News transmission method and system based on big data processing

Abstract

The invention relates to the technical field of news transmission, in particular to a news transmission method and system based on big data processing. The method comprises the following steps: collecting user data, establishing a hierarchical label system, and generating a multi-dimensional label user portrait; acquiring news content features, predicting a news propagation path in combination with the user portrait, and performing path layering on the propagation path; extracting a discrimination keyword from the propagation content, matching the discrimination keyword with the user portrait, and monitoring a propagation termination triggering condition according to a matching condition; acquiring a propagation termination triggering condition, requesting a data propagation channel according to the propagation termination triggering condition, and ending the propagation process; the system comprises a user portrait construction module, a news propagation path prediction module, a monitoring and discrimination cooperation module and a data propagation channel control module. According to the method, the data is discriminated for the user portraits in the news transmission, the discriminated keywords are used as the user news blacklist, the data transmission channel is terminated in time, and the news transmission accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the field of news dissemination technology, and in particular to a news dissemination method and system based on big data processing. Background Art

[0002] With the rapid development and widespread application of Internet technology, news websites and software platforms have gradually become the core platforms for people to obtain information and share opinions. In order to improve the efficiency of news information dissemination, we now use keyword matching based on big data processing to disseminate news to users.

[0003] However, in the above-mentioned news dissemination method, the news dissemination path is relatively fixed. When the news content is in the process of dissemination, it is impossible to terminate the dissemination path in time when the user changes the keywords in the middle of the dissemination, which reduces the accuracy of news dissemination. Summary of the Invention

[0004] The purpose of the present invention is to provide a news dissemination method and system based on big data processing, aiming to solve the technical problem in the prior art that the news dissemination path is relatively fixed, and when the news content is being disseminated, the dissemination path cannot be terminated in time when the user changes the keywords in the middle of the dissemination, thereby reducing the accuracy of news dissemination.

[0005] To achieve the above-mentioned purpose, the present invention adopts a news dissemination method based on big data processing, which includes the following steps:

[0006] Collect user registration data, behavior data, and geographic location data, establish a hierarchical tag system, and generate multi-dimensional tag user portraits based on user data;

[0007] Obtain news content features, combine them with user portraits to predict news dissemination paths, and stratify the dissemination paths;

[0008] Extract discriminant keywords from the dissemination content, match them with user portraits, and monitor the termination trigger of the dissemination based on the matching results;

[0009] Obtain the propagation termination trigger condition, request the data propagation channel according to the propagation termination trigger condition, and end the propagation process.

[0010] Among them, in the steps of collecting user registration data, behavior data and geographic location data, establishing a hierarchical label system, and generating multi-dimensional label user portraits based on user data:

[0011] Trigger user data request instructions to collect user registration data, behavior data and geographic location data respectively;

[0012] Divide the label system into levels, clean and pre-process the data, extract user features, and generate labels based on the extracted features;

[0013] Assign data objects, associate tags with data objects, and generate user portraits with multi-dimensional tags.

[0014] After allocating data objects, associating tags with data objects, and generating user profiles with multi-dimensional tags:

[0015] Monitor label changes at each layer separately and update user profiles in real time.

[0016] Among them, in the steps of obtaining news content features, combining user portraits to predict news dissemination paths, and stratifying the dissemination paths:

[0017] Acquire news content data, analyze the news content features, match the news content features with user profiles, and match user preferred news;

[0018] According to the matching results, the news propagation path is output;

[0019] The propagation path is layered; the path includes the core layer, the extension layer, and the peripheral layer.

[0020] Among them, after the step of layering the propagation paths:

[0021] Provides discrimination layer and monitoring layer for propagation path.

[0022] The core layer is used to provide news content to users with stable preferences.

[0023] The extension layer is used to provide news content to users with interest preferences;

[0024] The peripheral layer is used to provide news content to users in the local area;

[0025] The discrimination layer is used to match news content features with updated user profiles and output discrimination keywords;

[0026] The monitoring layer is used to obtain the identification keywords and operate the transmission status of the propagation path.

[0027] Among them, in the steps of extracting discriminant keywords from the dissemination content, matching them with user portraits, and monitoring the termination trigger of the dissemination based on the matching situation:

[0028] Monitor news dissemination data in real time, obtain news content features, extract discriminant keywords based on news content features, and match the discriminant keywords with the updated user profile;

[0029] Set a similarity threshold, calculate a first trigger condition value based on the matching result, and output a first trigger result;

[0030] Set the blacklist trigger threshold, calculate the second trigger condition value based on the matching result, and output the second trigger result;

[0031] A propagation termination trigger operation is performed according to the first trigger result and the second trigger result.

[0032] Among them, in the steps of setting the similarity threshold, calculating the first trigger condition value according to the matching result, and outputting the first trigger result:

[0033] When the first trigger result is that the first trigger condition value is less than the similarity threshold, the news dissemination is terminated.

[0034] Among them, in the step of setting the blacklist trigger threshold, calculating the second trigger condition value according to the matching result, and outputting the second trigger result:

[0035] When the second trigger result is that the second trigger condition value is equal to the blacklist trigger threshold, the news dissemination is terminated.

[0036] The present invention also provides a news dissemination system based on big data processing, including a user portrait construction module, a news dissemination path prediction module, a monitoring and discrimination collaboration module, and a data dissemination channel control module; wherein:

[0037] The user portrait construction module is used to collect user registration data, behavior data and geographic location data, establish a hierarchical label system, and generate a multi-dimensional label user portrait based on the user data;

[0038] The news propagation path prediction module is used to obtain news content features, predict news propagation paths based on user portraits, and perform path stratification on the propagation paths;

[0039] The monitoring and discrimination collaborative module is used to extract discrimination keywords from the dissemination content, match them with the user profile, and monitor the termination trigger of the dissemination based on the matching situation;

[0040] The data propagation channel control module is used to obtain a propagation termination trigger condition, request a data propagation channel according to the propagation termination trigger condition, and end the propagation process.

[0041] The present invention provides a news dissemination method and system based on big data processing, which respectively uses the user portrait construction module, the news dissemination path prediction module, the monitoring and discrimination collaborative module, and the data dissemination channel control module to perform the following processes: collecting user registration data, behavior data, and geographic location data, establishing a hierarchical label system, and generating a multi-dimensional label user portrait for user data; obtaining news content features, combining the user portrait to predict the news dissemination path, and performing path stratification on the dissemination path; extracting discriminant keywords from the dissemination content, matching them with the user portrait, and monitoring the dissemination termination trigger according to the matching situation; obtaining the dissemination termination trigger, requesting a data dissemination channel according to the dissemination termination trigger, and ending the dissemination process. Based on user usage data, a user portrait is constructed, and the news dissemination path is predicted, and the dissemination path is layered. The monitoring layer is used to obtain the discriminant keywords of the data from the discriminant layer, monitoring the dissemination termination trigger, and requesting a data dissemination channel; by discriminating the data according to the user portrait in news dissemination, the discriminant keywords are used as the user news blacklist, and the data dissemination channel is terminated in time, thereby improving the accuracy of news dissemination. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. 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 any creative work.

[0043] Figure 1 It is a flowchart of the steps of the news dissemination method based on big data processing of the present invention.

[0044] Figure 2 It is a step flow chart of S100 of the present invention.

[0045] Figure 3 It is a step flow chart of S200 of the present invention.

[0046] Figure 4 It is a step flow chart of S300 of the present invention.

[0047] Figure 5 It is a structural principle diagram of the news dissemination system based on big data processing of the present invention.

[0048] Figure 6 It is a structural principle diagram of the electronic device of the present invention.

[0049] 501-User portrait construction module, 502-News dissemination path prediction module, 503-Monitoring and discrimination collaboration module, 504-Data dissemination channel control module. DETAILED DESCRIPTION

[0050] Exemplary embodiments are described in detail herein, with examples illustrated in the accompanying drawings. When the following description refers to the drawings, identical numerals in different drawings represent identical or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with this application.

[0051] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. As used in this application and the appended claims, the singular forms "a," "an," "the," and "the" are intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0052] It should be understood that although the terms first, second, third, etc. may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0053] See also Figures 1 to 4 The present invention provides a news dissemination method based on big data processing, comprising the following steps:

[0054] S100: Collect user registration data, behavior data, and geographic location data, establish a hierarchical labeling system, and generate multi-dimensional label user portraits based on user data.

[0055] In this implementation, user registration data, behavior data, and geographic location data are collected to establish a hierarchical tag system, and a multi-dimensional tag user profile is generated based on the user data. The specific process is as follows:

[0056] S101: Triggering a user data request instruction to collect user registration data, behavior data, and geographic location data;

[0057] S102: Divide the label system into layers, clean and pre-process the data, extract user features, and generate labels based on the extracted features;

[0058] S103: Allocate data objects, associate tags with data objects, and generate user profiles with multi-dimensional tags;

[0059] S104: Monitor the label changes of each layer separately and update the user portrait in real time.

[0060] In the above process, the user data request instruction is first triggered to collect user registration data, behavioral data and geographic location data respectively; the registration data includes user name, gender, date of birth, and email address; the behavioral data includes user browsing behavior, search behavior, click behavior, interaction behavior, and purchase behavior; the geographic location data is obtained through GPS, mobile phone signal towers, WiFi, and IP address.

[0061] The collected data is evaluated using the quality assessment formula:

[0062] Completeness = number of non-empty fields / total number of fields.

[0063] Then, the label system is divided into multiple layers, such as: the first-level label is user attributes, the second-level label is geographic location, and the third-level label is interest preference; after cleaning and preprocessing the data, user features are extracted and labels are generated based on the extracted features. The feature extraction formula uses PCA dimensionality reduction:

[0064] Feature extraction formula, taking the user behavior matrix as an example:

[0065] T = XP;

[0066] where X∈R n×m is the original behavior matrix, P∈R m×k is the projection matrix, T∈R n×k is the feature after dimensionality reduction.

[0067] At the same time, the data is processed for outlier judgment. For example, if the GPS coordinate jumps, the previous coordinate is Hangzhou, and the coordinate changes to Beijing in a short time, the previous coordinate will be retained.

[0068] Assign data objects and associate tags with them to generate a multi-dimensional user profile. The data object uniquely identifies the user, which can be generated from a phone number and timestamp. Associate each tag with the unique user to generate a user profile composed of multi-dimensional tags.

[0069] In user portrait generation, association rule mining (Apriori algorithm) is used:

[0070] Support calculation:

[0071]

[0072] Confidence calculation:

[0073]

[0074] Monitor changes in labels at each layer separately and update user portraits in real time. For example, if behavioral changes are detected, such as users browsing maternal and child products or clicking on video explanations, the label will be updated after the same behavior is obtained multiple times in a row, where the number of consecutive times can be 3, 5, etc.

[0075] In the update tag, the real-time change monitoring formula and dynamic weight adjustment (exponential smoothing method) are used respectively:

[0076] Real-time change monitoring, label change rate:

[0077] Δt=|v t -v t-1 ∣ / v t-1 ;

[0078] where v t is the value of the label at time t.

[0079] Dynamic weight adjustment and image attribute update:

[0080] New Value=αOld Value+(1-α)·Current Value;

[0081] Where α is the smoothing coefficient, which is usually between 0.1 and 0.3.

[0082] The system uses trigger user data request instructions to collect user registration data, behavior data and geographic location data respectively; divide the label system into levels, clean and pre-process the data, extract user features, and generate labels based on the extracted features; allocate data objects, associate labels with data objects, and generate user portraits with multi-dimensional labels; monitor changes in each layer of labels separately, and update user portraits in real time; by integrating registration, behavior and geographic location data, a more comprehensive user portrait can be constructed. This combination of multi-source data can make up for the limitations of a single data source; the hierarchical label system can more accurately segment users, and accurate user portraits can help provide personalized services to users; the structured design of the label system can improve data processing and retrieval efficiency; hierarchical labels enable faster indexing and return of results when querying specific user groups, optimizing resource utilization.

[0083] S200: Obtain news content features, combine them with user portraits to predict news dissemination paths, and stratify the dissemination paths.

[0084] In this embodiment, news content features are obtained, news propagation paths are predicted based on user profiles, and the propagation paths are layered. The specific process is as follows:

[0085] S201: Acquire news content data, analyze the news content features, match the news content features with the user profile, and match the user's preferred news;

[0086] S202: Outputting the news dissemination path based on the matching results;

[0087] S203: Layering the propagation path; the path includes a core layer, an extension layer, and a peripheral layer;

[0088] S204: Provide a discrimination layer and a monitoring layer for the propagation path.

[0089] In the above process, news content data is obtained and its features are analyzed. The TF-IDF algorithm is used to extract the top-N keywords from the news content features:

[0090]

[0091] Match news content features with user portraits to match user preferred news.

[0092] According to the matching results, the news propagation path is output;

[0093] Node definition: users as nodes;

[0094] The SIR model is used to simulate the propagation process:

[0095] Infection rate: β = forwarding probability × social influence;

[0096] Recovery rate: γ = 1 / news life cycle (usually 72 hours).

[0097] Calculate key figures:

[0098] Transmission rate: v = Δ number of infected users / Δt;

[0099] Coverage breadth: C = number of independent users reached / total number of users.

[0100] The propagation path is layered; the path includes a core layer, an extension layer, and a peripheral layer; and a discrimination layer and a monitoring layer are provided for the propagation path.

[0101] The core layer is used to provide news content to users with stable preferences;

[0102] The extension layer is used to provide news content to users with interest preferences;

[0103] The peripheral layer is used to provide news content to users in the local area;

[0104] The discrimination layer is used to match news content features with updated user profiles and output discrimination keywords;

[0105] The monitoring layer is used to obtain the identification keywords and operate the transmission status of the propagation path.

[0106] S300: Extract discriminant keywords from the dissemination content, match them with the user portrait, and monitor the termination trigger of the dissemination based on the matching situation.

[0107] In this embodiment, the discriminant keywords are extracted from the dissemination content, matched with the user profile, and the termination trigger of the dissemination is monitored based on the matching situation. The specific process is:

[0108] S301: Real-time monitoring of news dissemination data, obtaining news content features, extracting discriminant keywords based on the news content features, and matching the discriminant keywords with the updated user profile;

[0109] S302: Setting a similarity threshold, calculating a first trigger condition value based on the matching result, and outputting a first trigger result;

[0110] S303: Setting a blacklist trigger threshold, calculating a second trigger condition value according to the matching result, and outputting a second trigger result;

[0111] S304: Execute a propagation termination trigger operation according to the first trigger result and the second trigger result.

[0112] In this process, news content data is acquired in real time through crawler technology. News dissemination data includes the source, time, and scope of dissemination. The TF-IDF or TextRank algorithms are used to identify discriminative keywords that represent the core content of the news. Based on the characteristics of the news content, discriminative keywords are extracted and matched with the updated user profiles. A similarity threshold is set and cosine similarity is calculated. The similarity threshold can be set to 0.3. Based on the matching results, the first trigger condition value is calculated and the first trigger result is output. If the first trigger condition value is less than the similarity threshold, the news dissemination is terminated.

[0113] Set the blacklist trigger threshold to query whether the blacklist content is triggered. The blacklist trigger threshold can be set to a logical value of 0 or 1, and the second trigger condition value is calculated based on the matching result, and the second trigger result is output; when the second trigger result is that the second trigger condition value is equal to the blacklist trigger threshold, the news dissemination is terminated.

[0114] A propagation termination trigger operation is performed according to the first trigger result and the second trigger result.

[0115] S400: Acquire a propagation termination trigger condition, request a data propagation channel according to the propagation termination trigger condition, and end the propagation process.

[0116] In this embodiment, news dissemination data is first monitored in real time, and news content features are obtained. Based on the news content features, discriminant keywords are extracted and matched with the updated user profile. A similarity threshold is set, a first trigger condition value is calculated based on the matching results, and a first trigger result is output. A blacklist trigger threshold is set, a second trigger condition value is calculated based on the matching results, and a second trigger result is output. The dissemination termination trigger condition is obtained, and a data dissemination channel is requested based on the dissemination termination trigger condition. The data dissemination channel includes terminating dissemination and continuing dissemination.

[0117] In the present invention, user registration data, behavior data, and geographic location data are collected to establish a hierarchical label system, and a multi-dimensional label user portrait is generated for the user data. The specific process is as follows: triggering a user data request instruction, collecting user registration data, behavior data, and geographic location data respectively; dividing the label system layers, cleaning and pre-processing the data, extracting user features, and generating labels based on the extracted features; allocating data objects, associating labels with data objects, and generating a multi-dimensional label user portrait; monitoring changes in each layer of labels, and updating the user portrait in real time. Acquiring news content features, combining user portraits to predict news propagation paths, and layering the propagation paths; the specific process is as follows: acquiring news content data, analyzing the news content features, matching the news content features with user portraits, and matching user preferred news; outputting news propagation paths based on the matching results; layering the propagation paths; wherein the paths include a core layer, an extension layer, and a peripheral layer; providing a discriminant layer and a monitoring layer for the propagation paths. Extract discriminant keywords from disseminated content, match them with user profiles, and monitor dissemination termination triggers based on the matching results. The specific process is as follows: real-time monitoring of news dissemination data, obtaining news content features, extracting discriminant keywords based on the news content features, and matching these keywords with updated user profiles; setting a similarity threshold, calculating a first trigger condition value based on the matching results, and outputting a first trigger result; setting a blacklist trigger threshold, calculating a second trigger condition value based on the matching results, and outputting a second trigger result; and executing a dissemination termination trigger operation based on the first and second trigger results. Dissemination termination trigger status is obtained, and based on this, a data dissemination channel is requested, terminating the dissemination process. Based on user usage data, a user profile is constructed, and the news dissemination path is predicted. The dissemination path is layered, and the monitoring layer uses the discriminant keywords from the data to obtain the discriminant keywords from the discriminant layer. Dissemination termination triggers are monitored and requested based on the data dissemination channel. By discriminating data based on user profiles during news dissemination, the discriminant keywords are used as a user news blacklist, terminating the data dissemination channel in a timely manner, and improving the accuracy of news dissemination.

[0118] Corresponding to the aforementioned embodiment of the news dissemination method based on big data processing, the present application also provides an embodiment of a news dissemination system based on big data processing.

[0119] Figure 5 This is a block diagram of a news dissemination system based on big data processing according to an exemplary embodiment. Figure 5 The system may include: a user portrait construction module 501, a news dissemination path prediction module 502, a monitoring and discrimination coordination module 503, and a data dissemination channel control module 504; wherein:

[0120] The user profile building module 501 is used to collect user registration data, behavior data and geographic location data, establish a hierarchical label system, and generate a multi-dimensional label user profile based on the user data;

[0121] The news propagation path prediction module 502 is used to obtain news content features, predict news propagation paths based on user portraits, and perform path stratification on the propagation paths;

[0122] The monitoring and discrimination collaborative module 503 is used to extract discrimination keywords from the dissemination content, match them with the user profile, and monitor the termination trigger of the dissemination based on the matching situation;

[0123] The data propagation channel control module 504 is used to obtain a propagation termination trigger condition, request a data propagation channel according to the propagation termination trigger condition, and terminate the propagation process.

[0124] In this embodiment, the user profile construction module 501 collects user registration data, behavior data, and geographic location data, establishes a hierarchical label system, and generates a multi-dimensional label user profile for the user data. The specific process is as follows: triggering a user data request instruction, collecting user registration data, behavior data, and geographic location data respectively; dividing the label system layers, cleaning and preprocessing the data, extracting user features, and generating labels based on the extracted features; assigning data objects, associating labels with data objects, and generating a multi-dimensional label user profile; monitoring changes in each layer of labels, and updating the user profile in real time. The news propagation path prediction module 502 obtains news content features, predicts the news propagation path based on the user profile, and stratifies the propagation path. The specific process is as follows: obtaining news content data, analyzing the news content features, matching the news content features with the user profile, and matching the user's preferred news; outputting the news propagation path based on the matching results; stratifying the propagation path; wherein the path includes a core layer, an extension layer, and a peripheral layer; and providing a discriminant layer and a monitoring layer for the propagation path. The monitoring and discrimination collaborative module 503 extracts discriminative keywords from the dissemination content, matches them with the user profile, and monitors the termination trigger of dissemination based on the matching results. The specific process is as follows: real-time monitoring of news dissemination data, obtaining news content features, extracting discriminative keywords based on the news content features, and matching the discriminative keywords with the updated user profile; setting a similarity threshold, calculating a first trigger condition value based on the matching results, and outputting a first trigger result; setting a blacklist trigger threshold, calculating a second trigger condition value based on the matching results, and outputting a second trigger result; and executing a dissemination termination trigger operation based on the first and second trigger results. The data dissemination channel control module 504 obtains the dissemination termination trigger, requests a data dissemination channel based on the dissemination termination trigger, and terminates the dissemination process. Based on user usage data, a user profile is constructed, and the news dissemination path is predicted. The dissemination path is layered, and the monitoring layer is used to obtain the discriminative keywords for the data from the discriminative layer. The dissemination termination trigger is monitored and a data dissemination channel is requested. By discriminating the data based on the user profile during news dissemination, the discriminative keywords are used as the user news blacklist, and the data dissemination channel is terminated in a timely manner, thereby improving the accuracy of news dissemination.

[0125] Regarding the system in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0126] For the system embodiment, since it basically corresponds to the method embodiment, the relevant parts can be referred to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this application. A person of ordinary skill in the art can understand and implement it without paying any creative work.

[0127] Accordingly, the present application also provides an electronic device, comprising: one or more processors; a memory for storing one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the above-mentioned news dissemination method based on big data processing. Figure 6 As shown in the figure, a hardware structure diagram of a news dissemination system based on big data processing provided by an embodiment of the present invention is provided, in which any device with data processing capability is provided. Figure 6 In addition to the processor, memory, and network interface shown, any device with data processing capabilities in which the apparatus in the embodiment is located may also include other hardware, generally based on the actual functions of the device with data processing capabilities, which will not be described in detail.

[0128] Accordingly, the present application also provides a computer-readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the news dissemination method based on big data processing as described above. The computer-readable storage medium may be an internal storage unit of any device with data processing capabilities as described in any of the aforementioned embodiments, such as a hard disk or memory. The computer-readable storage medium may also be an external storage device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), an SD card, a flash card (Flash Card), etc. equipped on the device. Furthermore, the computer-readable storage medium may also include both an internal storage unit and an external storage device of any device with data processing capabilities. The computer-readable storage medium is used to store the computer program and other programs and data required by any device with data processing capabilities, and may also be used to temporarily store data that has been output or is to be output.

[0129] Those skilled in the art will readily conceive of other embodiments of the present application after considering the specification and practicing the contents disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of this application and include common knowledge or customary techniques in the art that are not disclosed in this application.

[0130] It will be understood that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.

Claims

1. A news dissemination method based on big data processing, characterized in that: The steps include: Collect user registration data, behavior data, and geographic location data, establish a hierarchical tag system, and generate multi-dimensional tag user portraits based on user data; Obtain news content features, combine them with user portraits to predict news dissemination paths, and stratify the dissemination paths; Extract discriminant keywords from the dissemination content, match them with user portraits, and monitor the termination trigger of the dissemination based on the matching results; Obtain the propagation termination trigger condition, request the data propagation channel according to the propagation termination trigger condition, and end the propagation process.

2. The news dissemination method based on big data processing according to claim 1, characterized in that: In the steps of collecting user registration data, behavior data, and geographic location data, establishing a hierarchical tag system, and generating multi-dimensional tag user portraits based on user data: Trigger user data request instructions to collect user registration data, behavior data and geographic location data respectively; Divide the label system into levels, clean and pre-process the data, extract user features, and generate labels based on the extracted features; Assign data objects, associate tags with data objects, and generate user portraits with multi-dimensional tags.

3. The news dissemination method based on big data processing according to claim 2, characterized in that: After assigning data objects, associating tags with data objects, and generating user profiles with multi-dimensional tags: Monitor label changes at each layer separately and update user profiles in real time.

4. The news dissemination method based on big data processing according to claim 3, characterized in that: In the steps of obtaining news content features, predicting news dissemination paths based on user profiles, and stratifying the dissemination paths: Acquire news content data, analyze the news content features, match the news content features with user profiles, and match user preferred news; According to the matching results, the news propagation path is output; Stratify the propagation paths; The path includes the core layer, the extension layer, and the peripheral layer.

5. The news dissemination method based on big data processing according to claim 4, characterized in that: After the step of stratifying the propagation paths: Provides discrimination layer and monitoring layer for propagation path.

6. The news dissemination method based on big data processing according to claim 5, characterized in that: The core layer is used to provide news content to users with stable preferences; The extension layer is used to provide news content to users with interest preferences; The peripheral layer is used to provide news content to users in the local area; The discrimination layer is used to match news content features with updated user profiles and output discrimination keywords; The monitoring layer is used to obtain the identification keywords and operate the transmission status of the propagation path.

7. The news dissemination method based on big data processing according to claim 6, characterized in that: In the steps of extracting discriminant keywords from the dissemination content, matching them with user portraits, and monitoring the termination trigger of the dissemination based on the matching situation: Monitor news dissemination data in real time, obtain news content features, extract discriminant keywords based on news content features, and match the discriminant keywords with the updated user profile; Set a similarity threshold, calculate a first trigger condition value based on the matching result, and output a first trigger result; Set the blacklist trigger threshold, calculate the second trigger condition value based on the matching result, and output the second trigger result; A propagation termination trigger operation is performed according to the first trigger result and the second trigger result.

8. The news dissemination method based on big data processing according to claim 7, characterized in that: In the steps of setting the similarity threshold, calculating the first trigger condition value according to the matching result, and outputting the first trigger result: When the first trigger result is that the first trigger condition value is less than the similarity threshold, the news dissemination is terminated.

9. The news dissemination method based on big data processing according to claim 7, characterized in that: In the steps of setting the blacklist trigger threshold, calculating the second trigger condition value based on the matching result, and outputting the second trigger result: When the second trigger result is that the second trigger condition value is equal to the blacklist trigger threshold, the news dissemination is terminated.

10. A news dissemination system based on big data processing, applied to the news dissemination method based on big data processing according to claim 1, characterized in that: It includes a user portrait construction module, a news dissemination path prediction module, a monitoring and discrimination collaboration module, and a data dissemination channel control module; among which: The user portrait construction module is used to collect user registration data, behavior data and geographic location data, establish a hierarchical label system, and generate a multi-dimensional label user portrait based on the user data; The news propagation path prediction module is used to obtain news content features, predict news propagation paths based on user portraits, and perform path stratification on the propagation paths; The monitoring and discrimination collaborative module is used to extract discrimination keywords from the dissemination content, match them with the user profile, and monitor the termination trigger of the dissemination based on the matching situation; The data propagation channel control module is used to obtain a propagation termination trigger condition, request a data propagation channel according to the propagation termination trigger condition, and end the propagation process.

Images