the t-th technical information file written by the a-th user in history, ly y is the preset number of sharing times, ly b is the standard value of ly at is the standard value of f k ​To define a function, if f k (x)≥x, then let f k (x)=x; otherwise, let f k (x)=1.

[0027] Furthermore, the analysis process in S4 also includes:

[0028] Obtain the quality qualification index p of the technical information file to be released by the a-th user through the formula ; a

[0029] And compare the quality qualification index p of the technical information file to be released by the a-th user a with the preset quality qualification index threshold p 01 ;

[0030] If p a ≥p 01 , determine that the technical information file to be released by the a-th user is a high-quality file;

[0031] If p a <p 01 , determine that the technical information file to be released by the a-th user is a medium-quality file.

[0032] Furthermore, the adjustment process in S5 includes:

[0033] When it is determined that the technical information file to be released by the a-th user is a high-quality file, various channels can be considered for wide release to ensure that it is known and used by more people;

[0034] When it is determined that the technical information file to be released by the a-th user is a medium-quality file, release it for specific audiences or needs, and establish a feedback mechanism to collect the opinions and suggestions of the audiences.

[0035] A technical information release system based on a data processing platform, the system includes:

[0036] An AI evaluation module, used to verify and score the text content of the technical information files to be released by different users in combination with AI algorithms, and record this data;

[0037] A data acquisition module, used to collect the interaction data of the technical information files previously released by the user who writes the technical information in the information platform;

[0038] A data evaluation module, used to analyze the quality of the technical information file to be released by this user in combination with the data recorded in the AI evaluation module;

[0039] ​A data processing platform is used to analyze the user profile and the quality of technical information of the user by combining the analysis results of the data evaluation module with the data recorded by the data collection module.

[0040] An adjustment module is used to adjust the release channel of the technical information file to be released by the user in combination with the analysis results of the user profile and the quality of technical information of the user by the data processing platform.

[0041] Advantages of the present invention:

[0042] (1) By combining the AI evaluation module to verify and score the text content of the technical information file, the present invention can make a preliminary judgment on the quality of the technical information file to be released. The interaction data of the technical information files written by the user in the past can reflect the writing quality of the user. Subsequently, by combining diversified data to analyze the quality of the technical information file, the reliability of the analysis results can be improved, so as to optimize the technical information release method, avoid the situation that the technical information is stolen and transported to other channels after being released on a single release channel, and thus avoid the infringement of the rights of the original user.

[0043] (2) By comparing the qualified index of the technical information file to be released by the a-th user with the preset file qualified index threshold , through this comparison method, an accurate judgment can be made on whether the technical information file to be released by the a-th user is qualified, so as to make a decision on whether the technical information is eligible for release. And when it is judged that the technical information file to be released by the a-th user is qualified, it means that the quality of the technical information file is at a qualified level. Based on this data, diversified data support can be provided for adjusting the release method of the technical information file subsequently, so as to improve the reliability of the adjustment results.

[0044] (3) By comparing the quality qualified index p a of the technical information file to be released by the a-th user with the preset quality qualified index threshold p 01 , a judgment can be made on whether the technical information file to be released by the a-th user is a high-quality file by combining diversified data. The diversified data can improve the accuracy and reliability of the judgment results, so as to optimize the technical information release method, avoid the situation that the technical information is stolen and transported to other channels after being released on a single release channel, and thus avoid the infringement of the rights of the original user.

[0045] (4) The present invention makes different release methods for technical documents of different qualities by combining the judgment result of whether the technical information file to be released by the a-th user is a high-quality file. For high-quality technical information files, they are directly and widely released through multiple channels to ensure that they are known and used by more people, and to avoid the situation where the high-quality content in the technical information may be stolen and transported to other channels after being released through a single release channel, thereby avoiding the infringement of the rights of the original users. For medium-quality technical information files, they are released for specific audiences or needs, and a feedback mechanism is established to facilitate the improvement and optimization of the files.

[0046] (5) The present invention can make a preliminary judgment on whether the technical information file to be released is qualified by first verifying and scoring the text content of the technical information file to be released by the user. Then, through the data processing platform, by combining the analysis results of the data evaluation module and the data recorded by the data acquisition module, the user profile and the quality of the technical information of the user can be analyzed, so as to analyze the overall quality of the technical information file to be released. Then, by combining this data to adjust the release channel, the optimization of the technical information release method can be achieved, and the situation where the rights of the original user are infringed, resulting in economic losses to the original creator, can be avoided. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] The present invention will be further described below with reference to the accompanying drawings.

[0048] Figure 1 is a flowchart of the steps of a technical information release algorithm based on a data processing platform in the present invention;

[0049] Figure 2 is a schematic block diagram of a technical information release system based on a data processing platform in the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0050] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0051] Please refer to Figure 1 As shown, in one embodiment, the present application provides a technical information release algorithm and system based on a data processing platform. The algorithm includes the following steps:

[0052] S1: The text content of the technical information files to be released by different users is verified and scored by the AI evaluation module in combination with the AI algorithm, and this data is recorded;

[0053] S2: The interaction data of the technical information files published by the user who wrote the technical information is collected by the data collection module in the information platform;

[0054] S3: The quality of the technical information files to be released by this user is analyzed by the data evaluation module in combination with the data recorded in the AI evaluation module;

[0055] S4: The user profile of this user and the quality of the technical information files to be released are analyzed by the data processing platform in combination with the analysis results of the data evaluation module and the data recorded by the data collection module;

[0056] S5: The release channels of the technical information files to be released by the user are adjusted by the adjustment module in combination with the user profile of this user and the analysis results of the quality of the technical information files to be released;

[0057] Through the above technical solution, this example provides a technical information release algorithm based on a data processing platform, including the following steps. First, the text content of the technical information files to be released by different users is verified and scored by the AI evaluation module in combination with the AI algorithm, and this data is recorded. Then, the interaction data of the technical information files published by the user who wrote the technical information is collected by the data collection module in the information platform, and the quality of the technical information files to be released by this user is analyzed by the data evaluation module in combination with the data recorded in the AI evaluation module. Subsequently, the user profile of this user and the quality of the technical information files to be released are analyzed by the data processing platform in combination with the analysis results of the data evaluation module and the data recorded by the data collection module. Finally, the release channels of the technical information files to be released by the user can be adjusted by the adjustment module in combination with the user profile of this user and the analysis results of the quality of the technical information files to be released;

[0058] By setting it like this, when a user submits a technical information file to be released, first, by combining the AI evaluation module to verify and score the text content of the technical information file, a preliminary judgment can be made on the quality of the technical information file to be released. And the interaction data of the technical information files written by this user in history can reflect the writing quality of this user. Subsequently, by combining diversified data to analyze the quality of the technical information file, the reliability of the analysis results can be improved, so as to realize the optimization of the technical information release method, avoid the situation that the technical information is stolen and transported to other channels after being released through a single release channel, and thus avoid the infringement of the rights of the original users.

[0059] The analysis process in the above S3 includes:

[0060] The eligibility index of the technical information file to be released by the ath user is calculated through the formula where a is any user, cc

[0061] is the duplicate check rate of the technical information file to be released by the ath user, cc a is the preset duplicate check rate, cw y is the number of text spelling mistakes in the technical information file to be released by the ath user, px a is the number of text spelling errors in the technical information file to be released by the ath user, bd a is the number of text punctuation errors in the technical information file to be released by the ath user, cw a is the preset number of text errors, x1, x2, and x3 are the first weight coefficients, which are set by fitting empirical data, i is any AI algorithm, and n is the total number of AI algorithms used for scoring, fs y is the practicality score of the technical information file to be released by the ath user by the ith AI algorithm, wz ai is the innovation score of the technical information file to be released by the ath user by the ith AI algorithm, γ ai is the empirical weight coefficient of the ith AI algorithm, which is set by empirical fitting; i Through the above technical solution, this example provides the eligibility index of the technical information file to be released by the ath user

[0062] which can be calculated through the formula Obviously, when the duplicate check rate of the technical information file to be released by the ath user is lower, the number of text punctuation errors, text spelling mistakes, and text spelling errors in the technical information file is less, and the average score of the practicality and innovation of the technical information file is higher, then the eligibility index of the technical information file to be released by the ath user is higher, and the quality of the technical information file to be released by the ath user is higher. On the contrary, when the duplicate check rate of the technical information file to be released by the ath user is higher, the number of text punctuation errors, text spelling mistakes, and text spelling errors in the technical information file is more, and the average score of the practicality and innovation of the technical information file is lower, then the eligibility index of the technical information file to be released by the ath user is lower, which means that the writing of this technical information is poor; That is, it represents that the writing of this technical information is poor;

[0063] Through this calculation method, accurate data can be provided for subsequent judgment on whether the technical information file to be released by the ath user is qualified, thereby improving the accuracy and reliability of the judgment result.

[0064] The analysis process in S3 further includes;

[0065] By comparing the qualified index of the technical information file to be released by the a-th user with the preset threshold of the file qualified index for comparison;

[0066] If it is determined that the technical information file to be released by the a-th user is unqualified, and the release of this file is refused;

[0067] If it is determined that the technical information file to be released by the a-th user is qualified, the file can be released;

[0068] Through the above technical solution, in this example, by comparing the qualified index of the technical information file to be released by the a-th user with the preset threshold of the file qualified index for comparison, through this comparison method, an accurate judgment can be made on whether the technical information file to be released by the a-th user is qualified, so as to make a decision on whether the technical information is eligible for release. And when it is determined that the technical information file to be released by the a-th user is qualified, it means that the quality of the technical information file is at a qualified level. Based on this data, it can provide diversified data support for subsequent adjustment of the release method of the technical information file, thereby improving the reliability of the adjustment result.

[0069] The analysis process in S3 further includes:

[0070] By assigning a value to the qualified index of the technical information file to be released by the a-th user to generate a quality weight value of the technical information file to be released by the a-th user that is between 1 and 3 and increases as the qualified index of the technical information file to be released by the a-th user increases;

[0071] Among them, the quality weight value of the technical information file to be released by the a-th user corresponding to the qualified index of the technical information file to be released by the a-th user is set as

[0072] Through the above technical solution, this example provides the assignment process of the qualified index of the technical information file to be released by the a-th user ;

[0073] As an embodiment, the value-taking standard of

[0074]

[0075] It should be noted that as the qualification index of the technical information file to be released by the a-th user increases, the quality weight value of the technical information file to be released by the a-th user will increase. That is because when the qualification index of the technical information file to be released by the a-th user is higher, it means that the quality of the technical information to be released by the a-th user is higher. Then, it is necessary to increase the release channels of this technical information to ensure that more people can know about high-quality technical information. Therefore, when the quality weight value of the technical information file to be released by the a-th user increases with the increase, it can reflect the quality of this technical information, thus providing diversified data support for subsequent adjustment of the release method of the technical information file and improving the reliability of the adjustment result.

[0076] The analysis process in S4 includes:

[0077] Calculate the readability index μ of the technical information files historically written by the a-th user through the formula ; a

[0078] where t is any one of the technical information files historically written, z is the total number of historical writings of the technical information files, djl at is the click volume of the t-th technical information file historically written by the a-th user, djl b is the standard value of djl at , and the above standard value can be selected and set according to the allowable error in the empirical data. fw at is the total viewing duration of the t-th technical information file historically written by the a-th user, fw b is the standard value of fw at , and the above standard value can be selected and set according to the allowable error in the empirical data. fsj at is the number of days since the release of the t-th technical information file historically written by the a-th user, ly at is the number of shares of the t-th technical information file historically written by the a-th user, ly y is the preset number of shares, ly b is the standard value of ly at , and the above standard value can be selected and set according to the allowable error in the empirical data. f k is a defined function. If f k (x)≥x, then let f k (x) = x; otherwise, let f k (x) = 1;

[0079] ​Through the above technical solution, this example provides the readability index μ of the technical information file written by the a-th user in the past a , which can be obtained by the formula . Obviously, when the click-through rate, total viewing duration, and sharing times of the t-th technical information file written by the a-th user in the past are more, and the release days of the t-th technical information file written by the a-th user in the past are shorter, then the readability index μ of the technical information file written by the a-th user in the past a is higher. This is because the more the click-through rate, total viewing duration, and sharing times of the technical information file are, it means the higher the content quality of this historical technical information file. And the shorter the release days are, it means the more interactive data per day, further reflecting the high content quality of this historical technical information file. Therefore, when the click-through rate, total viewing duration, and sharing times of the t-th technical information file written by the a-th user in the past are less, and the release days of the t-th technical information file written by the a-th user in the past are longer, then the readability index μ of the technical information file written by the a-th user in the past a is lower;

[0080] Through this calculation method, the interactive data of the technical information files written by the a-th user in the past on this platform can be combined to analyze the overall writing quality of this user. By combining the analysis results, the user portrait of the user based on the overall writing quality can be analyzed, so as to further provide diversified data support for adjusting the release method of technical information files subsequently, thereby improving the reliability of the adjustment results.

[0081] The analysis process in S4 further includes:

[0082] The quality qualification index p of the technical information file to be released by the a-th user is obtained by the formula ; a ;

[0083] And by comparing the quality qualification index p of the technical information file to be released by the a-th user a with the preset quality qualification index threshold p 01 ;

[0084] If p a ≥ p 01 , it is determined that the technical information file to be released by the a-th user is a high-quality file;

[0085] If p a < p 01 , it is determined that the technical information file to be released by the a-th user is a medium-quality file;

[0086] Through the above technical solution, this example provides the quality qualification index p of the technical information file to be published by the a-th user a , which can be obtained by the formula . After that, by comparing the quality qualification index p a of the technical information file to be published by the a-th user with the preset quality qualification index threshold p 01 , a judgment can be made on whether the technical information file to be published by the a-th user is a high-quality file by combining diversified data. The diversified data can improve the accuracy and reliability of the judgment result, so as to optimize the technical information publishing method, avoid the situation that the technical information is stolen and transported to other channels after being published on a single publishing channel, and thus avoid the infringement of the rights of the original user.

[0087] The adjustment process in S5 includes:

[0088] When it is judged that the technical information file to be published by the a-th user is a high-quality file, it can be considered to be widely published through multiple channels to ensure that it is known and used by more people;

[0089] When it is judged that the technical information file to be published by the a-th user is a medium-quality file, it is published for specific audiences or needs, and a feedback mechanism is established to collect the opinions and suggestions of the audiences;

[0090] Through the above technical solution, this example makes different publishing methods for technical files of different qualities by combining the judgment result of whether the technical information file to be published by the a-th user is a high-quality file. For high-quality technical information files, they are directly widely published through multiple channels to ensure that they are known and used by more people, and to avoid the situation that the high-quality content in the technical information may be stolen and transported to other channels after being published on a single publishing channel, thus avoiding the infringement of the rights of the original user. For medium-quality technical information files, they are published for specific audiences or needs, and a feedback mechanism is established to improve and optimize the files.

[0091] Please refer to Figure 2 as shown. A technical information publishing system based on a data processing platform, the system includes:

[0092] An AI evaluation module, used to verify and score the text content of the technical information files to be published by different users in combination with AI algorithms, and record this data;

[0093] A data acquisition module, used to acquire the interaction data of the technical information files previously published by the technical information writing user in the information platform;

[0094] A data evaluation module, which is used to analyze the quality of the technical information file to be published by the user in combination with the data recorded in the AI evaluation module;

[0095] A data processing platform, which is used to analyze the user portrait and the quality of technical information of the user in combination with the analysis results of the data evaluation module and the data recorded by the data collection module;

[0096] An adjustment module, which is used to adjust the publishing channel of the technical information file to be published by the user in combination with the analysis results of the user portrait and the quality of technical information of the user by the data processing platform;

[0097] Through the above technical solutions, this example provides an AI evaluation module. Before publishing the technical information of the user, first, the AI evaluation module combines the AI algorithm to check and score the text content of the technical information files to be published by different users, and records this data. Then, the data collection module collects the interaction data of the technical information files published by the user in the information platform when writing the historical technical information, and the data evaluation module combines the data recorded in the AI evaluation module to analyze the quality of the technical information file to be published by the user. Then, through the data processing platform, combining the analysis results of the data evaluation module and the data recorded by the data collection module, the user portrait and the quality of technical information of the user can be analyzed. Finally, through the adjustment module, combining the analysis results of the user portrait and the quality of technical information of the user by the data processing platform, the publishing channel of the technical information file to be published by the user can be adjusted;

[0098] By setting it like this, by first checking and scoring the text content of the technical information file to be published by the user, a preliminary judgment can be made on whether the technical information file to be published is qualified. Then, through the data processing platform, combining the analysis results of the data evaluation module and the data recorded by the data collection module, the user portrait and the quality of technical information of the user can be analyzed, so as to analyze the overall quality of the technical information file to be published. Then, by combining this data to adjust the publishing channel, the optimization of the technical information publishing method can be realized, and the situation of the original user's rights being violated and the original author suffering economic losses can be avoided.

[0099] The above has described an embodiment of the present invention in detail, but the content described is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the application of the present invention should still fall within the scope covered by the patent of the present invention.

Claims

1. A technical information publishing algorithm based on a data processing platform, characterized in that, The algorithm includes the following steps: S1: The AI evaluation module combines with the AI algorithm to verify and score the text content of the technical information files to be published by different users, and record this data; S2: The data acquisition module collects the interaction data of the technical information files published by the user who writes this technical information in the information platform; S3: The data evaluation module combines with the data recorded in the AI evaluation module to analyze the quality of the technical information files to be published by this user; S4: The data processing platform combines the analysis results of the data evaluation module with the data recorded by the data acquisition module to analyze the user portrait of this user and the quality of the technical information files to be published; S5: The adjustment module combines the user portrait of this user with the analysis results of the quality of the technical information files to be published to adjust the publishing channels of the technical information files to be published by the user.

2. The technical information publishing algorithm based on a data processing platform according to claim 1, characterized in that, The analysis process in S3 includes: Obtained through the formula Calculate the qualification index θ of the technical information file to be released by the a-th user a ; Among them, a is any user, cc a is the duplicate check rate of the technical information file to be released by the a-th user, cc y is the preset duplicate check rate, cw a is the number of text misspelled words in the technical information file to be released by the a-th user, px a is the number of text spelling mistakes in the technical information file to be released by the a-th user, bd a is the number of text punctuation mistakes in the technical information file to be released by the a-th user, cw y is the preset number of text mistakes, x1, x2 and x3 are the first weight coefficients, set by fitting according to empirical data, i is any one of the AI algorithms, n is the total number of AI algorithms used for scoring, fs ai is the practicality score of the i-th AI algorithm for the technical information file to be released by the a-th user, wz ai is the innovation score of the i-th AI algorithm for the technical information file to be released by the a-th user, γ i is the empirical weight coefficient of the i-th AI algorithm, set by empirical fitting.

3. The technical information publishing algorithm based on a data processing platform according to claim 2, characterized in that, The analysis process in S3 also includes; By comparing the qualified index θ of the technical information file to be released by the a-th user a with the preset threshold value θ of the file qualified index 01 for comparison; If θ a < θ 01 , it is determined that the technical information file to be released by the a-th user is unqualified, and the release of this file is refused; If θ a ≥ θ 01 , it is determined that the technical information file to be released by the a-th user is qualified, and the file can be released.

4. The technical information publishing algorithm based on a data processing platform according to claim 3, wherein The analysis process in S3 also includes: By assigning a value to the qualification index θ of the technical information file to be released by the a-th user a a quality weight value of the technical information file to be released by the a-th user has been generated, which is between 1 and 3 and increases as the qualification index θ a of the technical information file to be released by the a-th user increases; Among them, the qualified index θ of the technical information file to be released by the a-th user a The quality weight value of the technical information file to be released by the a-th user corresponding to it is set as f(θ a ).

5. The technical information publishing algorithm based on a data processing platform according to claim 4, characterized in that, The analysis process in S4 includes: Through the formula calculate the readability index μ of the technical information file written by the a-th user in history a ; Among them, t is any technical information document written in history, z is the total number of historical writings of technical information documents, and djl at is the click volume of the t-th technical information document written by the a-th user in history, djl b is the standard value of djl at and fw at is the total viewing duration of the t-th technical information document written by the a-th user in history, fw b is the standard value of fw at and fsj at is the number of days since the release of the t-th technical information document written by the a-th user in history, ly at is the number of shares of the t-th technical information document written by the a-th user in history, ly y is the preset number of shares, ly b is the standard value of ly at and f k is a defined function. If f k (x) ≥ x, then let f k (x) = x; otherwise, let f k (x) = 1.

6. The technical information publishing algorithm based on a data processing platform according to claim 5, characterized in that The analysis process in S4 also includes: Obtained through the formula Calculate the quality qualification index p of the technical information file to be released by the a-th user a ; And by comparing the quality qualification index p of the technical information file to be released by the a-th user a with the preset quality qualification index threshold p 01 for comparison; If p a ≥ p 01 , it is determined that the technical information file to be released by the a-th user is a high-quality file; If p a <p 01 , it is determined that the technical information file to be released by the a-th user is a medium-quality file.

7. A technical information publishing algorithm based on a data processing platform according to claim 6, characterized in that, The adjustment process in S5 includes: When it is determined that the technical information file to be published by the a-th user is a high-quality file, extensive publication through multiple channels can be considered to ensure that it is known and used by more people; When it is determined that the technical information file to be published by the a-th user is a medium-quality file, publish it for specific audiences or needs, and establish a feedback mechanism to collect the opinions and suggestions of the audiences.

8. A technical information publishing system based on a data processing platform, applicable to the system of a technical information publishing algorithm based on a data processing platform as described in claims 1-7, characterized in that, The system includes: An AI evaluation module, which is used to combine with the AI algorithm to verify and score the text content of the technical information files to be published by different users, and record this data; A data acquisition module, which is used to collect the interaction data of the technical information files published by the user who writes this technical information in the information platform; A data evaluation module, which is used to combine with the data recorded in the AI evaluation module to analyze the quality of the technical information files to be published by this user; A data processing platform, which is used to combine the analysis results of the data evaluation module with the data recorded by the data acquisition module to analyze the user portrait of this user and the quality of the technical information; An adjustment module, which is used to combine the analysis results of the user portrait of this user and the quality of the technical information by the data processing platform to adjust the publishing channels of the technical information files to be published by the user.

Images