Social platform opinion optimization system and method based on alliance chain and text classification

Abstract

The application discloses a social platform public opinion optimization system and method based on a consortium chain and text classification, which stores platform user published information in a decentralized manner based on a Fabric consortium chain, guarantees user privacy security, and returns value to individuals. On the other hand, user behavior is recorded on the blockchain to generate tamper-proof behavior history records, supplemented by the characteristics of blockchain record traceability and automatic execution of smart contracts, and the ability to automatically trace and hold accountable low-quality information in the social platform can be realized. The application uses the Fabric network to store We Media information and user comments, article and other information, and through the deployment and execution of the smart contract, the control and quality optimization of the social platform public opinion can be realized on the basis of guaranteeing user privacy. At the same time, the text classification technology based on machine learning is adopted to identify and analyze the information such as user published topics and comments, so that the filtering processing of low-quality information is realized, and a social platform with an optimized public opinion environment is constructed.

Description

Technical Field

[0001] This invention belongs to the field of cross-application technology combining blockchain and machine learning, and specifically relates to a method and system for optimizing public opinion on social platforms based on consortium blockchain and text classification. Background Technology

[0002] Driven by mobile networks and Web 2.0 technologies, social media has experienced unprecedented growth globally. This evolution has transformed how information is disseminated and gradually formed an information ecosystem centered around it. After nearly a decade of development, as users place increasing emphasis on personal privacy and information security, issues such as uncontrolled user information, rampant misinformation, unequal distribution of benefits, and privacy breaches have become acute pain points for social media development. To fundamentally solve these problems, transformative technological support is needed to achieve decentralized platform operation.

[0003] By using blockchain as a technological breakthrough, a decentralized social media platform is built, offering an innovative solution to the current bottlenecks in social media development. Additionally, a method for identifying low-quality comments in forums is provided, aiming to automatically filter out comments that meet predefined low-quality characteristics, improve the accuracy of forum comment data, and enhance the overall public opinion environment of social media platforms. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for optimizing public opinion on social platforms based on consortium blockchains and text classification, which solves the technical problems of centralized network operation, platform control of user data, proliferation of fake news, deterioration of information flow quality, excessive exposure of personal information, and threats to user privacy in traditional social media platforms.

[0005] To address the problems in traditional social media platforms, the present invention adopts the following technical solution:

[0006] A social platform opinion optimization system based on consortium blockchain and text classification is developed. The system builds a blockchain network for storing core data based on the Hyperledger Fabric consortium blockchain and develops a social platform with multiple types of users on the network. The social platform includes an administrative backend and a community frontend.

[0007] The system's users include system administrators, community administrators, self-media organizations, and individual users;

[0008] The system management users are initialized by default in the system, while community management users are dynamically assigned and generated in the management backend. Self-media organization users and individual users can become system users through two methods: adding in the backend and registering in the frontend. Each type of user corresponds to a specific organization in the blockchain network, thereby realizing permission control at both the network layer and the application layer.

[0009] This system records user behavior on the blockchain, generating an immutable and undeletable historical record. Combined with the traceability of blockchain records and the automatic execution of smart contracts, it enables automatic tracing and accountability for low-quality information on social media platforms. This invention uses a Fabric network to store self-media information, user comments, articles, and other data. Through the deployment and execution of smart contracts, it can control and optimize the quality of public opinion on social media platforms while protecting user privacy. Simultaneously, it employs machine learning-based text classification technology to identify and analyze user-posted topics and comments, filtering out low-quality information to build a social platform with an optimized public opinion environment.

[0010] Further optimizations include a comment section management module, a user permission management module, and a blog management module in the management backend.

[0011] The comment section management module includes a section management submodule, a category management submodule, a post management submodule, and a comment reply management submodule.

[0012] The user permission management module includes a user management submodule, a self-media management submodule, and an individual user management submodule;

[0013] The blog management module includes an article management submodule and a comment management submodule.

[0014] Further optimizations include providing comprehensive forum and blog functions for self-media organizations and individual users, offering convenient channels for platform users to share knowledge and record their lives;

[0015] The community front-end includes user-related function modules, post browsing function modules, post editing function modules, and article management function modules;

[0016] The user-related functional modules include user login, user registration, user settings, user homepage, and private messages;

[0017] The post browsing module includes custom browsing, post search, post favorites, post comments, and post replies;

[0018] The post editing module includes post publishing and post editing;

[0019] The article management module includes article publishing, article editing, article commenting, and article collection.

[0020] Further optimizations include granting the system management user the highest level of management authority over the entire system; this includes role menu management, adding and modifying the other three types of users, and various support functions at the system development level.

[0021] The community management users are administrators who directly control the core services of the system; the social platform functions include forums and blogs; in the forums, community management users control the sections and categories existing in the system and conduct quality reviews of posts and comment replies; in the blogs, community management users review the content published by self-media and user articles to ensure the authenticity and accuracy of information shared on the platform;

[0022] The self-media organizations mentioned above obtain platform incentives by publishing valuable technical articles or promotions on the platform, thereby achieving the goals of self-promotion, profit generation, and popularization of technical knowledge.

[0023] Individual users can ask questions or share knowledge in the technical forum according to their needs, based on the sections and categories. They can also use blog posts to record their technical insights.

[0024] Because consortium blockchains inherently possess MSP member services and Fabric-ca modules, blockchain nodes can be divided into multiple organizations at the network layer according to design requirements: community management organizations, self-media organizations, and individual user organizations. When application layer users access data stored on the blockchain by calling chaincode through the SDK, different types of users connect to channel clients constructed by nodes and users from different organizations. Based on the pre-set permissions in the network configuration, different permission levels for users in different organizations are implemented at the network layer.

[0025] The construction method of a social platform public opinion optimization system based on consortium blockchain and text classification includes the following steps.

[0026] S1: Build the Fabric network and enable blockchain support for applications:

[0027] S11: Deployment of Hyperledger-Fabric via Docker containers:

[0028] S12: Blockchain support for implementing core system functions through smart contracts:

[0029] S2: Use Fabric-java-sdk to establish a connection between the application system and the blockchain network;

[0030] S3: Develop the upper-layer web application system and implement the main functions of the system; use Java language and the Spring Boot framework for the system backend functions; use Bootstrap + Vue to implement the front-end page display and data rendering.

[0031] S4: Introducing Fabric platform components to optimize system performance.

[0032] Further optimization, step S11: the deployment method of Hyperledger-Fabric via Docker containers, includes the following steps:

[0033] a. Pull Fabric image version 2.0 or higher from the Docker Hub online repository;

[0034] b. Write relevant configuration files; configure the organization-related information of the blockchain network, the information of internal nodes and users, and the various policy mechanisms used by the blockchain network. Generate the corresponding blockchain network dependency files through the commands provided by Hyperledger-Fabric, including channel files, genesis block files, blockchain network organization node-related files, and organization node identity authentication information-related files.

[0035] c. Write smart contracts; use Go to write chaincode, introduce relevant APIs into the smart contract and call them in the chaincode interface, and define the storage structure of certificates in the blockchain in the chaincode file;

[0036] d. Write a docker-compose file whose content is related to the blockchain organization's node structure and is used to start the docker image environment that the blockchain network depends on; map the relevant files generated in step b above into the image as runtime dependencies for different images; configure the couchDB database in the docker-compose file to replace Fabric's default levelDB state database.

[0037] e. Start the Docker image environment corresponding to docker-compose. If the image environment starts normally, the blockchain network is successfully built.

[0038] The method uses the Explorer component provided by Hyperledger to view block-related information, and the Caliper component to perform performance evaluation and continuously optimize the system.

[0039] The Explorer project is positioned as a browser for blockchain platforms, allowing users to view, access, deploy, or query blocks, network information, chaincode, transaction sequences, and other related information stored in the blockchain. This method uses Explorer to view block-related information in the backend development, improving system performance.

[0040] Further optimization involves implementing blockchain support for the core system functions through smart contracts in step S12. This involves using Go to write smart contracts, which on one hand implement basic CRUD (Create, Read, Update, Delete) functions for the core data entities stored in the system. Each function is implemented by calling APIs provided by Fabric. The query method relies on the CouchDB state database, defining different query conditions using the query string format specified by CouchDB to achieve the query function. On the other hand, it integrates the basic operations of various entities, implementing corresponding data operations and functional support based on the specific processes of public opinion control, data analysis, and information filtering algorithms.

[0041] This invention utilizes smart contracts written in Go to implement the core functions of the system, including public opinion control. By writing and deploying smart contracts, the system incorporates the public opinion circuit breaker mechanism of existing mainstream online public opinion platforms. Through the introduction of big data analysis mechanisms, it extracts trending topics and analyzes netizens' stances. For events with ongoing controversy, it adjusts the activity coefficient of messages to reduce the probability of related messages appearing, thereby stabilizing online public opinion. By writing and deploying smart contracts, it implements an extreme information filtering algorithm based on sentiment analysis. Most low-quality information exhibits exaggerated language style and abnormal emotional expression. Through methods such as sentence segmentation and sentiment dictionary comparison, it quantifies and analyzes the sentiment indicators of the text, providing a reference for judging information quality. Using an analysis algorithm based on sentiment polarity, the contract layer analyzes the sentiment indicators of messages in real time. For messages with extreme sentiment, it promptly publishes warning information on the blockchain and records the author information, reposting history, and version control information of the message, updating them in real time. Furthermore, the contract can be used to extend more complex control mechanisms.

[0042] Further optimization involves step S2 using the Fabric-java-sdk to establish a connection between the application system and the blockchain network, including the following steps:

[0043] S21: SDK initialization operations:

[0044] After the blockchain network is successfully built for the first time, it needs to be initialized, including channel creation, node joining, chaincode installation, and Fabric-ca registration. The initialization operation is implemented through command line or Fabric-java-sdk. The initialization operation of the blockchain network can be completed by passing specific parameters when starting the application.

[0045] S21: Implementation of SDK-related business logic

[0046] Clients are created for different organizations, users, and nodes, and the SDK-provided interfaces are used to initiate chaincode calls to the blockchain. At this level, methods for supporting key functions such as public opinion control and information filtering are implemented through the combined use of multiple smart contracts.

[0047] Further optimization involves step S3, which develops an upper-layer web application system to implement the system's main functions, including the following steps:

[0048] S31: Develop an RBAC system on the backend using the Spring Security framework; first complete the general basic functions of each type of function, including simple CRUD operations for each entity, and then complete the specific algorithms for data analysis, public opinion control and information filtering.

[0049] For comments, since each section and category has its own content characteristics, either pre-training a corresponding low-quality commenter or online training can be used.

[0050] S32: After the front-end uses Bootstrap and other frameworks to write the page, Vue is introduced to make requests to the back-end interface to obtain data for page rendering, realizing the separation of front-end and back-end, and displaying different interfaces according to different user types.

[0051] Further optimization involves introducing Fabric platform components to improve system performance in step S4, including the following steps:

[0052] S41: Integrates the Fabric blockchain explorer to monitor network conditions in real time, such as information on blocks, transactions, nodes, organizations, and chaincode, achieving system traceability and tamper-proof features;

[0053] S42: The system uses the Caliper component provided by Hyperledger for performance evaluation and continuous system optimization. Through blockchain performance indicators such as TPS, latency, and system resource consumption generated by Caliper, the system is continuously tested and parameters are adjusted during development to ensure that the system meets development requirements.

[0054] Caliper is a blockchain benchmarking framework. Its key component is the adapter layer, which integrates different blockchain implementations with the Caliper framework. An adapter is implemented for each test environment, translating Caliper's northbound interface into the protocol of the corresponding blockchain implementation. Caliper's northbound interface is a series of general blockchain interfaces, encompassing various operations such as installing smart contracts, calling smart contracts, and querying ledger status during interaction with the backend blockchain system.

[0055] This invention also employs an identification method for effective filtering of low-quality comments. The method is specifically described as follows: selecting and labeling samples; generating various strong feature classifiers to identify the characteristics of each low-quality comment; forming a strong comment classifier based on these strong feature classifiers; and identifying low-quality comments based on this strong comment classifier.

[0056] The beneficial results of this invention are as follows:

[0057] This method establishes a social platform for analyzing and exchanging high-quality knowledge based on the Hyperledger Fabric consortium blockchain. The system records user behavior on the blockchain, generating an immutable and undeletable historical record. Combined with the traceability of blockchain records and the automatic execution of smart contracts, it enables automatic tracing and accountability for low-quality information on the social platform. Simultaneously, it employs machine learning-based text classification technology to identify and analyze user-posted topics and comments, filtering out low-quality information. Attached Figure Description

[0058] Figure 1 This is a functional module diagram of the social platform public opinion optimization system based on consortium blockchain and text classification as described in this invention;

[0059] Figure 2 This is a technical architecture diagram of the social platform public opinion optimization system based on consortium blockchain and text classification as described in this invention;

[0060] Figure 3 These are screenshots of some pages of the system described in this invention;

[0061] Figure 4 This is a screenshot of a portion of the system described in this invention. Detailed Implementation

[0062] To make the objectives and technical solutions of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with the embodiments of this invention.

[0063] Example 1:

[0064] like Figure 1-4 As shown, a social platform opinion optimization system based on consortium blockchain and text classification is built on a blockchain network for storing core data based on Hyperledger Fabric consortium blockchain. On the basis of the network, a social platform with multiple types of users is developed. The social platform includes an administrative backend and a community frontend.

[0065] The system's users include system administrators, community administrators, self-media organizations, and individual users;

[0066] The system management users are initialized by default in the system, while community management users are dynamically assigned and generated in the management backend. Self-media organization users and individual users can become system users through two methods: adding in the backend and registering in the frontend. Each type of user corresponds to a specific organization in the blockchain network, thereby realizing permission control at both the network layer and the application layer.

[0067] This system records user behavior on the blockchain, generating an immutable and undeletable historical record. Combined with the traceability of blockchain records and the automatic execution of smart contracts, it enables automatic tracing and accountability for low-quality information on social media platforms. This invention uses a Fabric network to store self-media information, user comments, articles, and other data. Through the deployment and execution of smart contracts, it can control and optimize the quality of public opinion on social media platforms while protecting user privacy. Simultaneously, it employs machine learning-based text classification technology to identify and analyze user-posted topics and comments, filtering out low-quality information to build a social platform with an optimized public opinion environment.

[0068] In this embodiment, the management backend includes a comment section management module, a user permission management module, and a blog management module;

[0069] The comment section management module includes a section management submodule, a category management submodule, a post management submodule, and a comment reply management submodule.

[0070] The user permission management module includes a user management submodule, a self-media management submodule, and an individual user management submodule;

[0071] The blog management module includes an article management submodule and a comment management submodule.

[0072] In this embodiment, the community front-end provides comprehensive forum and blog functions for self-media organizations and individual users, providing a convenient channel for platform users to share knowledge and record their lives;

[0073] The community front-end includes user-related function modules, post browsing function modules, post editing function modules, and article management function modules;

[0074] The user-related functional modules include user login, user registration, user settings, user homepage, and private messages;

[0075] The post browsing module includes custom browsing, post search, post favorites, post comments, and post replies;

[0076] The post editing module includes post publishing and post editing;

[0077] The article management module includes article publishing, article editing, article commenting, and article collection.

[0078] In this embodiment, the system management user has the highest level of management authority over the entire system, including role menu management, adding and modifying the other three types of users, and various support functions at the system development level.

[0079] The community management users are administrators who directly control the core services of the system; the social platform functions include forums and blogs; in the forums, community management users control the sections and categories existing in the system and conduct quality reviews of posts and comment replies; in the blogs, community management users review the content published by self-media and user articles to ensure the authenticity and accuracy of information shared on the platform;

[0080] The self-media organizations mentioned above obtain platform incentives by publishing valuable technical articles or promotions on the platform, thereby achieving the goals of self-promotion, profit generation, and popularization of technical knowledge.

[0081] Individual users can ask questions or share knowledge in the technical forum according to their needs, based on the sections and categories. They can also use blog posts to record their technical insights.

[0082] Because consortium blockchains inherently possess MSP member services and Fabric-ca modules, blockchain nodes can be divided into multiple organizations at the network layer according to design requirements: community management organizations, self-media organizations, and individual user organizations. When application layer users access data stored on the blockchain by calling chaincode through the SDK, different types of users connect to channel clients constructed by nodes and users from different organizations. Based on the pre-set permissions in the network configuration, different permission levels for users in different organizations are implemented at the network layer.

[0083] Example 2:

[0084] The construction method of a social platform public opinion optimization system based on consortium blockchain and text classification includes the following steps:

[0085] S1: Build the Fabric network and enable blockchain support for applications:

[0086] S11: Deployment of Hyperledger-Fabric via Docker containers:

[0087] a. Pull Fabric image version 2.0 or higher from the Docker Hub online repository;

[0088] b. Write relevant configuration files; configure the organization-related information of the blockchain network, the information of internal nodes and users, and the various policy mechanisms used by the blockchain network. Generate the corresponding blockchain network dependency files through the commands provided by Hyperledger-Fabric, including channel files, genesis block files, blockchain network organization node-related files, and organization node identity authentication information-related files.

[0089] c. Write smart contracts; use Go to write chaincode, introduce relevant APIs into the smart contract and call them in the chaincode interface, and define the storage structure of certificates in the blockchain in the chaincode file;

[0090] d. Write a docker-compose file whose content is related to the blockchain organization's node structure and is used to start the docker image environment that the blockchain network depends on; map the relevant files generated in step b above into the image as runtime dependencies for different images; configure the couchDB database in the docker-compose file to replace Fabric's default levelDB state database.

[0091] e. Start the Docker image environment corresponding to docker-compose. If the image environment starts normally, the blockchain network is successfully built.

[0092] S12: Blockchain support for implementing core system functions through smart contracts:

[0093] Using Go to write smart contracts, on the one hand, we can implement basic CRUD (Create, Read, Update, Delete) functions for the core data entities stored in the system. Each function is implemented by calling the API provided by Fabric. The query method relies on the CouchDB state database. Different query conditions are defined by the query string format specified by CouchDB to achieve the query function. On the other hand, we can integrate the basic operations of various entities and implement corresponding data operations and functional support according to the specific processes of public opinion control, data analysis, and information filtering algorithms.

[0094] S2: Using Fabric-java-sdk, establish the connection between the application system and the blockchain network:

[0095] S21: SDK initialization operations:

[0096] After the blockchain network is successfully built for the first time, it needs to be initialized, including channel creation, node joining, chaincode installation, and Fabric-ca registration. The initialization operation is implemented through command line or Fabric-java-sdk. The initialization operation of the blockchain network can be completed by passing specific parameters when starting the application.

[0097] S21: Implementation of SDK-related business logic

[0098] Clients are created for different organizations, users, and nodes, and the SDK-provided interfaces are used to initiate chaincode calls to the blockchain. At this level, methods for supporting key functions such as public opinion control and information filtering are implemented through the combined use of multiple smart contracts.

[0099] S3: Develop the upper-layer web application system and implement the main functions of the system:

[0100] The system backend functionality is implemented using the Java language and the Spring Boot framework; the frontend page display and data rendering are implemented using Bootstrap and Vue.

[0101] S31: Develop an RBAC system on the backend using the Spring Security framework; first complete the general basic functions of each type of function, including simple CRUD operations for each entity, and then complete the specific algorithms for data analysis, public opinion control and information filtering.

[0102] For comments, since each section and category has its own content characteristics, either pre-training a corresponding low-quality commenter or online training can be used.

[0103] S32: After the front-end uses Bootstrap and other frameworks to write the page, Vue is introduced to make a request to the back-end interface to obtain data for page rendering, realizing the separation of front-end and back-end, and displaying different interfaces according to different user types;

[0104] S4: Introducing Fabric platform components to optimize system performance:

[0105] S41: Integrates the Fabric blockchain explorer to monitor network conditions in real time, such as information on blocks, transactions, nodes, organizations, and chaincode, achieving system traceability and tamper-proof features;

[0106] S42: The system uses the Caliper component provided by Hyperledger for performance evaluation and continuous system optimization. Through blockchain performance indicators such as TPS, latency, and system resource consumption generated by Caliper, the system is continuously tested and parameters are adjusted during development to ensure that the system meets development requirements.

[0107] The above embodiments are only for illustrating the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They should not be construed as limiting the scope of protection of the present invention. All equivalent changes or modifications made in accordance with the spirit and essence of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A social platform public opinion optimization system based on consortium blockchain and text classification, characterized by: A blockchain network for storing core data is built on the Hyperledger Fabric consortium blockchain. On the basis of the network, a social platform with multiple types of users is developed. The social platform includes an administrative backend and a community frontend. The system's users include system administrators, community administrators, self-media organizations, and individual users; The system management users are initialized by default in the system, while community management users are dynamically assigned and generated in the management backend. Self-media organization users and individual users can become system users through two methods: adding in the backend and registering in the frontend. Each type of user corresponds to a specific organization in the blockchain network, thereby realizing access control at both the network layer and the application layer. The system also includes a network built on the Hyperledger-Fabric consortium blockchain, smart contracts deployed on the Fabric network, a Fabric-java-sdk, an upper-layer web application system, and Fabric platform components. The Fabric network is deployed via Docker containers and enables blockchain application support. The smart contracts are used to implement blockchain support for the system's core functions. The Fabric-java-sdk is used to establish the connection between the application system and the Fabric network. The upper-layer web application system uses Java and the Spring Boot framework to develop the system's backend functions; it uses Bootstrap + Vue to implement the frontend page display and data rendering. The Fabric platform components are used to optimize system performance. The upper-layer web application system is used to implement the following functions: Develop an RBAC system on the backend using the Spring Security framework; first complete the common basic functions of each type of function, including simple CRUD operations for each entity, and then complete the specific algorithms for data analysis, public opinion control and information filtering. For comments, since each section and category has its own content characteristics, either pre-training a corresponding low-quality commenter or online training can be used. After writing the front-end page using Bootstrap in conjunction with other frameworks, Vue is introduced to initiate requests to the back-end interface to obtain data for page rendering, thus achieving front-end and back-end separation and displaying different interfaces according to different user types.

2. The social platform public opinion optimization system based on consortium blockchain and text classification according to claim 1, characterized in that: The management backend includes a comment section management module, a user permission management module, and a blog management module; The comment section management module includes a section management submodule, a category management submodule, a post management submodule, and a comment reply management submodule. The user permission management module includes a user management submodule, a self-media management submodule, and an individual user management submodule; The blog management module includes an article management submodule and a comment management submodule.

3. The social platform public opinion optimization system based on consortium blockchain and text classification according to claim 1, characterized in that: The community front end provides comprehensive forum and blog functions for self-media organizations and individual users, offering a convenient channel for platform users to share knowledge and record their lives; The community front-end includes user-related function modules, post browsing function modules, post editing function modules, and article management function modules; The user-related functional modules include user login, user registration, user settings, user homepage, and private messages; The post browsing module includes custom browsing, post search, post favorites, post comments, and post replies; The post editing module includes post publishing and post editing; The article management module includes article publishing, article editing, article commenting, and article collection.

4. The social platform public opinion optimization system based on consortium blockchain and text classification according to claim 1, characterized in that: The system management user has the highest level of management authority over the entire system, including role menu management, adding and modifying three types of users, and various support functions at the system development level. The community management users are administrators who directly control the core services of the system; the social platform functions include forums and blogs; in the forums, community management users control the sections and categories existing in the system and conduct quality reviews of posts and comment replies; in the blogs, community management users review the content published by self-media and user articles to ensure the authenticity and accuracy of information shared on the platform; The self-media organizations mentioned above obtain platform incentives by publishing valuable technical articles or promotions on the platform, thereby achieving the goals of self-promotion, profit generation, and popularization of technical knowledge. Individual users can ask questions or share knowledge in the technical forum according to their needs, based on the sections and categories. They can also use blog posts to record their technical insights.

5. The method for constructing a social platform public opinion optimization system based on consortium blockchain and text classification according to any one of claims 1-4, characterized in that: Includes the following steps S1: Build the Fabric network and enable blockchain support for applications: S11: Deploying Hyperledger-Fabric via Docker containers: S12: Blockchain support for implementing core system functions through smart contracts: S2: Use Fabric-java-sdk to establish a connection between the application system and the blockchain network; S3: Develop the upper-layer web application system and implement system functions; use Java and the Spring Boot framework to develop the system backend functions; use Bootstrap + Vue to implement the front-end page display and data rendering; S4: Introducing Fabric platform components to optimize system performance.

6. The method for constructing a social platform public opinion optimization system based on consortium blockchain and text classification according to claim 5, characterized in that: Step S11: The Hyperledger-Fabric deployment method using Docker containers includes the following steps: a. Pull Fabric image version 2.0 or higher from the Docker Hub online repository; b. Write relevant configuration files; configure the organization-related information of the blockchain network, the information of internal nodes and users, and the various policy mechanisms used by the blockchain network. Generate the corresponding blockchain network dependency files through the commands provided by Hyperledger-Fabric, including channel files, genesis block files, blockchain network organization node-related files, and organization node identity authentication information-related files. c. Write smart contracts; use Go to write chaincode, introduce relevant APIs into the smart contract and call them in the chaincode interface, and define the storage structure of certificates in the blockchain in the chaincode file; d. Write a docker-compose file whose content is related to the blockchain organization's node structure and is used to start the docker image environment that the blockchain network depends on; map the relevant files generated in step b above into the image as runtime dependencies for different images; configure the couchDB database in the docker-compose file to replace Fabric's default levelDB state database. e. Start the Docker image environment corresponding to docker-compose. If the image environment starts normally, the blockchain network is successfully built.

7. The method for constructing a social platform public opinion optimization system based on consortium blockchain and text classification according to claim 5, characterized in that: Step S12 describes a blockchain-supported method for implementing the core functions of the system through smart contracts: Smart contracts are written in Go to implement basic CRUD (Create, Read, Update, Delete) functions for the core data entities stored in the system. Each function is implemented by calling APIs provided by Fabric. The query method relies on the CouchDB state database, defining different query conditions using the query string format specified by CouchDB to achieve the query function. The system integrates basic operations on various entities, implementing corresponding data operations and functional support based on the specific processes of public opinion control, data analysis, and information filtering algorithms.

8. The method for constructing a social platform public opinion optimization system based on consortium blockchain and text classification according to claim 5, characterized in that: Step S2 uses Fabric-java-sdk to establish a connection between the application system and the blockchain network, including the following steps: S21: SDK initialization operations: After the blockchain network is successfully built for the first time, it needs to be initialized, including channel creation, node joining, chaincode installation, and Fabric-ca registration. The initialization operation is implemented through command line or Fabric-java-sdk. The initialization operation of the blockchain network can be completed by passing specific parameters when starting the application. S22: SDK-related business logic implementation: Create corresponding clients based on different organizations, users, and nodes, and use the interfaces provided by the SDK to initiate chaincode call requests for blockchain operations; implement key functions such as public opinion control and information filtering by combining multiple smart contracts.

9. The method for constructing a social platform public opinion optimization system based on consortium blockchain and text classification according to claim 5, characterized in that: Step S4 introduces Fabric platform components to optimize system performance, including the following steps: S41: Integrates a Fabric blockchain explorer for real-time monitoring of network conditions; S42: The performance evaluation is carried out using the Caliper component provided by Hyperledger, and the system is continuously optimized. The blockchain performance indicators generated by Caliper include TPS, latency, and system resource consumption. During development, the parameters are continuously tested and adjusted to ensure that the system meets the development requirements.

Images