Digital economic information management method and system based on big data analysis

Abstract

The invention provides a digital economic information management method and system based on big data analysis, and the method comprises the steps: integrating the data of a consumption end, a production end and a server end through a unified platform, guaranteeing the transmission safety through the combination of differential privacy and TLS 1.3, recording the full life cycle of data through a segmented block chain, supporting the dynamic authorization and fine-grained authority control, and improving the efficiency. The method comprises the following steps: verifying data purposes by combining zero-knowledge proof, fusing NLP, image processing and time sequence analysis, constructing a knowledge graph, generating dynamic tags and composite features, simulating decision influence through anti-fact analysis, supporting generation of a Pareto optimal solution by an interactive decision tree, and providing a market trend thermodynamic diagram and a supply chain risk graph. Customized suggestions are pushed to an ERP system in real time, sensitive data are protected based on role permission and lightweight homomorphic encryption, abnormal behavior monitoring and zero-knowledge proof compliance review are combined, a service scene is simulated through an A / B sandbox, and low-code community drives demand iteration.

Description

Technical Field

[0001] This invention relates to the field of digital economy information big data management technology, and in particular to a digital economy information management method and system based on big data analysis. Background Technology

[0002] The current digital economy faces core challenges such as data silos, low circulation efficiency, high security risks, and insufficient intelligent decision-making. Existing technologies often rely on single channels for data collection, such as traditional ERP systems or IoT devices, making it difficult to integrate multimodal data from the consumer, production, and service ends, including text, images, and time series. This leads to fragmented data governance, a lack of full lifecycle tracking for data sharing and use, and rigid authorization mechanisms such as static permission allocation, which are difficult to dynamically adapt to complex business scenarios. Furthermore, traditional data cleaning technologies lack the ability to semantically correlate cross-modal data, making it difficult to extract composite features to support in-depth analysis. Moreover, data security relies on centralized encryption, posing a risk of sensitive information leakage.

[0003] Existing predictive models are mostly based on single-dimensional data such as transaction records, making it difficult to simulate the synergistic effects of multiple factors, including pricing strategies and supply chain adjustments. Decision support lacks real-time capabilities and the ability to balance conflicting objectives, such as cost versus carbon emissions. Data visualization tools are mostly static reports, unable to dynamically respond to real-time data fluctuations, such as sudden increases in supply chain latency, and lack deep integration with business processes. Although access control technologies exist in the security field, role-based access management is disconnected from business rules, making it difficult to achieve fine-grained dynamic authorization. Furthermore, encryption schemes cannot balance privacy protection and computational efficiency.

[0004] To address the aforementioned shortcomings, this method utilizes technologies such as blockchain full lifecycle management, cross-modal cleaning and knowledge graph semantic association, lightweight homomorphic encryption and zero-knowledge proof to construct a data closed-loop governance and intelligent decision-making system, thereby solving the bottleneck problems of data circulation, security and decision-making efficiency. Summary of the Invention

[0005] The purpose of this invention is to provide a digital economy information management method and system based on big data analysis.

[0006] The problem this invention aims to solve is: the difficulty in integrating multi-source heterogeneous data, insufficient data security and compliance, delayed decision-making response, and poor adaptability to business scenarios in the digital economy.

[0007] A digital economy information management method based on big data analysis, employing the following technical solution: S1: Collect data related to the digital economy, including user behavior data, transaction records, supply chain data, market sentiment, and IoT device data, covering the consumer, production, and service ends; S2: Uses blockchain technology to record the entire lifecycle, including the entire lifecycle of data collection, sharing and use, and adopts dynamic authorization, including restrictions on access frequency and usage. Data providers adjust the scope of authorization according to digital economy rules. S3: Construct a cross-modal cleaning model to jointly process text, image and time series data, formulate data standardization rules specific to the digital economy, introduce knowledge graphs for semantic association, dynamically generate data tags based on business needs, and extract composite features; S4: Based on the data processed in S3, train a digital economy-specific prediction model to simulate the impact of different decision-making schemes on the digital economy; S5: Develop customized data dashboards, including market trend heatmaps and supply chain risk maps, generate customized suggestions for the digital economy based on analysis results, and embed decision-making suggestions into digital economy business processes; S6: Assigns data access permissions based on roles, uses lightweight homomorphic encryption technology to protect sensitive data, automatically identifies abnormal access behavior and blocks unauthorized operations, and combines zero-knowledge proof technology to design compliance review tools to monitor data usage behavior in real time; S7: Design A / B tests for digital economy scenarios, simulate digital economy scenarios for low-cost testing, build a user co-creation community, collect requirements and iterate quickly through crowdsourcing, and design functions based on the digital economy industry.

[0008] Furthermore, S1 collects digital economy-related data, covering the consumer, production, and service ends, including: Build a unified data collection platform to collect data in real time and offline from consumer, production, service and IoT device scenarios. Collect user click, browsing and search behavior data based on web page and APP tracking, browser plugins and log system. Based on payment system and order management system, collect transaction amount, time and product category information synchronously through API interface and database. IoT sensors are used to monitor logistics status, inventory levels, and production progress in real time and perform preliminary cleaning. Web crawlers are used to collect public comments from social media, news websites, and forums. NLP technology is used to extract keywords and sentiment. An industrial IoT gateway is deployed, and based on the MQTT protocol, equipment operating parameters are collected. Data is transmitted and processed in real time using the TLS 1.3 protocol and a streaming data processing framework. Differential privacy technology is used to blur sensitive information.

[0009] Furthermore, S2 employs blockchain technology to record the entire lifecycle and uses dynamic authorization, including restrictions on access frequency and usage, including: The data collected in S1 is divided into independent segments according to business scenarios using segmented blockchain, including consumer data segment and supply chain data segment. Each segment contains timestamp, data source, operator identity metadata, and records the modification history of each piece of data. Set up multi-condition trigger rules, including only allowing access to transaction data from 9:00 to 18:00 on weekdays and only allowing access to supply chain data from specific IP addresses. Deployed through blockchain, it supports fine-grained authorization, including allowing suppliers to view logistics status, prohibiting the export of complete order details, and dynamically adjusting permissions based on multiple dimensions such as role, time, and location. Access counters are deployed on blockchain nodes to count the number of accesses for each user and the system in real time. When the number of accesses reaches a threshold, an alert is triggered. The purpose of data use is verified based on zero-knowledge proofs, including that it is used only for market trend analysis.

[0010] Furthermore, in S3, a cross-modal cleaning model is constructed, and a knowledge graph is introduced for semantic association. Data tags are dynamically generated based on business needs, and composite features are extracted, including: Natural language processing (NLP) techniques are used to process text data, including user behavior data and market sentiment text, which are segmented, stop words removed, and stemmed. Semantic encoding is performed based on word vectors to eliminate ambiguity and noise. Image data collected by IoT devices is processed by grayscale conversion, edge detection, and image normalization to remove low-quality and duplicate images. Time-aligned timestamp information in transaction records and supply chain data is performed, and trend features are extracted using the sliding window moving average method. Hash encoding technology is used to map text, images, and time series data to a unified low-dimensional vector space to perform cross-modal feature association. Construct a knowledge graph, define the ontology in the digital economy field. Entity types include users, products, supply chain nodes, and IoT devices. Relationship types include purchasing, supply, monitoring, and influence. Attribute definitions include user ID, product price, and supply chain delay days. Semantically map objects in image data to entities in the knowledge graph. Based on business needs, design a tag rule base, generate active user tags based on user behavior data, and generate high-risk device tags based on IoT device data. Periodic features are extracted from transaction records, sentiment trends and keyword popularity are extracted from public opinion data, and color histograms and texture features are extracted from IoT monitoring images. Composite features are generated using feature cross-referencing techniques.

[0011] Furthermore, the simulation of the impact of different decision-making schemes on the digital economy in S4 includes: Transform business decisions into quantifiable parameters, including setting price increase strategies to simulate the impact on user purchase intentions and profits, inputting supply chain change plans, predicting changes in inventory turnover and logistics costs, and using counterfactual analysis to evaluate the effectiveness of decisions. An interactive decision tree is constructed, allowing users to drag and drop nodes to adjust decision parameters and display prediction results in real time. For conflicting objectives, a Pareto optimal solution set is generated.

[0012] Furthermore, S5 develops customized data dashboards, generates customized digital economy recommendations based on analysis results, and embeds these recommendations into digital economy business processes, including: The user behavior data, market sentiment, and transaction records collected as described in S1 are used as the data source for the market trend heatmap. Based on the cross-modal cleaning model described in S3, the data is standardized and divided into regions. The color intensity represents market activity. By clicking on a heatmap region, detailed data can be viewed and different indicators can be overlaid. IoT device data and sentiment data are continuously received. The heatmap is updated every minute. When there is a sudden drop in sales in a region, the heatmap is automatically marked as a risk area. The supply chain data, IoT device data and cleaned composite features described in S1 and S3 are used as the data sources for the supply chain risk map. Low risk is defined as a delay rate of <5% and a supplier score of >90, and high risk is defined as a delay rate of >20% and a supplier score of <60. The map is based on modeling supply chain nodes and relationships using graph neural networks. Node color represents risk, edge thickness represents logistics volume, and label represents delay days. Based on market trend heatmaps, supply chain risk maps, and Pareto frontier analysis, customized suggestions for the digital economy are generated and pushed to the enterprise ERP system via the API interface of the customized data dashboard.

[0013] Furthermore, in S6, data access permissions are assigned based on roles, lightweight homomorphic encryption technology is used to protect sensitive data, and data usage behavior is monitored in real time, including: User roles are preset and their permissions are dynamically bound using the blockchain technology described in S2. When a role's permissions change, the blockchain records the permission adjustment event and triggers a data tag update in S3. Access policies are dynamically generated by combining user attributes, data attributes, and environmental attributes. Attribute definitions are semantically associated based on the knowledge graph described in S3. The blockchain records the inheritance relationship between roles based on the blockchain described in S2. Access is prohibited by default, and only the minimum permissions required to complete the business are granted. The creation, modification, and revocation events of permissions are recorded. The encryption strategy is determined based on the dynamically generated data tags described in S3, and a lightweight homomorphic encryption scheme is adopted. The encryption parameters, including the key length, are optimized based on the prediction model described in S4. Detect abnormal access outside of working hours, combine IP address and geographical location, and analyze and identify risky behaviors based on the composite features extracted in S3. When high-risk behavior is detected, trigger the blocking mechanism, lock the user account and send an alarm to the administrator's email. Before accessing data, users submit a compliance statement using a zero-knowledge proof tool. Combined with authorization records, the system verifies whether the user's current access permissions comply with digital economy rules.

[0014] Furthermore, S7 includes A / B testing designed for digital economy scenarios, the construction of a user co-creation community, and functional design based on the digital economy industry, including: Test scenarios are divided according to the types of digital economy businesses, including the consumer end and the production end. Based on the S4 prediction model, a virtual sandbox is built to simulate real business scenarios. Different supply chain strategies are simulated in the sandbox, and changes in the inventory risk map are compared in real time. Based on permission management, test variables are controlled to only affect the target user group and multi-dimensional variable combinations are generated. Test result reports are automatically generated. Based on the test results, the system automatically triggers the parameter update of the prediction model described in S4 to optimize subsequent business strategies. A user co-creation community is built on a low-code platform. Based on the knowledge graph described in S3, user needs are semantically associated and tags are automatically generated. Based on the digital economy industry and the role and permission system described in S6, different roles are assigned to practitioners for functional design. Based on the data dashboard described in S5, the design is optimized based on data feedback.

[0015] Furthermore, a digital economy information management system based on big data analytics is provided to implement the digital economy information management method based on big data analytics described above. The big data analytics-based digital economy information management system includes: a data acquisition and preprocessing module, a blockchain full lifecycle management module, a cross-modal data cleaning and semantic association module, a prediction and decision simulation module, a visualization and decision support module, a security and access control module, and a testing and co-creation module. Data Acquisition and Preprocessing Module: Constructs a unified data acquisition platform, utilizing multi-source heterogeneous data acquisition technologies such as webpage tracking, APP logs, browser plugins, IoT sensors, and web crawlers to cover consumer, production, and server scenarios. Employs differential privacy technology to obfuscate sensitive information and ensures secure streaming data transmission through the TLS 1.3 protocol. Performs preliminary cleaning and structuring processing on IoT device data and achieves real-time acquisition of transaction data by synchronizing with API interfaces and databases. Blockchain full lifecycle management module: Based on segmented blockchain technology, data is divided according to business scenarios. Each data record includes timestamp, operator identity, and modification history metadata. The dynamic authorization mechanism supports multi-condition triggering rules and deploys fine-grained access control through the blockchain. Access counters and zero-knowledge proof verification modules are deployed on blockchain nodes to monitor the purpose of data use in real time and trigger an early warning mechanism. Cross-modal data cleaning and semantic association module: Natural language processing is used to segment text data, remove stop words, and semantically encode it. Combined with image processing technology and time series analysis, multimodal data cleaning is carried out to construct a knowledge graph in the digital economy field, define entities, relationships and attributes, and map text, image and time series data to a unified low-dimensional vector space through hash encoding. Data labels are dynamically generated based on the business rule base and composite features are extracted. Prediction and Decision Simulation Module: Based on cleaned composite features, a prediction model specifically for the digital economy is trained. The effect of decision-making is evaluated through counterfactual analysis. An interactive decision tree is constructed, which allows users to drag and drop to adjust parameters and display prediction results in real time. Pareto optimal solution sets are generated for conflicting objectives. Combined with supply chain risk maps and market trend heat maps, multi-dimensional decision simulation scenarios are provided. Visualization and Decision Support Module: Develop customized data dashboards, including market trend heatmaps and supply chain risk maps. Model the relationships between supply chain nodes through graph neural networks, update the heatmaps in real time and automatically mark risk areas, generate customized suggestions based on Pareto front analysis, and push them to the enterprise ERP system through API interface; Security and Access Management Module: Role-based access control combined with blockchain for dynamic permission binding; generating least privilege access policies by semantically associating user attributes, data attributes, and environment attributes through knowledge graphs; protecting sensitive data with lightweight homomorphic encryption technology; optimizing encryption parameters by combining predictive models; monitoring abnormal access behavior in real time; verifying user compliance claims through zero-knowledge proof tools; and triggering account locking and alarm mechanisms. Testing and Co-creation Module: Design an A / B testing sandbox to simulate real business scenarios based on a predictive model. Generate test reports through multi-dimensional variable combinations and automatically optimize predictive model parameters. Build a low-code user co-creation community, use knowledge graphs to semantically associate and tag user needs, and combine a role-based access control system to assign functional design tasks to practitioners. Feedback on test results through data dashboards drives rapid iteration of functional design and adaptation to industry scenarios.

[0016] The beneficial effects of this invention are as follows: By using a unified data collection platform, it covers data from all scenarios, including consumer, production, and service ends; by combining differential privacy and TLS 1.3 protocol to ensure data security; by combining cross-modal cleaning models and knowledge graphs to achieve data standardization, noise reduction, and semantic mapping, thereby improving data quality and interpretability; by using segmented blockchain to record the entire process of data collection, sharing, and use; and by combining dynamic authorization and zero-knowledge proof verification to ensure data transparency and compliance and prevent data abuse. By training a digital economy-specific prediction model and combining it with Pareto optimal solution set generation, we can support multi-objective decision-making balance and improve the scientific nature of decision-making. Market trend heatmaps and supply chain risk maps are modeled and updated in real time through graph neural networks, which can intuitively present key indicators such as regional activity and supply chain delay rate, automatically mark risk areas and trigger warnings, shorten response time. The user co-creation community, combined with A / B testing sandbox, collects requirements and verifies them quickly through crowdsourcing, reducing trial and error costs and accelerating feature iteration and business scenario adaptation. Role-based access control combined with blockchain-based dynamic permission binding protects highly sensitive data through lightweight homomorphic encryption, ensuring that aggregation and analysis are still possible even in the encrypted state, balancing security and availability. Combined with composite feature analysis and zero-knowledge proof compliance review, it identifies and blocks high-risk operations in real time, reducing the risk of data leakage and unauthorized use. The blockchain records operation logs such as permission changes, data access, and abnormal events, forming an immutable audit trail that meets the regulatory requirements of the digital economy. The modular design of the system can be flexibly adapted to the needs of various fields such as manufacturing, retail, and service industries, promoting collaboration between upstream and downstream industries. Through application scenarios such as supply chain optimization and energy monitoring, it reduces resource waste and carbon emissions, helps enterprises achieve ESG goals, and lowers the technical threshold with low-cost A / B testing, low-code platforms, and inclusive computing power, helping SMEs quickly access the digital economy ecosystem and narrow the digital divide. Attached Figure Description

[0017] Figure 1 This is a flowchart of a digital economy information management method based on big data analysis. Figure 2 This is a module diagram of a digital economy information management system based on big data analysis. Detailed Implementation

[0018] The present invention will be further described clearly and completely below, but the scope of protection of the present invention is not limited thereto.

[0019] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0020] Example 1 A digital economy information management method based on big data analysis, employing the following technical solution: S1: Collect data related to the digital economy, including user behavior data, transaction records, supply chain data, market sentiment, and IoT device data, covering the consumer, production, and service ends; S2: Uses blockchain technology to record the entire lifecycle, including the entire lifecycle of data collection, sharing and use, and adopts dynamic authorization, including restrictions on access frequency and usage. Data providers adjust the scope of authorization according to digital economy rules. S3: Construct a cross-modal cleaning model to jointly process text, image and time series data, formulate data standardization rules specific to the digital economy, introduce knowledge graphs for semantic association, dynamically generate data tags based on business needs, and extract composite features; S4: Based on the data processed in S3, train a digital economy-specific prediction model to simulate the impact of different decision-making schemes on the digital economy; S5: Develop customized data dashboards, including market trend heatmaps and supply chain risk maps, generate customized suggestions for the digital economy based on analysis results, and embed decision-making suggestions into digital economy business processes; S6: Assigns data access permissions based on roles, uses lightweight homomorphic encryption technology to protect sensitive data, automatically identifies abnormal access behavior and blocks unauthorized operations, and combines zero-knowledge proof technology to design compliance review tools to monitor data usage behavior in real time; S7: Design A / B tests for digital economy scenarios, simulate digital economy scenarios for low-cost testing, build a user co-creation community, collect requirements and iterate quickly through crowdsourcing, and design functions based on the digital economy industry.

[0021] refer to Figure 1 This is a flowchart of a digital economy information management method based on big data analysis.

[0022] Furthermore, S1 collects digital economy-related data, covering the consumer, production, and service ends, including: Build a unified data collection platform to collect data in real time or offline from consumer user behavior, production supply chain, service market sentiment and IoT device industrial sensor scenarios. Collect user click, browsing and search behavior data based on web page and APP tracking, browser plugins and ELK Stack log system. Based on payment systems including Alipay, WeChat Pay and ERP order management system, collect transaction amount, time and product category information synchronously through API interface and database. IoT sensors, including RFID tags and temperature and humidity sensors, are used to monitor logistics status, inventory levels, and production progress in real time and perform initial cleaning. Public comments from social media, news websites, and forums are crawled using the Scrapy web crawler framework, and keywords and sentiment are extracted using NLP technology. An industrial IoT gateway, Siemens SIMATIC IOT2000, is deployed based on the MQTT protocol to collect equipment operating parameters, including temperature, pressure, and vibration frequency. Data is transmitted and processed in real time using the TLS 1.3 protocol and the Apache Kafka streaming data processing framework. Differential privacy technology, the Laplace mechanism, is used to obfuscate sensitive information.

[0023] Furthermore, S2 employs blockchain technology to record the entire lifecycle and uses dynamic authorization, including restrictions on access frequency and usage, including: The data collected in S1 is divided into independent segments according to business scenarios, such as consumer data segment and supply chain data segment, using segmented blockchain. Each segment contains metadata such as timestamp, data source, operator identity, etc., and records the modification history of each piece of data. Set up multi-condition trigger rules, including only allowing access to transaction data from 9:00 to 18:00 on weekdays and only allowing access to supply chain data from specific IP addresses. Deployed through blockchain, it supports fine-grained authorization, including allowing suppliers to view logistics status, prohibiting the export of complete order details, and dynamically adjusting permissions based on multiple dimensions such as role, time, and location. Access counters are deployed on blockchain nodes to count the number of accesses by each user and system in real time. When the number of accesses reaches a threshold, an alert is triggered. For example, if a system accesses market sentiment data more than 100 times in one hour, the purpose of the data use is verified based on zero-knowledge proof, including that it is only used for market trend analysis.

[0024] Furthermore, in S3, a cross-modal cleaning model is constructed, and a knowledge graph is introduced for semantic association. Data tags are dynamically generated based on business needs, and composite features are extracted, including: Natural language processing (NLP) technology is used to process text data, including user behavior data and market sentiment text, which are segmented, stop words removed, and stemmed. Semantic encoding is performed based on word vectors to eliminate ambiguity and noise. Image data collected by IoT devices, including supply chain logistics monitoring images, are processed by grayscale conversion, edge detection, and image normalization to remove low-quality and duplicate images. Time-aligned timestamp information in transaction records and supply chain data is performed, and trend features are extracted using the sliding window moving average method. Hash encoding technology is used to map text, images, and time series data to a unified low-dimensional vector space to perform cross-modal feature association. Construct a knowledge graph and define the ontology of the digital economy. Entity types include users, products, supply chain nodes, and IoT devices. Relationship types include purchasing, supply, monitoring, and impact. Attribute definitions include user ID, product price, and supply chain delay days. Semantically map objects in image data, such as faulty devices in IoT monitoring screens, to entities in the knowledge graph, such as device faults. Based on business needs, such as highly active users and high-risk nodes in the supply chain, design a tag rule base. Generate active user tags based on user behavior data, such as average daily access time >30 minutes, and generate high-risk device tags based on IoT device data, such as temperature anomaly frequency >5 times / week. Periodic features, including fluctuations in average monthly spending per user, are extracted from transaction records. Sentimental tendencies, including positive, negative, and keyword popularity, are extracted from public opinion data. Color histograms and texture features are extracted from IoT monitoring images. Composite features are generated using feature cross-referencing techniques, such as user activity (text) × supply chain delay rate (time series) as a consumer-side supply chain sensitivity feature, and product price (numerical) + user evaluation sentiment (text) as a comprehensive product attractiveness feature.

[0025] Furthermore, the simulation of the impact of different decision-making schemes on the digital economy in S4 includes: Transform business decisions into quantifiable parameters, including setting price increase strategies to simulate the impact on user purchase intentions and profits, such as setting a 10% price reduction or a 5% price increase strategy; input supply chain change scenarios, such as increasing inventory by 20% or switching suppliers; predict changes in inventory turnover and logistics costs; and use counterfactual analysis to evaluate the effectiveness of decisions, such as sales may decrease by 15% if no promotional activities are launched, while user conversion rate may increase by 8% if advertising is increased. Construct an interactive decision tree where users can drag and drop nodes to adjust decision parameters, such as increasing the advertising budget from 500,000 to 800,000. Real-time prediction results are displayed, such as the expected number of new users. For conflicting objectives, generate Pareto optimal solution sets. For example, choosing to expand production scale may increase profits by 20%, but carbon emissions will increase by 12%. Choosing a green supply chain may decrease profits by 5%, but carbon emissions will decrease by 30%.

[0026] Furthermore, S5 develops customized data dashboards, generates customized digital economy recommendations based on analysis results, and embeds these recommendations into digital economy business processes, including: The user behavior data, market sentiment, and transaction records collected as described in S1 are used as the data source for the market trend heatmap. Based on the cross-modal cleaning model described in S3, the data is standardized and divided by region, such as city and province. The color intensity represents market activity, such as red representing high growth potential and blue representing low demand. Clicking on a heatmap area, such as a city, allows viewing detailed data, such as user profiles and sentiment keywords for that area. Different indicators, such as sales, user activity, and sentiment, are overlaid. Data from IoT devices, such as store traffic and sentiment data, are continuously received. The heatmap is updated every minute. When there is a sudden drop in sales in a region, the heatmap is automatically marked as a risk area. The supply chain data, IoT device data, and cleaned composite features described in S1 and S3 are used as the data sources for the supply chain risk map. Low risk is defined as a delay rate of <5% and a supplier score of >90, and high risk is defined as a delay rate of >20% and a supplier score of <60. The map is based on graph neural network modeling of supply chain nodes such as factories and warehouses and relationships such as factory A → supplier B. Node colors, such as green for low risk and red for high risk, edge thickness for logistics volume, and labels for delay days. Based on market trend heatmaps, supply chain risk maps, and Pareto frontier analysis, customized suggestions for the digital economy are generated and pushed to the enterprise ERP system via the API interface of the customized data dashboard.

[0027] Furthermore, in S6, data access permissions are assigned based on roles, lightweight homomorphic encryption technology is used to protect sensitive data, and data usage behavior is monitored in real time, including: The system includes preset user roles such as market analyst and supply chain manager. Role permissions are dynamically bound using the blockchain technology described in S2. For example, a market analyst can access market sentiment and transaction records, but cannot view IoT device data. When role permissions change, the blockchain records the permission adjustment event and triggers a data tag update in S3. Access policies are dynamically generated based on user attributes (department, job level), data attributes (sensitivity level, data type), and environmental attributes (access time, IP address). For example, if a user is in the finance department and accesses the system during weekdays (9:00-18:00), they are allowed to view transaction records. If the access time is late at night, they are only allowed to view the hidden amount field in anonymized data. Attribute definitions are semantically linked based on the knowledge graph described in S3, such as finance department → transaction records → high sensitivity level. Inheritance relationships between roles are recorded based on the blockchain described in S2, such as senior manager → inheriting ordinary manager permissions. Access is prohibited by default, and only the minimum permissions required to complete business tasks are granted. For example, a supply chain manager can only view logistics data related to their assigned region and cannot access data from other departments across regions. The system also records the creation, modification, and revocation of permissions. The dynamically generated data tags described in S3 include high-sensitivity and low-sensitivity to determine the encryption strategy. For example, high-sensitivity data, including user mobile phone numbers, uses homomorphic encryption, allowing aggregation analysis and statistics of the number of users in a certain area under encrypted conditions. Low-sensitivity data, including public market sentiment, does not require encryption and uses a lightweight homomorphic encryption scheme, such as a ring-based encryption algorithm. The encryption parameters, including key length, are optimized based on the prediction model described in S4. Detect abnormal access outside of working hours, such as accessing the financial system at 3 a.m., and combine IP address and geographical location, such as sudden login from an overseas IP. Based on the composite feature analysis extracted in S3, identify risky behaviors, such as user A → high-frequency download of transaction records → non-working device. When high-risk behavior is detected, trigger the blocking mechanism, lock the user account and send an alarm to the administrator's email. Before accessing data, users submit a compliance statement through a zero-knowledge proof tool, stating that the data is only for market analysis. Combined with authorization records, the system verifies whether the user's current access permissions comply with digital economy rules, such as prohibiting the use of user behavior data for advertising.

[0028] Furthermore, S7 includes A / B testing designed for digital economy scenarios, the construction of a user co-creation community, and functional design based on the digital economy industry, including: Test scenarios are divided according to the types of digital economy businesses, including e-commerce promotions and supply chain optimization, including the consumer end and the production end. For example, on the consumer end, the impact of different promotional copy on conversion rate is tested, and the effect of supply chain scheduling strategies on improving inventory turnover rate is tested. Based on the prediction model of S4, a virtual sandbox is built to simulate real business scenarios, such as a 10% increase in logistics delay rate in a certain region. Different supply chain strategies are simulated in the sandbox, including strategy A adding backup suppliers and strategy B adjusting transportation routes. The changes in the inventory risk map are compared in real time. Based on permission management, test variables such as price and interface layout are controlled to only affect the target user group. For example, only the new promotional page is pushed to the test group A, while the control group B retains the old version. Multi-dimensional variable combinations such as price + copy + button color are generated, and test result reports are automatically generated, such as strategy A increasing the conversion rate by 12%. Based on the test results, the system automatically triggers the parameter update of the prediction model described in S4 to optimize subsequent business strategies. A user co-creation community is built on a low-code platform. Based on the knowledge graph described in S3, semantic associations are made for user needs, such as supply chain visualization → inventory risk graph. Tags such as high priority and low implementation cost are automatically generated. Based on the digital economy industry and combined with the role and permission system described in S6, different roles are assigned to practitioners for functional design. Based on the data dashboard described in S5, the design is optimized based on data feedback.

[0029] Furthermore, a digital economy information management system based on big data analytics is provided to implement the digital economy information management method based on big data analytics described above. The big data analytics-based digital economy information management system includes: a data acquisition and preprocessing module, a blockchain full lifecycle management module, a cross-modal data cleaning and semantic association module, a prediction and decision simulation module, a visualization and decision support module, a security and access control module, and a testing and co-creation module. Data Acquisition and Preprocessing Module: Constructs a unified data acquisition platform, utilizing multi-source heterogeneous data acquisition technologies such as webpage tracking, APP logs, browser plugins, IoT sensors, and web crawlers to cover consumer, production, and server scenarios. Employs differential privacy technology to obfuscate sensitive information and ensures secure streaming data transmission through the TLS 1.3 protocol. Performs preliminary cleaning and structuring processing on IoT device data and achieves real-time acquisition of transaction data by synchronizing with API interfaces and databases. Blockchain full lifecycle management module: Based on segmented blockchain technology, data is divided according to business scenarios. Each data record includes timestamp, operator identity, and modification history metadata. The dynamic authorization mechanism supports multi-condition triggering rules and deploys fine-grained access control through the blockchain. Access counters and zero-knowledge proof verification modules are deployed on blockchain nodes to monitor the purpose of data use in real time and trigger an early warning mechanism. Cross-modal data cleaning and semantic association module: Natural language processing is used to segment text data, remove stop words, and semantically encode it. Combined with image processing technology and time series analysis, multimodal data cleaning is carried out to construct a knowledge graph in the digital economy field, define entities, relationships and attributes, and map text, image and time series data to a unified low-dimensional vector space through hash encoding. Data labels are dynamically generated based on the business rule base and composite features are extracted. Prediction and Decision Simulation Module: Based on cleaned composite features, a prediction model specifically for the digital economy is trained. The effect of decision-making is evaluated through counterfactual analysis. An interactive decision tree is constructed, which allows users to drag and drop to adjust parameters and display prediction results in real time. Pareto optimal solution sets are generated for conflicting objectives. Combined with supply chain risk maps and market trend heat maps, multi-dimensional decision simulation scenarios are provided. Visualization and Decision Support Module: Develop customized data dashboards, including market trend heatmaps and supply chain risk maps. Model the relationships between supply chain nodes through graph neural networks, update the heatmaps in real time and automatically mark risk areas, generate customized suggestions based on Pareto front analysis, and push them to the enterprise ERP system through API interface; Security and Access Management Module: Role-based access control combined with blockchain for dynamic permission binding; generating least privilege access policies by semantically associating user attributes, data attributes, and environment attributes through knowledge graphs; protecting sensitive data with lightweight homomorphic encryption technology; optimizing encryption parameters by combining predictive models; monitoring abnormal access behavior in real time; verifying user compliance claims through zero-knowledge proof tools; and triggering account locking and alarm mechanisms. Testing and Co-creation Module: Design an A / B testing sandbox to simulate real business scenarios based on a predictive model. Generate test reports through multi-dimensional variable combinations and automatically optimize predictive model parameters. Build a low-code user co-creation community, use knowledge graphs to semantically associate and tag user needs, and combine a role-based access control system to assign functional design tasks to practitioners. Feedback on test results through data dashboards drives rapid iteration of functional design and adaptation to industry scenarios.

[0030] refer to Figure 2 The diagram shown is a module diagram of a digital economy information management system based on big data analysis.

[0031] Example 2 A food company is facing inventory backlog in its East China warehouse and shortages in South China. It needs to integrate consumer sales data, production IoT device data, and public opinion data to dynamically optimize its supply chain strategy.

[0032] In e-commerce apps, the Apache SDK is used to collect user clicks and add-to-cart behaviors, such as clicks on chocolate products, and transaction records, including timestamps, amounts, and SKUs, are synchronized via the payment API. Deploy temperature and humidity sensors (model: Sensirion SHT45) and RFID tags in the East China warehouse, and transmit inventory data through the MQTT protocol. For example, the inventory of chocolate is 1,200 boxes, and the temperature is 18°C. Use Scrapy crawler to capture comments on the Weibo topic "#Snack Recommendation#", with a daily capture volume of 5,000 comments. Use the BERT model for NLP sentiment analysis to output the proportion of positive evaluations of chocolate: 72%. Add Laplace noise (ε = 0.5) to the transaction amount to achieve differential privacy. After the data is encrypted by TLS 1.3, it is transmitted to the central node through Kafka streams.

[0033] Create an independent blockchain segment: Consumption segment: Store transaction data, block record: {tx_hash: "0x3a1b…", operator: "Finance Department", time: 2023-08-22 10:00}; Supply chain segment: Store warehouse temperature and humidity data, block record: {sensor_id: "WH-E1", value: "18°C", source: "IoT gateway"}. Set rules: if (role == "supplier" and time not in ["9:00-18:00"]) or (IP!= "192.168.1.0 / 24"): deny_access("supply chain data") # Prohibit access to supply chain data outside working hours / from non-intranet IPs. Suppliers can only view logistics status such as the number of in-transit vehicles and cannot export order details. The Marketing Department submits a statement before accessing data: Purpose: Demand forecasting. The system verifies its permission scope, such as prohibiting use for advertising.

[0034] Tokenize Weibo comments → Remove stop words (such as "de", "a") → Stem extraction (e.g., "delicious" → "good") → Word vector encoding (dimension: 300); Grayscale the warehouse monitoring screen → Apply Canny edge detection → Normalize to 224×224 pixels; Align the timestamps of sales data and calculate the weekly average sales using a sliding window (window = 7 days). Knowledge graph construction: Entities: [User ID: U1001, Product: Chocolate A, Warehouse: WH-E1]; Relationships: (Chocolate A)-[Stored in]->(WH-E1), (U1001)-[Negative review]->(Chocolate A); Attributes: {Chocolate A: {Price: ￥25, Weekly sales volume: 1,200}}. When the inventory in the East China warehouse > 1,500 boxes and the weekly sales volume < 800 boxes → Generate a high inventory risk label. When the negative comments on Weibo suddenly increase by 50% → Generate a public opinion warning label. Feature crossing: (User click volume × Warehouse temperature and humidity volatility) → Supply chain sensitivity feature.

[0035] Simulated decision-making: If 20% of the inventory in East China is transferred to South China, the predicted results are: inventory turnover rate in East China will increase by 15%, but logistics costs will increase by ¥5,000. The Pareto optimal solution set is generated as follows: Option A (land transportation) profit change +¥8,000, carbon emission change +12%; Option B (sea transportation) profit change +¥3,000, carbon emission change -5%. Users can drag and drop parameters (such as a 10% promotional discount) to view real-time forecast results: sales in South China are expected to increase by 18%.

[0036] Divided by province, the shade of color equals the sales growth rate. Clicking on a region displays details: {User profile: Generation Z accounts for 68%, public opinion keyword: moderate sweetness}. When sales drop by 20% within 1 hour, the heat map icon turns red and a text message alert is sent. Supply Chain Risk Map: Nodes: Warehouse (size = inventory level, color = risk level), Edges: Logistics route (thickness = transportation volume, label: delay days), Rules: Delay rate > 20% + score < 60 points → marked as high risk (red node); Decision suggestions are pushed to the SAP system via API.

[0037] When a user's supply chain manager role changes, the blockchain records the event and updates the data tags. For example, if the user loses the right to modify inventory, the user's mobile phone number is encrypted (Solution: TFHE). The key length is dynamically adjusted to 256 bits. Security requirements are calculated based on a predictive model. If the R&D department's account is detected downloading transaction data from a Vietnamese IP at 3:00 AM, the following will be triggered: 1. Account is locked. 2. Email alert: [High Risk] Abnormal access: User X downloaded 1000 transaction records.

[0038] Simulated Scenario: The impact of excessive temperature and humidity in an East China warehouse on chocolate spoilage rate. Group A installed refrigeration equipment and offered a promotional scheme of 20 RMB off for every 100 RMB spent; Group B added ventilation and offered a limited-time discount. Results: Group A's spoilage rate decreased by 40%, and its conversion rate increased by 12% → The system automatically updated the prediction model parameters. The low-code platform OutSystems generates a request form where users submit urgent requests for cold chain transportation monitoring functionality. This functionality is then linked to a knowledge graph with the following tags: [Priority: High, Related Module: Supply Chain Graph]. Based on feedback, the new feature is iterated and launched within 72 hours.

[0039] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0040] This invention provides a digital economy information management method and system based on big data analysis. It integrates consumer, production, and service data through a unified platform, combines differential privacy and TLS 1.3 to ensure transmission security, employs segmented blockchain to record the entire data lifecycle, supports dynamic authorization and fine-grained access control, verifies data usage using zero-knowledge proofs, integrates NLP, image processing, and time series analysis to construct a knowledge graph, and generates dynamic tags and composite features. It simulates the impact of decisions through counterfactual analysis, supports Pareto optimal solution generation through interactive decision trees, provides market trend heatmaps and supply chain risk maps, pushes customized suggestions to the ERP system in real time, protects sensitive data based on role-based permissions and lightweight homomorphic encryption, combines abnormal behavior monitoring and zero-knowledge proof compliance review, simulates business scenarios through A / B sandboxes, and drives requirement iteration through a low-code community.

Claims

1. A digital economy information management method based on big data analysis, characterized in that, include: S1: Collect data related to the digital economy, including user behavior data, transaction records, supply chain data, market sentiment, and IoT device data, covering the consumer, production, and service ends; S2: Uses blockchain technology to record the entire lifecycle, including the entire lifecycle of data collection, sharing and use, and adopts dynamic authorization, including restrictions on access frequency and usage. Data providers adjust the scope of authorization according to digital economy rules. S3: Construct a cross-modal cleaning model to jointly process text, image and time series data, formulate data standardization rules specific to the digital economy, introduce knowledge graphs for semantic association, dynamically generate data tags based on business needs, and extract composite features; S4: Based on the data processed in S3, train a digital economy-specific prediction model to simulate the impact of different decision-making schemes on the digital economy; S5: Develop customized data dashboards, including market trend heatmaps and supply chain risk maps, generate customized suggestions for the digital economy based on analysis results, and embed decision-making suggestions into digital economy business processes; S6: Assigns data access permissions based on roles, uses lightweight homomorphic encryption technology to protect sensitive data, automatically identifies abnormal access behavior and blocks unauthorized operations, and combines zero-knowledge proof technology to design compliance review tools to monitor data usage behavior in real time; S7: Design A / B tests for digital economy scenarios, simulate digital economy scenarios for low-cost testing, build a user co-creation community, collect requirements and iterate quickly through crowdsourcing, and design functions based on the digital economy industry.

2. The digital economy information management method based on big data analysis as described in claim 1, characterized in that, The S1 process collects digital economy-related data, covering the consumer, production, and service sectors, including: Build a unified data collection platform to collect data in real time and offline from consumer, production, service and IoT device scenarios. Collect user click, browsing and search behavior data based on web page and APP tracking, browser plugins and log system. Based on payment system and order management system, collect transaction amount, time and product category information synchronously through API interface and database. IoT sensors are used to monitor logistics status, inventory levels, and production progress in real time and perform preliminary cleaning. Web crawlers are used to collect public comments from social media, news websites, and forums. NLP technology is used to extract keywords and sentiment. An industrial IoT gateway is deployed, and based on the MQTT protocol, equipment operating parameters are collected. Data is transmitted and processed in real time using the TLS 1.3 protocol and a streaming data processing framework. Differential privacy technology is used to blur sensitive information.

3. The digital economy information management method based on big data analysis as described in claim 1, characterized in that, S2 uses blockchain technology to record the entire lifecycle and employs dynamic authorization, including restrictions on access frequency and usage, including: The data collected in S1 is divided into independent segments according to business scenarios using segmented blockchain, including consumer data segment and supply chain data segment. Each segment contains timestamp, data source, operator identity metadata, and records the modification history of each piece of data. Set up multi-condition trigger rules, including only allowing access to transaction data from 9:00 to 18:00 on weekdays and only allowing access to supply chain data from specific IP addresses. Deployed through blockchain, it supports fine-grained authorization, including allowing suppliers to view logistics status, prohibiting the export of complete order details, and dynamically adjusting permissions based on multiple dimensions such as role, time, and location. Access counters are deployed on blockchain nodes to count the number of accesses for each user and the system in real time. When the number of accesses reaches a threshold, an alert is triggered. The purpose of data use is verified based on zero-knowledge proofs, including that it is used only for market trend analysis.

4. The digital economy information management method based on big data analysis as described in claim 1, characterized in that, The S3 section constructs a cross-modal cleaning model and introduces a knowledge graph for semantic association. It dynamically generates data tags based on business needs and extracts composite features, including: Natural language processing (NLP) techniques are used to process text data, including user behavior data and market sentiment text, which are segmented, stop words removed, and stemmed. Semantic encoding is performed based on word vectors to eliminate ambiguity and noise. Image data collected by IoT devices is processed by grayscale conversion, edge detection, and image normalization to remove low-quality and duplicate images. Time-aligned timestamp information in transaction records and supply chain data is performed, and trend features are extracted using the sliding window moving average method. Hash encoding technology is used to map text, images, and time series data to a unified low-dimensional vector space to perform cross-modal feature association. Construct a knowledge graph, define the ontology in the digital economy field. Entity types include users, products, supply chain nodes, and IoT devices. Relationship types include purchasing, supply, monitoring, and influence. Attribute definitions include user ID, product price, and supply chain delay days. Semantically map objects in image data to entities in the knowledge graph. Based on business needs, design a tag rule base, generate active user tags based on user behavior data, and generate high-risk device tags based on IoT device data. Periodic features are extracted from transaction records, sentiment trends and keyword popularity are extracted from public opinion data, and color histograms and texture features are extracted from IoT monitoring images. Composite features are generated using feature cross-referencing techniques.

5. The digital economy information management method based on big data analysis as described in claim 1, characterized in that, The S4 simulation of the impact of different decision-making schemes on the digital economy includes: Transform business decisions into quantifiable parameters, including setting price increase strategies to simulate the impact on user purchase intentions and profits, inputting supply chain change plans, predicting changes in inventory turnover and logistics costs, and using counterfactual analysis to evaluate the effectiveness of decisions. An interactive decision tree is constructed, allowing users to drag and drop nodes to adjust decision parameters and display prediction results in real time. For conflicting objectives, a Pareto optimal solution set is generated.

6. The digital economy information management method based on big data analysis as described in claim 1, characterized in that, The S5 system develops customized data dashboards, generates customized suggestions for the digital economy based on analysis results, and embeds these suggestions into digital economy business processes, including: The user behavior data, market sentiment, and transaction records collected as described in S1 are used as the data source for the market trend heatmap. Based on the cross-modal cleaning model described in S3, the data is standardized and divided into regions. The color intensity represents market activity. By clicking on a heatmap region, detailed data can be viewed and different indicators can be overlaid. IoT device data and sentiment data are continuously received. The heatmap is updated every minute. When there is a sudden drop in sales in a region, the heatmap is automatically marked as a risk area. The supply chain data, IoT device data and cleaned composite features described in S1 and S3 are used as the data sources for the supply chain risk map. Low risk is defined as a delay rate of <5% and a supplier score of >90, and high risk is defined as a delay rate of >20% and a supplier score of <60. The map is based on modeling supply chain nodes and relationships using graph neural networks. Node color represents risk, edge thickness represents logistics volume, and label represents delay days. Based on market trend heatmaps, supply chain risk maps, and Pareto frontier analysis, customized suggestions for the digital economy are generated and pushed to the enterprise ERP system via the API interface of the customized data dashboard.

7. The digital economy information management method based on big data analysis as described in claim 1, characterized in that, The S6 mechanism assigns data access permissions based on roles, employs lightweight homomorphic encryption to protect sensitive data, and monitors data usage behavior in real time, including: User roles are preset and their permissions are dynamically bound using the blockchain technology described in S2. When a role's permissions change, the blockchain records the permission adjustment event and triggers a data tag update in S3. Access policies are dynamically generated by combining user attributes, data attributes, and environmental attributes. Attribute definitions are semantically associated based on the knowledge graph described in S3. The blockchain records the inheritance relationship between roles based on the blockchain described in S2. Access is prohibited by default, and only the minimum permissions required to complete the business are granted. The creation, modification, and revocation events of permissions are recorded. The encryption strategy is determined based on the dynamically generated data tags described in S3, and a lightweight homomorphic encryption scheme is adopted. The encryption parameters, including the key length, are optimized based on the prediction model described in S4. Detect abnormal access outside of working hours, combine IP address and geographical location, and analyze and identify risky behaviors based on the composite features extracted in S3. When high-risk behavior is detected, trigger the blocking mechanism, lock the user account and send an alarm to the administrator's email. Before accessing data, users submit a compliance statement using a zero-knowledge proof tool. Combined with authorization records, the system verifies whether the user's current access permissions comply with digital economy rules.

8. The digital economy information management method based on big data analysis as described in claim 1, characterized in that, The S7 design incorporates A / B testing for digital economy scenarios, builds a user co-creation community, and designs functionalities based on the digital economy industry, including: Test scenarios are divided according to the types of digital economy businesses, including the consumer end and the production end. Based on the S4 prediction model, a virtual sandbox is built to simulate real business scenarios. Different supply chain strategies are simulated in the sandbox, and changes in the inventory risk map are compared in real time. Based on permission management, test variables are controlled to only affect the target user group and multi-dimensional variable combinations are generated. Test result reports are automatically generated. Based on the test results, the system automatically triggers the parameter update of the prediction model described in S4 to optimize subsequent business strategies. A user co-creation community is built on a low-code platform. Based on the knowledge graph described in S3, user needs are semantically associated and tags are automatically generated. Based on the digital economy industry and the role and permission system described in S6, different roles are assigned to practitioners for functional design. Based on the data dashboard described in S5, the design is optimized based on data feedback.

9. A digital economy information management system based on big data analysis, characterized in that, To implement the digital economy information management method based on big data analysis as described in any one of claims 1-8, the digital economy information management system based on big data analysis includes: a data acquisition and preprocessing module, a blockchain full lifecycle management module, a cross-modal data cleaning and semantic association module, a prediction and decision simulation module, a visualization and decision support module, a security and access control module, and a testing and co-creation module. Data Acquisition and Preprocessing Module: Constructs a unified data acquisition platform, utilizing multi-source heterogeneous data acquisition technologies such as webpage tracking, APP logs, browser plugins, IoT sensors, and web crawlers to cover consumer, production, and server scenarios. Employs differential privacy technology to obfuscate sensitive information and ensures secure streaming data transmission through the TLS 1.3 protocol. Performs preliminary cleaning and structuring processing on IoT device data and achieves real-time acquisition of transaction data by synchronizing with API interfaces and databases. Blockchain full lifecycle management module: Based on segmented blockchain technology, data is divided according to business scenarios. Each data record includes timestamp, operator identity, and modification history metadata. The dynamic authorization mechanism supports multi-condition triggering rules and deploys fine-grained access control through the blockchain. Access counters and zero-knowledge proof verification modules are deployed on blockchain nodes to monitor the purpose of data use in real time and trigger an early warning mechanism. Cross-modal data cleaning and semantic association module: Natural language processing is used to segment text data, remove stop words, and semantically encode it. Combined with image processing technology and time series analysis, multimodal data cleaning is carried out to construct a knowledge graph in the digital economy field, define entities, relationships and attributes, and map text, image and time series data to a unified low-dimensional vector space through hash encoding. Data labels are dynamically generated based on the business rule base and composite features are extracted. Prediction and Decision Simulation Module: Based on cleaned composite features, a prediction model specifically for the digital economy is trained. The effect of decision-making is evaluated through counterfactual analysis. An interactive decision tree is constructed, which allows users to drag and drop to adjust parameters and display prediction results in real time. Pareto optimal solution sets are generated for conflicting objectives. Combined with supply chain risk maps and market trend heat maps, multi-dimensional decision simulation scenarios are provided. Visualization and Decision Support Module: Develop customized data dashboards, including market trend heatmaps and supply chain risk maps. Model the relationships between supply chain nodes through graph neural networks, update the heatmaps in real time and automatically mark risk areas, generate customized suggestions based on Pareto front analysis, and push them to the enterprise ERP system through API interface; Security and Access Management Module: Role-based access control combined with blockchain for dynamic permission binding; generating least privilege access policies by semantically associating user attributes, data attributes, and environment attributes through knowledge graphs; protecting sensitive data with lightweight homomorphic encryption technology; optimizing encryption parameters by combining predictive models; monitoring abnormal access behavior in real time; verifying user compliance claims through zero-knowledge proof tools; and triggering account locking and alarm mechanisms. Testing and Co-creation Module: Design an A / B testing sandbox to simulate real business scenarios based on a predictive model. Generate test reports through multi-dimensional variable combinations and automatically optimize predictive model parameters. Build a low-code user co-creation community, use knowledge graphs to semantically associate and tag user needs, and combine a role-based access control system to assign functional design tasks to practitioners. Feedback on test results through data dashboards drives rapid iteration of functional design and adaptation to industry scenarios.

Images