Judicial big data management method and system oriented to price discrimination analysis

Abstract

The invention relates to the technical field of big data management, and discloses a judicial big data management method and system for price discrimination analysis, and the method comprises the steps: obtaining an e-commerce platform transaction data flow; constructing a compliance correction rule set, wherein the compliance correction rule set comprises a price discrimination judgment standard and a correction action; detecting a potential price discrimination risk based on the e-commerce platform transaction data stream and the compliance correction rule set; according to the potential price discrimination risk and the compliance correction rule set, generating a pricing parameter adjustment suggestion; and sending the pricing parameter adjustment suggestion to a pricing system to correct a pricing strategy, and receiving an adjustment result after the pricing system executes the correction to form a compliance evaluation and response closed loop. According to the invention, the problem that an electronic commerce platform dynamic pricing strategy in the prior art may touch a legal red line, price discrimination occurs, and an effective evaluation and correction mechanism is lacked is effectively solved.

Description

Technical Field

[0001] This application relates to the field of big data management technology, and more specifically, to a judicial big data management method and system for price discrimination analysis. Background Technology

[0002] With the booming development of the digital economy, e-commerce platforms are playing an increasingly important role in market competition. To maintain a competitive edge in this fiercely competitive environment, many platforms adopt flexible pricing strategies, known as dynamic pricing. However, while these strategies bring commercial benefits, they may also cross legal lines, such as price discrimination. To avoid unnecessary legal trouble and regulatory penalties, platforms need an effective method to assess the compliance of their pricing strategies.

[0003] Against the backdrop of the rapid development of the platform economy, e-commerce platforms often adopt dynamic pricing strategies to enhance their market competitiveness. Some of these strategies may involve price discrimination, requiring compliance assessments to mitigate legal risks. Judicial big data contains a wealth of case law, legal provisions, and regulatory updates related to price discrimination. Deeply integrating this data with the platform's real-time transaction data can provide quantitative support for the legality and rationality of pricing strategies, thereby encouraging platforms to optimize business decisions within a compliant framework.

[0004] However, existing data management systems fall short in supporting the complex needs of price discrimination analysis. Specifically, legal documents (such as court judgments and legal provisions) and real-time transaction data on the platform (such as user purchasing behavior, payment prices, and purchase times) differ significantly in structure, expression, and volume. Traditional storage methods struggle to quickly integrate and manage this heterogeneous data from diverse sources. Furthermore, the lack of specialized analytical methods to link legal precedents with specific transaction behaviors makes it difficult for the system to accurately identify risk points related to pricing behavior from a large number of legal precedents. This results in low query efficiency and coarse, unrefined analytical results. More importantly, traditional databases are slow to respond to constantly changing compliance requirements, leaving platforms always a step behind in adjusting pricing strategies, thus facing a higher risk of legal disputes and regulatory penalties. Summary of the Invention

[0005] The purpose of this invention is to address the legal compliance risks and price discrimination issues that may arise when e-commerce platforms adopt dynamic pricing strategies, and to propose a judicial big data management method and system for price discrimination analysis.

[0006] This invention is achieved through the following technical solution: A judicial big data management method for price discrimination analysis includes the following steps: S1. Obtain transaction data streams from e-commerce platforms; S2. Construct a compliance correction rule set, which includes price discrimination judgment criteria and correction actions; S3. Based on the e-commerce platform transaction data stream and the compliance correction rule set, detect potential price discrimination risks; S4. Based on the potential price discrimination risk and the compliance correction rule set, generate pricing parameter adjustment suggestions; S5. Send the proposed pricing parameter adjustment to the pricing system to correct the pricing strategy, and receive the result of the pricing system executing the correction to form a compliance assessment and response closed loop.

[0007] A judicial big data management system for price discrimination analysis, the system comprising: The data acquisition module is used to acquire transaction data streams from e-commerce platforms; The rule building module is used to build a compliance correction rule set, which includes price discrimination judgment criteria and correction actions; The risk detection module is used to detect potential price discrimination risks based on the e-commerce platform's transaction data stream and the compliance correction rule set. The suggestion generation module is used to generate pricing parameter adjustment suggestions based on the potential price discrimination risk and the compliance correction rule set; The strategy correction module is used to send the proposed adjustment of pricing parameters to the pricing system to correct the pricing strategy, and to receive the result of the adjustment after the pricing system has executed the correction, so as to form a compliance assessment and response closed loop.

[0008] This invention offers the following advantages: The judicial big data management method and system disclosed in this application, designed for price discrimination analysis, acquires transaction data streams from e-commerce platforms and constructs a compliance correction rule set containing price discrimination judgment standards and corrective actions. Based on this data and these rules, potential price discrimination risks can be detected. Subsequently, based on the detected risks and the rule set, pricing parameter adjustment suggestions are generated and sent to the pricing system to correct the pricing strategy. Finally, the system receives the adjusted results from the pricing system, thus forming a closed loop of compliance assessment and response. This method effectively solves the problem in existing technologies where dynamic pricing strategies of e-commerce platforms may cross legal red lines and exhibit price discrimination without effective assessment and correction mechanisms. By deeply integrating the case law, legal provisions, and regulatory dynamics contained in judicial big data with the platform's real-time transaction data, this application can provide quantitative support for the legality and rationality of pricing strategies, thereby promoting platforms to optimize business decisions within a compliance framework, effectively avoiding legal risks and regulatory penalties, and achieving a balance between business benefits and compliance. Attached Figure Description

[0009] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0011] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0012] To better understand the judicial big data management method for price discrimination analysis proposed in this application, it is necessary to explain some of the key terms involved.

[0013] Please see Figure 1 As shown, a judicial big data management method for price discrimination analysis includes the following steps: S1. Obtain transaction data streams from e-commerce platforms; The "e-commerce platform transaction data stream" mentioned in this application refers to transaction-related data generated and continuously transmitted in real time on e-commerce platforms, including but not limited to product browsing history, purchase history, payment information, user profiles, geographical location, device information, historical orders, and coupon usage. This data is continuously fed into the system in a streaming format, providing real-time data support for subsequent price discrimination analysis.

[0014] S2. Construct a compliance correction rule set, which includes price discrimination judgment criteria and correction actions; The "Compliance Correction Rule Set" is a core component of this application, containing two key types of information: first, "Price Discrimination Judgment Standards," which are specific rules and thresholds extracted from laws, regulations, industry standards, and judicial precedents used to identify and quantify price discrimination behavior; and second, "Corrective Actions," which are pricing parameter adjustment strategies or suggestions that the system should take in response to detected price discrimination risks, such as adjusting product prices, restricting coupon usage, or modifying recommendation algorithm weights. This rule set is dynamically updatable and can be adjusted according to the latest laws, regulations, and market environment.

[0015] S3. Based on the e-commerce platform transaction data stream and the compliance correction rule set, detect potential price discrimination risks; "Potential price discrimination risk" refers to the possibility that the platform may be deemed to be engaging in price discrimination under the current pricing strategy. This risk is detected by comparing and analyzing the e-commerce platform's transaction data stream with its compliance modification rule set, aiming to identify pricing patterns or differentiated pricing for user groups that may violate fair trade principles.

[0016] S4. Based on the potential price discrimination risk and the compliance correction rule set, generate pricing parameter adjustment suggestions; "Pricing parameter adjustment suggestions" are a series of specific operational instructions automatically generated by the system based on detected potential price discrimination risks and compliance correction rule sets. These instructions guide the pricing system to modify existing pricing strategies to eliminate or reduce price discrimination risks. These suggestions can be direct price adjustment instructions or instructions that affect internal parameters of the pricing model.

[0017] S5. Send the proposed pricing parameter adjustment to the pricing system to correct the pricing strategy, and receive the result of the pricing system executing the correction to form a compliance assessment and response closed loop.

[0018] The "pricing system" refers to the automated system used by e-commerce platforms to perform functions such as product pricing, coupon distribution, and promotional activity management. This system receives pricing parameter adjustment suggestions generated by this method and modifies its pricing strategy accordingly.

[0019] The "compliance assessment and response closed loop" refers to a continuously iterative cycle formed throughout the entire process of data acquisition, risk detection, recommendation generation, strategy revision, and result feedback. By receiving the adjustment results after the pricing system has implemented the corrections, this method can evaluate the effectiveness of the corrections and further optimize the compliance correction rule set based on the evaluation results, thereby achieving continuous management and dynamic response to price discrimination risks.

[0020] In practical implementation, firstly, the system continuously acquires transaction data streams from the e-commerce platform. This can be achieved by connecting to the e-commerce platform's data interface to receive user behavior data, product transaction data, payment data, etc., in real time. For example, a data collection agent can be deployed to capture data generated by users during browsing products, adding items to their carts, placing orders, and making payments, and transmit this data to the data processing center in streaming form. Secondly, a compliance correction rule set is constructed. This rule set is the core of this method; it defines what constitutes price discrimination and how to correct it. For example, definitions, constituent elements, and exemptions of price discrimination can be manually entered or extracted from laws, regulations, and judicial precedents using natural language processing technology, and then converted into calculable judgment standards. Simultaneously, corresponding correction actions are preset for different levels of price discrimination risk, such as "increasing the price of a specific product for a specific user by X%" or "disabling type Y coupons for a specific user group." Next, based on the e-commerce platform transaction data stream and the compliance correction rule set, potential price discrimination risks are detected. Specifically, the system analyzes the incoming transaction data stream in real time and compares it with the price discrimination judgment standards in the compliance correction rule set. For example, machine learning models can be used to compare the prices paid by different user groups when purchasing the same goods. If a significant and unjustified price difference is found, it is marked as a potential price discrimination risk. Subsequently, based on the potential price discrimination risk and the compliance correction rule set, pricing parameter adjustment suggestions are generated. When a potential price discrimination risk is detected, the system automatically generates specific pricing parameter adjustment suggestions based on the type and severity of the risk and the pre-set correction actions in the compliance correction rule set. For example, if price discrimination against new users is detected, the system may suggest adjusting the price of goods for new users to the same level as for existing users, or issuing equivalent coupons to new users to offset the price difference. Finally, the pricing parameter adjustment suggestions are sent to the pricing system to correct the pricing strategy, and the system receives the adjustment results after the pricing system has implemented the correction, thus forming a compliance assessment and response closed loop. The generated adjustment suggestions are sent to the e-commerce platform's pricing system via API interface, and the pricing system corrects its pricing strategy based on the suggestions. After the correction is implemented, the pricing system will feed back the adjustment results (such as actual price changes, user feedback, etc.) to this method. This method will then evaluate the effect of the correction and optimize the compliance correction rule set based on the evaluation results, thus forming a continuously iterative closed-loop management mechanism. This application's judicial big data management method for price discrimination analysis, by constructing a compliance correction rule set and combining it with the transaction data flow of e-commerce platforms, achieves proactive detection and correction of potential price discrimination risks. Compared to the passive response and post-event remedy models in existing technologies, this application can provide early warning and real-time intervention, thereby effectively reducing the legal risks and reputational damage faced by the platform.By establishing a closed loop for compliance assessment and response, this approach continuously optimizes the compliance of pricing strategies, ensuring that platforms, while pursuing commercial benefits, consistently adhere to laws and regulations and maintain a fair and competitive market environment. This proactive and intelligent compliance management significantly enhances e-commerce platforms' risk response capabilities and sustainable development potential in complex market environments.

[0021] In some of the embodiments described above in this application, pricing parameter adjustment suggestions are proposed based on potential price discrimination risks and compliance modification rule sets. However, in the implementation process, if only a static rule set is relied upon or a deep understanding of legal texts is lacking, the generated suggestions may fail to accurately capture complex price discrimination patterns and may be difficult to effectively address the dynamic changes in laws and regulations and the complexity of the market environment. This may result in insufficient targeting of the adjustment suggestions, failing to fundamentally solve the price discrimination problem, and may even cause unnecessary negative impacts on business benefits while pursuing compliance.

[0022] In response, this application further proposes a judicial big data management method for price discrimination analysis, wherein step S4 includes: S41. Perform semantic analysis on legal texts to identify legal concepts, contextual elements, and key values; S42. Based on the identified legal concepts, contextual elements, and key values, define contextualized rule templates to map legal principles to the observable input features, internal variables, or decision logic of specific modules in the pricing model. S43. Calculate the contribution of each input feature to the final price prediction and reveal the interaction between features; S44. Based on the calculated contribution and the revealed characteristic interactions, extract the impact links that lead to price differences, and quantify the contribution of each pricing module to the price differences. S45. Based on the quantified contribution, generate feature weight adjustment instructions or loss function penalty term instructions; store the generated feature weight adjustment instructions or loss function penalty term instructions as part of the compliance correction rule set; S46. Based on the changes in the compliance correction rule set, trigger the retraining of the pricing model; S47. Monitor the compliance performance of the pricing model and optimize the compliance correction rule set based on the monitoring results.

[0023] Specifically, semantic parsing of legal texts refers to using Natural Language Processing (NLP) technology to conduct in-depth analysis of relevant laws, regulations, judicial interpretations, precedents, and other texts. This involves automatically extracting legal concepts related to price discrimination (such as "equivalent terms of trade," "reasonable grounds," and "discriminatory pricing"), contextual elements (such as "user groups," "product types," and "transaction time"), and key numerical values ​​(such as "price difference threshold" and "discount cap"). The aim is to transform unstructured legal knowledge into structured information that can be understood and processed by machines.

[0024] In this process, defining contextualized rule templates based on identified legal concepts, contextual elements, and key values ​​concretizes abstract legal principles into actionable rules. Specifically, legal principles (such as "prohibition of discriminatory treatment without just cause") are mapped to observable input features of the pricing model (such as user location and purchase history), internal variables (such as user value ratings), or the decision logic of specific modules (such as weight allocation in dynamic pricing algorithms). For example, a template could be defined as: "If user A and user B are considered to have equal transaction conditions under contextual element X, and their product price difference exceeds key value Y, then this constitutes potential price discrimination unless there is a just cause permitted by legal concept Z." The aim is to establish a direct link between legal requirements and the behavior of the pricing model.

[0025] In practical applications, calculating the contribution of each input feature to the final price prediction and revealing the interactions between features can be understood as using interpretable artificial intelligence (XAI) techniques, such as LIME, SHAP values, or causal inference methods, to analyze the degree of influence of each input feature (such as user profile, product attributes, purchase history, etc.) in the pricing model on the final price prediction result. Simultaneously, revealing the interactions between features refers to identifying how different combinations of features collectively affect prices; for example, price differences may be amplified when users in a specific region purchase a specific product at a specific time. The aim is to provide transparency and interpretability in pricing decisions.

[0026] Furthermore, based on the calculated contribution and revealed feature interactions, the impact chains leading to price differences are extracted, and the contribution of each pricing module to price differences is quantified. This involves tracing the decision-making process of the pricing model to identify the complete path from input features to the final price difference. For example, it can be found that user geographic characteristics jointly lead to price differences by influencing the "freight calculation module" and the "promotion strategy module." Quantifying contribution means assessing the specific weight or responsibility of each module in the formation of price differences. The aim is to accurately pinpoint the root cause of price discrimination, rather than merely identifying the surface symptoms.

[0027] Therefore, generating feature weight adjustment instructions or loss function penalty instructions based on the quantified contribution refers to generating specific corrective instructions for the identified key features or pricing modules that lead to price discrimination. Feature weight adjustment instructions can instruct the pricing model to reduce the weight of a certain feature (e.g., user location) in pricing decisions, thereby reducing its impact on price differences. Loss function penalty instructions can introduce a penalty term during model training; when the price differences generated by the model exceed a compliance threshold, the model's loss is increased, thus prompting the model to pay more attention to compliance while optimizing business benefits. The aim is to provide executable and targeted technical correction solutions.

[0028] Storing the generated feature weight adjustment instructions or loss function penalty term instructions as part of the aforementioned compliance correction rule set means incorporating these dynamically generated instructions, used to correct pricing strategies, into the compliance correction rule set maintained by the system. This makes the rule set no longer static, but capable of self-updating and improving based on actual detection and analysis results. Its purpose is to achieve dynamic management and continuous optimization of the rule set.

[0029] Triggering pricing model retraining based on changes in the aforementioned compliance correction rule set means that when the instructions in the compliance correction rule set are updated, the system automatically initiates the retraining process of the pricing model. Through retraining, the pricing model can learn and internalize the new compliance requirements and correction instructions, thereby directly reflecting these adjustments in future pricing decisions. The purpose is to ensure that the pricing model always remains consistent with the latest compliance requirements.

[0030] Finally, monitoring the compliance performance of the pricing model and optimizing the aforementioned compliance correction rule set based on the monitoring results involves continuously collecting transaction data and user feedback after the pricing model is deployed and running, and evaluating the model's compliance indicators (such as price differences among different user groups and complaint rates) and business efficiency indicators in actual operation. Based on the monitored performance, the effectiveness of the rule set is further analyzed and iteratively optimized to form a closed loop of continuous improvement. The purpose is to ensure that the system can adapt to the ever-changing regulatory environment and market demands.

[0031] Through the aforementioned technical solution, this application significantly improves the accuracy and effectiveness of pricing parameter adjustment suggestions, ensuring that the generated suggestions not only strictly comply with legal and regulatory requirements but also consider commercial benefits. By providing transparency and explainability in pricing decisions, this solution enables the platform to clearly understand the sources of price differences, allowing for targeted corrections and avoiding the negative impacts of blind adjustments. Furthermore, by dynamically updating the rule set and triggering model retraining, this application endows the system with strong adaptive capabilities, enabling rapid responses to changes in the legal and regulatory environment and the emergence of new price discrimination patterns, thereby continuously reducing the platform's legal and operational risks and enhancing user trust.

[0032] In some of the embodiments described above in this application, simply monitoring and optimizing may not effectively balance the potential conflict between compliance and commercial benefits, nor may it be possible to find the optimal rule set configuration under a dynamically changing legal and regulatory environment and market competition landscape, which may lead to insufficient compliance or impaired commercial benefits. To address this, this application further proposes a more refined and intelligent method for monitoring the compliance performance of the pricing model and optimizing the compliance correction rule set based on the monitoring results, thereby achieving synergistic optimization of compliance and commercial benefits.

[0033] Step S47 includes: S4711. Define a multi-objective optimization function, which simultaneously considers compliance indicators and business benefit indicators; S4712. Set compliance hard constraints, wherein the compliance hard constraints transform the price discrimination behaviors prohibited by laws and regulations into the constraints of the multi-objective optimization function; S4713. Dynamically adjust the rule set weights, which are adjusted according to the legal and regulatory environment, market competition, and platform strategic objectives. S4714. Explore the rule set parameter space, wherein the rule set parameter space includes the price difference threshold, discount adjustment range, feature weight adjustment coefficient, or the λ value of the loss function penalty term; S4715. Using the improved multi-objective optimization algorithm NSGA-II, under the aforementioned compliance hard constraints, explore the rule set parameter space to find the Pareto optimal rule set configuration. S4716. Show the Pareto optimal rule set configuration and its corresponding compliance and business benefit indicators; S4717. Select an optimal configuration and deploy it to the compliance correction rule set; S4718. Continuously monitor the compliance performance and business benefits of the pricing model after applying the new rule set during operation.

[0034] Specifically, defining a multi-objective optimization function involves constructing a mathematical model that can simultaneously quantify and evaluate the compliance performance of a pricing strategy and its resulting business benefits. Compliance indicators may include, but are not limited to, a fairness index of price differences between different user groups, and a price discrimination risk score for specific products; business benefit indicators may include sales revenue, profit margin, market share growth rate, and user retention rate. By incorporating these interrelated but potentially conflicting objectives into a single optimization framework, it is possible to ensure strict compliance with laws and regulations while pursuing commercial interests.

[0035] Setting hard compliance constraints refers to transforming price discrimination practices explicitly prohibited by national or regional laws and regulations—such as differentiated pricing based on sensitive user characteristics like region, gender, or race—into insurmountable limitations in the optimization process. These constraints ensure that regardless of the optimization process, the final pricing strategy must meet basic legal compliance requirements to avoid crossing legal red lines.

[0036] In practical applications, dynamically adjusting the rule set weights refers to flexibly adjusting the relative importance of compliance indicators and business benefit indicators in a multi-objective optimization function based on changes in the external environment, such as the introduction of new laws and regulations, intensified market competition, or adjustments to the platform's business development priorities. For example, when regulations become stricter, the weight of compliance indicators can be increased; when market competition is fierce, the weight of business benefit indicators can be appropriately increased to adapt to different operational strategies and external environments.

[0037] Furthermore, exploring the rule set parameter space refers to systematically searching for adjustable parameter combinations within the compliance correction rule set. These parameters directly influence the behavior and output of the pricing model. For example, the price difference threshold can be defined as the maximum acceptable price difference between different user groups; the discount adjustment magnitude controls the flexibility of the pricing system when correcting prices; the feature weight adjustment coefficient affects the degree of influence of specific user characteristics on pricing decisions; and the λ value of the penalty term in the loss function is used to balance compliance penalties and prediction accuracy during model training. By comprehensively exploring these parameter combinations, potential optimal configurations can be discovered.

[0038] As a preferred implementation, the improved multi-objective optimization algorithm NSGA-II refers to a variant of the Non-dominated Sorting Genetic Algorithm II, which is particularly suitable for solving optimization problems with multiple conflicting objectives. Under the aforementioned compliance hard constraint, NSGA-II can efficiently search the parameter space of the rule set to identify a set of Pareto optimal solutions. Each configuration in the Pareto optimal solution set represents a solution that achieves the best balance between compliance and business benefits, i.e., one that cannot improve another objective without sacrificing one.

[0039] Therefore, displaying the Pareto optimal rule set configuration and its corresponding compliance and business benefit indicators means clearly presenting the multiple optimal configurations found by the algorithm and their performance on various objectives to decision-makers. This is usually achieved through a visual interface, enabling decision-makers to intuitively compare the advantages and disadvantages of different configurations and understand their trade-offs between compliance and business benefits.

[0040] Specifically, selecting an optimal configuration and deploying it to the compliance correction rule set means that the decision-maker, based on the presented Pareto optimal solution set, combined with current business needs and risk preferences, selects the most suitable rule set configuration from multiple optimal solutions and applies it to the actual pricing system.

[0041] Finally, continuous monitoring of the pricing model's compliance performance and business benefits after applying the new rule set refers to the system continuously collecting data after the new rule set is deployed to assess in real time whether the pricing model still meets compliance requirements in actual operation and its impact on business benefits. This continuous monitoring mechanism ensures the effectiveness and adaptability of the rule set and provides data support for subsequent iterative optimization.

[0042] In some of the embodiments described above in this application, a method for monitoring the compliance performance of a pricing model and optimizing the compliance correction rule set using a multi-objective optimization algorithm is proposed. However, in the actual e-commerce operation environment, user behavior patterns, product market dynamics, and legal regulatory requirements are constantly changing. If the monitoring and optimization process cannot respond to these changes in a timely and dynamic manner, the system may be unable to effectively identify new price discrimination patterns, thereby affecting the accuracy of compliance assessment and the effectiveness of correction strategies. Therefore, this application further proposes a more dynamic and adaptive monitoring and optimization mechanism. Step S47 further includes: S4721. Continuously receive transaction data, user behavior data, and legal and regulatory updates from e-commerce platforms; S4722. Based on the transaction data, user behavior data, and legal and regulatory dynamics, identify user behavior patterns, commodity market dynamics, and changes in legal and regulatory requirements; S4723. Generate or adjust compliance monitoring indicators and business benefit monitoring indicators based on identified user behavior patterns, product market dynamics, and changes in legal and regulatory requirements. S4724. Adjust the data collection frequency, analysis granularity, and risk warning threshold based on the generated compliance monitoring indicators and business benefit monitoring indicators. S4725. Identify new price discrimination patterns based on the adjusted data collection frequency, analysis granularity, and risk warning threshold. S4726. Based on the identified new price discrimination pattern, trace the impact chain and attribute the key user characteristics, product attributes and pricing module decisions that led to the new pattern. S4727. Generate new compliance correction rules or update existing rules based on the traced impact chain and attribution results; S4728. Adjust the weights of compliance and business benefit indicators in the multi-objective optimization function based on the generated new compliance correction rules or updated existing rules.

[0043] Specifically, continuously receiving transaction data, user behavior data, and legal and regulatory updates from e-commerce platforms means that the system is configured to acquire data in real-time or near real-time from the e-commerce platform's data interfaces, user behavior log systems, and external legal and regulatory databases or information sources released by regulatory agencies. Transaction data can include product prices, purchase times, user IDs, and product IDs; user behavior data can include browsing history, click-through rates, shopping cart addition records, and geographic location; legal and regulatory updates can include newly promulgated laws and regulations, judicial interpretations, and administrative penalty cases. The purpose is to provide comprehensive and real-time information input for subsequent analysis.

[0044] The process of identifying user behavior patterns, commodity market dynamics, and changes in legal and regulatory requirements based on transaction data, user behavior data, and legal and regulatory dynamics can be understood as processing the received data using data analysis and machine learning techniques. For example, identifying user behavior patterns can utilize methods such as cluster analysis and sequence pattern mining to discover the purchasing preferences and price sensitivity of different user groups; identifying commodity market dynamics can employ methods such as time series analysis and trend forecasting to monitor commodity price fluctuations, supply and demand changes, and competitor pricing strategies; and identifying changes in legal and regulatory requirements can utilize Natural Language Processing (NLP) technology to semantically analyze legal texts and extract key clauses, prohibited behaviors, and compliance requirements. The aim is to extract in-depth information from massive amounts of data that provides guidance for price discrimination analysis and compliance management.

[0045] In practical applications, generating or adjusting compliance monitoring indicators and business benefit monitoring indicators based on identified user behavior patterns, product market dynamics, and changes in legal and regulatory requirements means that the system can flexibly define or modify indicators used to evaluate the performance of pricing models based on an understanding of these dynamic changes. For example, when a change in the price sensitivity of a specific user group is identified, a price difference tolerance indicator can be generated for that group; when a new competitive landscape emerges in the market, the weight of market share or profit margin in the business benefit indicators can be adjusted. The purpose is to ensure that the monitoring indicators accurately reflect compliance risks and business opportunities in the current environment.

[0046] Furthermore, adjusting the data collection frequency, analysis granularity, and risk warning threshold based on the generated compliance monitoring indicators and business benefit monitoring indicators means that the system dynamically adjusts the granularity and response speed of data processing according to the current monitoring focus and the identified risk level. For example, when a high-risk area is identified, the collection frequency of relevant data can be increased, the analysis granularity can be refined from product category to specific product SKU, and the risk warning threshold can be lowered to detect potential problems earlier. The aim is to optimize resource allocation and improve the efficiency and accuracy of risk identification.

[0047] Therefore, identifying new price discrimination patterns based on adjusted data collection frequency, analysis granularity, and risk warning thresholds means that the system utilizes more refined data and a more sensitive warning mechanism to discover previously unidentified or newly emerging price discrimination behaviors. This may involve anomaly detection algorithms, pattern matching techniques, or rule-based inference engines used to compare actual pricing behavior with compliance requirements, thereby revealing new and hidden price discrimination patterns. The aim is to achieve continuous and comprehensive coverage of price discrimination risks.

[0048] This process involves tracing the impact chain based on identified new price discrimination patterns, attributing the causes to key user characteristics, product attributes, and pricing module decisions. This can be understood as conducting an in-depth analysis of the identified new patterns to find their root causes. This may involve techniques such as causal inference, feature importance analysis, and decision path tracing to determine which user characteristics (e.g., membership level, purchase history), product attributes (e.g., scarcity, brand effect), or specific module decisions within the pricing system (e.g., the setting of a certain algorithm parameter) led to this new price difference. The aim is to provide accurate attribution evidence for subsequent rule revisions.

[0049] Specifically, generating new compliance correction rules or updating existing rules based on the traced impact chain and attribution results means that, based on a clear understanding of the causes of price discrimination patterns, the system can automatically or semi-automatically generate specific correction instructions. For example, if it is found that a certain user characteristic is being overused to cause price differences, a new rule can be generated to limit the weight of that characteristic in pricing; if existing rules fail to cover a new scenario, existing rules can be updated to include that scenario. The aim is to intervene directly at the root cause of the problem and improve the effectiveness of the correction.

[0050] Finally, adjusting the weights of compliance and business benefit indicators in the multi-objective optimization function based on the generated new compliance correction rules or updated existing rules means feeding the new rule changes back into the overall optimization framework. For example, if the implementation of new rules has a significant impact on business benefits, the system can accordingly adjust the relative weights of compliance and business benefit indicators in the multi-objective optimization function to seek a new optimal balance under the new constraints. The aim is to form a complete adaptive closed loop, ensuring that the system pursues compliance while also considering business benefits.

[0051] In practical applications, if these monitoring indicators are not dynamically and meticulously constructed and adjusted for specific user groups and product categories, as well as constantly changing regulatory requirements, it may lead to inaccurate identification of price discrimination risks, or cause unnecessary negative impacts on business benefits while pursuing compliance. If the above problems are not addressed, it may be impossible to effectively balance compliance and business benefits, affecting the flexibility of pricing strategies and market competitiveness. To address this, this application further proposes a more refined indicator generation and adjustment mechanism. Specifically, step S4723 includes: S47231. Continuously receive transaction data, user behavior data, and legal and regulatory updates from e-commerce platforms; based on the transaction data, user behavior data, and legal and regulatory updates, identify user behavior patterns, commodity market dynamics, and changes in legal and regulatory requirements; S47232. Based on identified user behavior patterns, product market dynamics, and changes in legal and regulatory requirements, dynamically construct or adjust compliance monitoring indicators and business benefit monitoring indicators, wherein the dynamic construction or adjustment includes: S47233. Based on the identified user behavior patterns, generate compliance monitoring indicators for segmented user groups, wherein the compliance monitoring indicators include the difference ratio between the average payment price of a specific product and the control group. S47234. Based on the identified commodity market dynamics, generate business efficiency monitoring indicators for specific commodity categories. The business efficiency monitoring indicators include commodity sales growth rate and inventory digestion rate, and are associated with compliance constraints. S47235. Based on the generated compliance monitoring indicators and business benefit monitoring indicators, adjust the data collection frequency, analysis granularity, and risk warning threshold; S47236. Identify new price discrimination patterns based on the adjusted data collection frequency, analysis granularity, and risk warning threshold. S47237. Based on the identified new price discrimination pattern, trace the impact chain and attribute the user characteristics, product attributes and pricing module decisions that led to the new pattern. S47238. Based on the traced impact chain and attribution results, generate new compliance correction rules or update existing rules; S47239. Adjust the weights of compliance and business benefit indicators in the multi-objective optimization function based on the generated new compliance correction rules or updated existing rules.

[0052] The "dynamic construction or adjustment of compliance monitoring indicators and business benefit monitoring indicators" refers to the system's ability to flexibly create, modify, or delete quantitative standards for evaluating the compliance and business effectiveness of pricing strategies based on real-time or near-real-time data input. This dynamism ensures that the monitoring system can adapt to constantly changing market environments, user behaviors, and laws and regulations. Specifically, "generating compliance monitoring indicators for segmented user groups based on identified user behavior patterns" aims to refine the segmentation of user groups, for example, by creating different user profiles based on characteristics such as user purchase history, browsing preferences, geographical location, and device type. For these segmented groups, specific compliance indicators are constructed, such as the "difference ratio between the average payment price of a specific product and the control group." This ratio measures the degree of difference between the average payment price of a specific user group when purchasing a specific product and a comparable control group that has not been subjected to potentially discriminatory pricing strategies, in order to determine whether there are unreasonable pricing differences. "Generating business benefit monitoring indicators for segmented product categories based on identified product market dynamics" refers to classifying products based on market information such as sales trends, inventory status, seasonality, and competition. For these specific product categories, business performance indicators are constructed, such as "sales growth rate" and "inventory clearance rate." These indicators directly reflect the impact of pricing strategies on business operations. Furthermore, these business performance indicators are subject to "compliance constraints," meaning that while pursuing sales growth and inventory clearance, it is essential to ensure that pricing practices comply with laws and regulations to avoid triggering price discrimination risks.

[0053] This application's solution addresses the issue of insufficiently granular monitoring indicators by introducing a dynamic indicator construction mechanism targeting specific user groups and product categories. Specifically, by generating compliance monitoring indicators for specific user groups based on identified user behavior patterns—such as the difference ratio between the average payment price of a specific product and the control group—it can more accurately capture potential price differences between different user groups, thereby effectively identifying potential price discrimination risks. Simultaneously, based on identified product market dynamics, it generates business benefit monitoring indicators for specific product categories, such as product sales growth rate and inventory turnover rate, and links them to compliance constraints to ensure that compliance requirements are not violated while pursuing business benefits. This dynamic and granular indicator system enables the system to more sensitively perceive changes in the market and regulatory environment, and adjust data collection frequency, analysis granularity, and risk warning thresholds accordingly, thereby identifying more complex and hidden price discrimination patterns. By tracing the impact chain of these new patterns and attributing them to key factors, the system can generate or update compliance correction rules and adjust the weights of compliance and business benefit indicators in the multi-objective optimization function accordingly, forming a continuously optimizing, self-adaptive closed-loop management process.

[0054] In some of the above embodiments, although it is proposed to trace the influencing links and attribute the key user characteristics, product attributes, and pricing module decisions that lead to the new price discrimination pattern based on the identified new price discrimination pattern, in the complex e-commerce environment, the formation of price discrimination patterns is often the result of multiple factors and dynamic interactions. Simply tracing and attributing may not be enough to fully reveal its underlying causes and mechanisms. If these influencing links and their contributions are not accurately and deeply understood, the subsequently generated compliance correction rules may lack specificity or effectiveness, thereby affecting the efficiency and accuracy of the entire compliance assessment and response closed loop. In this regard, this application further proposes the above-mentioned method of tracing the new price discrimination pattern based on the identified new price discrimination pattern. Specifically, step S47239 includes: S472391. Conduct correlation analysis on transaction data, user behavior data, and pricing module decision logs to construct a dynamic causal network; the construction of the dynamic causal network adopts an improved information theory-based method. S472392. Perform an influence path search in the dynamic causal network to identify potential influence links; quantify the contribution of each influence link to the price difference; S472393. Based on the quantified contribution, identify the cumulative and synergistic effects that lead to the new price discrimination pattern; S472394. Based on the identified cumulative and synergistic effects, generate an attribution report that includes key user characteristics, product attributes, and pricing module decisions.

[0055] Specifically, correlation analysis is performed on transaction data, user behavior data, and pricing module decision logs to discover the inherent connections and interactions between different data sources. Transaction data can include product prices, purchase times, and purchase quantities; user behavior data can include browsing history, click behavior, search keywords, and geographic location; and the pricing module decision logs record the internal decision-making process of the pricing system when generating prices, the feature values ​​used, and the final pricing result. By integrating and analyzing these heterogeneous data, a dynamic causal network reflecting the dynamic causal relationships between various factors can be constructed. As a preferred implementation, the construction of the dynamic causal network employs an improved information theory method. This method, by calculating mutual information and conditional mutual information indicators between different variables, can more accurately capture the nonlinear dependencies between variables and the causal effects over time series, thus overcoming the limitations of traditional correlation analysis in revealing causal relationships and effectively handling high-dimensional and complex data.

[0056] After the dynamic causal network is constructed, an impact path search is performed to identify potential impact links. Impact path search refers to tracing the causes and consequences in the causal network, starting from a specific outcome node that leads to a price difference, either backwards or forwards, to reveal a series of intermediate steps and key factors leading to that outcome. For example, graph traversal algorithms (such as depth-first search or breadth-first search) combined with causal strength assessment can be used to discover the complete path from user characteristics, product attributes, or pricing module decisions to the final price difference. These potential impact links reveal how price discrimination patterns gradually form and evolve.

[0057] Furthermore, it is necessary to quantify the contribution of each of the aforementioned influencing links to the price difference. Contribution quantification aims to assess the importance of each identified influencing link in causing the price difference. This can be achieved through various techniques, such as using a Shapley value-based approach to fairly distribute the overall effect of the price difference across the various influencing links; utilizing counterfactual analysis to observe the impact on the price difference by changing the variable values ​​on specific links; or employing gradient-based or sensitivity analysis methods to assess the sensitivity of key nodes in the links to the final price. By quantifying contribution, it becomes clear which factors or paths have the greatest impact on the price difference.

[0058] Based on this, and according to the quantified contribution, the cumulative and synergistic effects leading to the new price discrimination pattern will be identified. The cumulative effect refers to the phenomenon where multiple factors or decisions gradually accumulate over time, ultimately resulting in price differences; the synergistic effect refers to the combined effect of multiple factors or decisions being greater than the sum of their individual effects. Identifying these effects helps to more comprehensively understand the complex mechanisms of price discrimination. For example, a user characteristic may have little impact on its own, but when combined with specific product attributes and pricing strategies, it may produce significant discriminatory pricing.

[0059] Finally, based on the identified cumulative and synergistic effects, an attribution report will be generated, including key user characteristics, product attributes, and pricing module decisions. This attribution report is a structured summary of the above analysis results, clearly listing the key factors leading to the new price discrimination pattern, their influence chains, their respective contributions, and the analysis results of cumulative and synergistic effects. This report provides data support and decision-making basis for the subsequent development of precise compliance correction rules.

[0060] In some of the embodiments described above in this application, attribution reports containing key user characteristics, product attributes, and pricing module decisions are generated. However, simply generating a report may not be sufficient to efficiently guide decision-makers in making precise strategy adjustments. In practical applications, decision-makers often need clearer, more prioritized insights, as well as tools that allow them to quickly obtain the necessary information based on their responsibilities, in order to avoid information overload or difficulty in locating key issues. If the above problems are not addressed, the generated attribution reports may be difficult to utilize effectively, thereby affecting the timeliness and accuracy of compliance corrections. Therefore, this application further proposes an optimized method for generating and displaying attribution reports. Specifically, step S472394 includes: S4723941. Based on the cumulative and synergistic effects of the identification, prioritize the key user characteristics, the product attributes, and the pricing module decisions to obtain the priority ranking results; S4723942. Based on the priority ranking results, generate a structured attribution report. The structured attribution report includes a description of the price discrimination pattern, the top N key user characteristics in terms of contribution and their impact on price differences, the top N key product attributes in terms of contribution and their impact on price differences, the top N key pricing module decisions in terms of contribution and their impact on price differences, and targeted correction suggestions. S4723943. Based on the structured attribution report, generate an interactive visualization interface. The interactive visualization interface allows decision-makers to explore the detailed information of the structured attribution report through filtering and drill-down operations, and supports customized display of the content of the structured attribution report according to user roles.

[0061] Specifically, prioritizing key user characteristics, product attributes, and pricing decisions involves evaluating them across multiple dimensions, including their contribution to price discrepancies, scope of impact, difficulty of correction, and potential legal risks. This process identifies the primary drivers of price discrimination patterns and prioritizes their attention and action. For example, characteristics or decisions with higher contributions, as well as those that would significantly improve compliance after correction, will be given higher priority.

[0062] The structured attribution report aims to provide a clear and easy-to-understand reporting format. This report not only describes the identified price discrimination patterns but also details the top N contributing key user characteristics, product attributes, and pricing module decisions, quantifying their specific impact on price differences. Furthermore, the report provides targeted corrective recommendations that can directly guide the pricing system in strategy adjustments, such as adjusting the weight of specific features or modifying pricing logic.

[0063] In practical applications, interactive visualization interfaces can be understood as a dynamic data display platform. This platform allows decision-makers to filter data by specific time periods, user groups, or product categories, and to perform drill-down operations to gain a deeper understanding of detailed data on a particular characteristic or decision. Its purpose is to enable users with different roles (such as legal compliance departments, business operations departments, and senior management) to customize their report viewing based on their interests, thereby improving the efficiency of information acquisition and the accuracy of decision-making.

[0064] This application's solution prioritizes identified cumulative and synergistic effects, ensuring a rapid focus on the most critical and impactful user characteristics, product attributes, and pricing decisions amidst complex attribution results. This prioritization allows the subsequently generated structured attribution reports to highlight key points and avoid information redundancy, providing decision-makers with concise and instructive corrective suggestions. Furthermore, by constructing an interactive visual interface, this application further addresses the static and impersonal nature of traditional reports, enabling decision-makers in different roles to flexibly explore and understand the detailed information in the attribution reports according to their responsibilities and concerns, thereby significantly improving the practicality of the attribution reports and the efficiency of decision support.

[0065] In some embodiments described above in this application, an attribution report containing key user characteristics, product attributes, and pricing module decisions is generated based on the identified cumulative and synergistic effects, and a structured attribution report is further generated. However, in practical applications, this structured attribution report may contain sensitive transaction data, user behavior data, or business strategy information. If not properly managed and protected, this sensitive information may face risks of data leakage, misuse, or improper disclosure when transmitted, stored, or displayed to different user roles, thereby causing compliance issues and business losses. Therefore, this application further proposes a more secure and refined method for generating structured attribution reports. Specifically, step S4723942 includes: S47239421. Based on the cumulative and synergistic effects of the identification, prioritize the key user characteristics, the product attributes, and the pricing module decisions to obtain the priority ranking result; S47239422. Based on the priority ranking results, generate a structured attribution report. The structured attribution report includes a description of the price discrimination pattern, the top N key user characteristics in terms of contribution and their impact on price differences, the top N key product attributes in terms of contribution and their impact on price differences, the top N key pricing module decisions in terms of contribution and their impact on price differences, and targeted correction suggestions. S47239423, Identify and classify sensitive information in the content of the structured attribution report; S47239424. Based on the preset user role permission matrix, sensitive data in the structured attribution report is anonymized or its display is restricted; the anonymization process adopts a dynamic anonymization method, specifically an AI-based adaptive anonymization method that intelligently selects anonymization strategies. S47239425. Generate a customized structured attribution report view for each user role; S47239426. Ensure the security of the structured attribution report content during transmission and storage by using encrypted transmission and access log auditing.

[0066] Specifically, the identification and classification of sensitive information in the structured attribution report refers to the system using Natural Language Processing (NLP) technology, machine learning models, or preset rules to automatically scan the text and data fields in the attribution report and identify personally identifiable information (PII), trade secrets, legally sensitive information, etc. For example, a user's name, ID number, mobile phone number, transaction amount, product cost, and profit margin can all be identified as sensitive information. Classification refers to categorizing the identified sensitive information according to its sensitivity level and type in order to take different protection measures subsequently.

[0067] The process of anonymizing or restricting the display of sensitive data in the structured attribution report based on a preset user role and permission matrix refers to the system maintaining a table mapping user roles to data access permissions. When different user roles (e.g., legal compliance personnel, business operations personnel, senior management personnel) request to view the attribution report, the system will process the sensitive data in the report accordingly based on their role permissions. Anonymization can employ various techniques, such as data masking, data generalization, data encryption, or data perturbation. As a preferred implementation, the anonymization process uses a dynamic anonymization method, specifically an AI-based adaptive anonymization method. This means that the system can intelligently select the most suitable anonymization strategy based on data context, user role, access situation, and data sensitivity to achieve the best balance between data availability and privacy protection. For example, for legal compliance personnel, only the user ID may be hashed; while for business operations personnel, a higher level of generalization may be applied to both the user ID and geographic location information.

[0068] Furthermore, generating customized structured attribution report views for each user role means that the system automatically generates or configures different report display interfaces based on the responsibilities and concerns of different user roles. For example, the legal compliance department may be more concerned with the legal basis, compliance indicators, and potential legal risks of price discrimination; the business operations department may be more concerned with the impact of price adjustments on sales, profits, and market share; while senior management may need macro-level compliance risk trends, an overview of overall business benefits, and the return on investment of key corrective strategies. Customized views ensure the accuracy and efficiency of information delivery.

[0069] Furthermore, ensuring the security of the structured attribution report content during transmission and storage through encrypted transmission and access log auditing means encrypting the data using industry-standard encryption technologies (e.g., Transport Layer Security (TLS) and Advanced Encryption Standard (AES)) when the report is transmitted from the generation end to the user end or stored in the database, preventing the data from being stolen or tampered with during transmission or storage. Access log auditing means that the system records all access behaviors to the attribution report in detail, including the visitor's identity, access time, accessed report content, operation type, etc., for security auditing, accountability, and abnormal behavior detection.

[0070] In some of the embodiments described above in this application, although an interactive visualization interface is proposed to be generated based on the structured attribution report and to support customized display of report content according to user roles, if the specific information needs of different user roles are not carefully distinguished and configured, the visualization interface may fail to efficiently meet the specific decision-making needs of various departments in practical applications. For example, the legal compliance department may focus more on compliance indicators and legal risks, while the business operations department focuses on business benefits and market performance, and senior management needs macro trends and return on investment. If all users see the same or overly generalized view, it will reduce the efficiency of information acquisition and the accuracy of decision-making. In this regard, this application further proposes an optimization scheme, which aims to generate a customized interactive visualization interface through a refined mapping of user roles and data views, thereby improving the relevance and effectiveness of information delivery. Specifically, step S4723943 includes: S47239431. Configure the mapping relationship between user roles and data views; S47239432. Based on the mapping relationship, a customized view is generated. The customized view is designed for the legal compliance department and displays price difference compliance indicators, legal risk thresholds, and legal citations. S47239433. Based on the mapping relationship, generate a customized view. The customized view is for the business operations department and shows the impact of business efficiency indicators, market share changes and correction suggestions on sales and profits. S47239434. Based on the mapping relationship, a customized view is generated. The customized view is designed for senior management personnel and displays the macro compliance risk trends, an overview of overall business benefits, and the input-output ratio of key corrective strategies. S47239435, supports custom chart types, data dimensions, and filtering conditions; S47239436, Save custom configuration.

[0071] The "configuring the mapping relationship between user roles and data views" refers to establishing an association rule that clarifies which specific data metrics, chart types, and report content different types of users (such as legal compliance personnel, business operations personnel, and senior management personnel) should be authorized to view when accessing the interactive visualization interface. This mapping relationship can be predefined or dynamically configured and adjusted through the management interface.

[0072] "Generate customized views" refers to dynamically building and presenting a unique data display interface for each specific user role based on the above mapping relationship. Specifically: The core of a customized view for legal compliance departments lies in highlighting indicators related to laws, regulations, and compliance. This includes, but is not limited to, price difference ratios among different user groups or product categories, risk thresholds related to legal provisions (such as antitrust laws, consumer protection laws, etc.), and directly cited relevant legal provisions, enabling legal professionals to quickly assess potential legal risks and compliance status.

[0073] The customized view for business operations focuses on business performance and market strategy. This view displays business performance metrics such as merchandise sales growth rate, market share changes, and inventory clearance rate, and provides a detailed analysis of the specific impact of different corrective measures (such as feature weight adjustment instructions or loss function penalty term instructions) on sales and profits, helping operations personnel optimize pricing strategies to achieve business goals. The customized view for senior management is designed to provide macro-level, strategic information. This view provides an overview of overall compliance risk trends, the platform's comprehensive business benefits, and the return on investment of key corrective measures (such as adjusting pricing model parameters), supporting senior management in strategic decision-making and resource allocation.

[0074] "Support for custom chart types, data dimensions, and filtering conditions" means that, based on the customized view, users can further flexibly choose different chart types (such as bar charts, line charts, pie charts, etc.), adjust the dimensions of data display (such as by time, by region, by user group, etc.), and set specific filtering conditions (such as specific products, specific time periods) to conduct more in-depth exploratory analysis according to their specific analytical needs.

[0075] "Save custom configurations" means that after a user makes custom settings, the system can save these personalized settings so that the user can directly load them the next time they visit the site, without having to repeat the settings, thereby improving efficiency.

[0076] This application also discloses a judicial big data management system for price discrimination analysis. The system includes: a data acquisition module for acquiring transaction data streams from an e-commerce platform; a rule construction module for constructing a compliance correction rule set, which includes price discrimination judgment criteria and correction actions; a risk detection module for detecting potential price discrimination risks based on the e-commerce platform transaction data streams and the compliance correction rule set; a suggestion generation module for generating pricing parameter adjustment suggestions based on the potential price discrimination risks and the compliance correction rule set; and a strategy correction module for sending the pricing parameter adjustment suggestions to a pricing system to correct the pricing strategy, and receiving the adjustment results from the pricing system after executing the correction, thus forming a compliance assessment and response closed loop.

[0077] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0078] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1.A judicial big data management method for price discrimination analysis, characterized in that, Comprising the following steps: S1, acquiring e-commerce platform transaction data stream; S2, constructing compliance correction rule set, the compliance correction rule set contains price discrimination judgment standard and correction action; S3, based on the e-commerce platform transaction data stream and the compliance correction rule set, detecting potential price discrimination risk; S4, generating pricing parameter adjustment suggestion according to the potential price discrimination risk and the compliance correction rule set; S5, sending the pricing parameter adjustment suggestion to the pricing system to correct the pricing strategy, and receiving the adjustment result of the pricing system after the correction, to form a compliance evaluation and response closed loop. 2.The judicial big data management method for price discrimination analysis according to claim 1, wherein, Step S4 includes: S41, performing semantic analysis on legal texts, identifying legal concepts, situational elements and key values; S42, defining situational rule templates according to the identified legal concepts, situational elements and key values, mapping legal principles with observable input features, internal variables or decision logic of specific modules of the pricing model; S43, calculating the contribution of each input feature to the final price prediction, and revealing the interaction between features; S44, according to the calculated contribution and the revealed feature interaction, extracting the influence link leading to price difference, and quantifying the contribution of each pricing module to the price difference; S45, according to the quantified contribution, generating feature weight adjustment instructions or loss function penalty item instructions; store the generated feature weight adjustment instructions or loss function penalty item instructions as part of the compliance correction rule set; S46, according to the change of the compliance correction rule set, trigger the retraining of the pricing model; S47, monitor the compliance performance of the pricing model, and optimize the compliance correction rule set according to the monitoring result. 3.The judicial big data management method for price discrimination analysis according to claim 2, wherein, Step S47 includes: S4711, define a multi-objective optimization function, which considers compliance indicators and business benefit indicators at the same time; S4712, set compliance hard constraints, which convert prohibited price discrimination behaviors in laws and regulations into constraint conditions of the multi-objective optimization function; S4713, dynamically adjust the rule set weight, which is adjusted according to the legal supervision environment, market competition situation and platform strategic target; S4714, explore the rule set parameter space, which contains price difference threshold, discount adjustment amplitude, feature weight adjustment coefficient or loss function penalty item λ value; S4715, use improved multi-objective optimization algorithm NSGA-II to explore the rule set parameter space under the compliance hard constraint, and find the Pareto optimal rule set configuration; S4716, display the Pareto optimal rule set configuration and its corresponding compliance indicators and business benefit indicators; S4717, select an optimal configuration and deploy it to the compliance correction rule set; S4718, continuously monitor the compliance performance and business benefit of the pricing model in operation after applying the new rule set. 4.The judicial big data management method for price discrimination analysis according to claim 2, wherein, Step S47 includes: S4721, continuously receive e-commerce platform transaction data, user behavior data and legal supervision dynamics; S4722, identify user behavior patterns, commodity market dynamics, and legal regulatory requirement changes based on the transaction data, user behavior data, and legal regulatory dynamics; S4723, generate or adjust compliance monitoring indicators and business benefit monitoring indicators based on the identified user behavior patterns, commodity market dynamics, and legal regulatory requirement changes; S4724, adjust data collection frequency, analysis granularity, and risk warning thresholds based on the generated compliance monitoring indicators and business benefit monitoring indicators; S4725, identify new price discrimination patterns based on the adjusted data collection frequency, analysis granularity, and risk warning thresholds; S4726, trace the impact link and attribute the key user features, commodity attributes, and pricing module decisions that lead to new patterns based on the identified new price discrimination patterns; S4727, generate new compliance correction rules or update existing rules based on the traced impact link and attribution results; S4728, adjust the weights of compliance and business benefit indicators in the multi-objective optimization function based on the generated new compliance correction rules or updated existing rules. 5.The judicial big data management method for price discrimination analysis according to claim 4, wherein, Step S4723 includes: S47231, continuously receive e-commerce platform transaction data, user behavior data, and legal regulatory dynamics; identify user behavior patterns, commodity market dynamics, and legal regulatory requirement changes based on the transaction data, user behavior data, and legal regulatory dynamics; S47232, dynamically build or adjust compliance monitoring indicators and business benefit monitoring indicators based on the identified user behavior patterns, commodity market dynamics, and legal regulatory requirement changes, wherein the dynamic building or adjustment includes: S47233, generate compliance monitoring indicators for sub-user groups based on the identified user behavior patterns, including the difference ratio of specific commodity average payment price and the control group; S47234, generate business benefit monitoring indicators for sub-commodity categories based on the identified commodity market dynamics, including commodity sales growth rate and inventory digestion rate, and related to compliance constraints; S47235, adjust data collection frequency, analysis granularity, and risk warning thresholds based on the generated compliance monitoring indicators and business benefit monitoring indicators; S47236, identify new price discrimination patterns based on the adjusted data collection frequency, analysis granularity, and risk warning thresholds; S47237, trace the impact link and attribute the user features, commodity attributes, and pricing module decisions that lead to new patterns based on the identified new price discrimination patterns; S47238, generate new compliance correction rules or update existing rules based on the traced impact link and attribution results; S47239, adjust the weights of compliance and business benefit indicators in the multi-objective optimization function based on the generated new compliance correction rules or updated existing rules. 6.The judicial big data management method for price discrimination analysis according to claim 5, wherein, Step S47239 includes: S472391, perform correlation analysis on transaction data, user behavior data, and pricing module decision logs to build a dynamic causal network; the dynamic causal network is built using an improved information theory method; S472392, performing influence path search in the dynamic causal network to identify potential influence links; quantifying the contribution of each of the influence links to the price difference; S472393, identifying the cumulative effect and synergistic effect leading to the new price discrimination pattern according to the quantified contribution; S472394, generating an attribution report containing key user features, product attributes, and pricing module decisions according to the identified cumulative effect and synergistic effect. 7.The judicial big data management method for price discrimination analysis according to claim 6, wherein, Step S472394 includes: S4723941, prioritizing the key user features, product attributes, and pricing module decisions according to the identified cumulative effect and synergistic effect to obtain a priority ranking result; S4723942, generating a structured attribution report according to the priority ranking result, the structured attribution report containing a description of the price discrimination pattern, the top N key user features and their impact on the price difference, the top N key product attributes and their impact on the price difference, the top N key pricing module decisions and their impact on the price difference, and targeted correction suggestions; S4723943, generating an interactive visualization interface according to the structured attribution report, the interactive visualization interface allowing decision-makers to explore detailed information of the structured attribution report through filtering and drilling operations, and supporting customized display of the content of the structured attribution report according to user roles. 8.The judicial big data management method for price discrimination analysis according to claim 7, wherein, Step S4723942 includes: S47239421, prioritizing the key user features, product attributes, and pricing module decisions according to the identified cumulative effect and synergistic effect to obtain a priority ranking result; S47239422, generating a structured attribution report according to the priority ranking result, the structured attribution report containing a description of the price discrimination pattern, the top N key user features and their impact on the price difference, the top N key product attributes and their impact on the price difference, the top N key pricing module decisions and their impact on the price difference, and targeted correction suggestions; S47239423, identifying and classifying sensitive information in the content of the structured attribution report; S47239424, desensitizing or limiting the display of sensitive data in the structured attribution report according to a pre-set user role permission matrix; the desensitization adopts a dynamic desensitization method, specifically an AI-based adaptive desensitization method that intelligently selects desensitization strategies; S47239425, generating a customized structured attribution report view for each user role; S47239426, ensuring the security of the content of the structured attribution report during transmission and storage through encrypted transmission and access log auditing. 9.The judicial big data management method for price discrimination analysis according to claim 7, wherein, Step S4723943 includes: S47239431, configuring a mapping relationship between user roles and data views; S47239432, generating a customized view according to the mapping relationship, the customized view showing a price difference compliance indicator, a legal risk threshold and a legal provision quotation for a legal compliance department; S47239433, generating a customized view according to the mapping relationship, the customized view showing a business benefit indicator, a market share change and an influence of a correction suggestion on sales and profits for a business operation department; S47239434, generating a customized view according to the mapping relationship, the customized view showing a macro compliance risk trend, an overall business benefit overview and an input-output ratio of a key correction strategy for a senior management; S47239435, supporting customizing a chart type, a data dimension and a filtering condition; S47239436, saving a customizing configuration. 10.A judicial big data management system for price discrimination analysis, characterized in that, The system comprises: a data acquisition module configured to acquire e-commerce platform transaction data streams; a rule construction module configured to construct a compliance correction rule set, the compliance correction rule set including price discrimination judgment criteria and correction actions; a risk detection module configured to detect potential price discrimination risks based on the e-commerce platform transaction data streams and the compliance correction rule set; a suggestion generation module configured to generate pricing parameter adjustment suggestions according to the potential price discrimination risks and the compliance correction rule set; a strategy correction module configured to send the pricing parameter adjustment suggestions to a pricing system to correct a pricing strategy, and receive adjustment results of the corrected pricing strategy executed by the pricing system to form a compliance evaluation and response closed loop.

Images