Big data transaction security control method and system

Abstract

The invention relates to the technical field of financial information management, and provides a big data transaction security control method and system. Obtaining user information and transaction data flow of two transaction parties, generating a dynamic quantum key from the polarization base sequence selected by the two transaction parties, encrypting the transaction data flow through the dynamic quantum key, and generating an encrypted ciphertext; performing dynamic desensitization on the user information and the transaction data flow based on an entity relationship network constructed by the transaction data flow and role information in the user information to obtain desensitization data; based on the encrypted ciphertext and the desensitized data, training through federal learning to generate a transaction risk identification model; and based on the obtained real-time transaction characteristics, determining a risk probability through the risk identification model, and generating a security management and control strategy according to the risk probability. The data security and privacy are protected by using the quantum encryption technology, and the accurate security control strategy is generated based on the real-time transaction characteristics, so that the security, privacy and reliability of the transaction are significantly improved.

Description

Technical Field

[0001] The present application relates to the field of financial information management technology, and more specifically, to a method and system for the security management and control of big data transactions. Background Art

[0002] With the acceleration of digitalization, data has become a core production factor, particularly in sectors like finance, healthcare, and e-commerce, where the scale of data transactions is rapidly growing. However, the high value of data also poses a variety of security threats. Large-scale data transactions involve sensitive information such as user identities, transaction records, and health data. Cross-institutional and cross-platform data sharing requires safeguards for data privacy. While financial institutions are improving their market competitiveness, they also face increasingly difficult-to-prevent transaction risks. To address these risks, financial institutions need to establish effective transaction security management and control systems.

[0003] In practical applications, data collected by big data systems is often stored in the cloud, which can pose potential security threats. If stored data is not encrypted and proper data security measures are not in place, data leaks are likely to occur. Consequently, transaction security management and control procedures can be challenging, both in terms of data security and accuracy. Summary of the Invention

[0004] The present application provides a method and system for big data transaction security management and control, which can at least to some extent solve the problem of low data security and low control accuracy during transaction security management and control.

[0005] Other features and advantages of the present application will become apparent from the following detailed description, or may be learned in part by practice of the present application.

[0006] According to one aspect of the present application, a method for security management and control of big data transactions is provided, including: obtaining user information and transaction data streams of both parties to the transaction, the transaction data streams including structured and unstructured transaction data; generating a dynamic quantum key using a polarization basis sequence selected by the two parties to the transaction, encrypting the transaction data stream using the dynamic quantum key to generate encrypted ciphertext; dynamically desensitizing the user information and transaction data stream based on an entity relationship network constructed from the transaction data stream and role information in the user information to obtain desensitized data; training based on the encrypted ciphertext and the desensitized data through federated learning to generate a transaction risk identification model; determining a risk probability using the risk identification model based on acquired real-time transaction features, and generating a security management and control strategy based on the risk probability.

[0007] In this application, based on the aforementioned solution, obtaining the user information and transaction data stream of both parties to the transaction includes: obtaining the user information and real-time transaction requests of both parties to the transaction from a mobile application or POS terminal; and obtaining the transaction data stream from a payment platform. In this application, based on the aforementioned solution, generating a dynamic quantum key from the polarization basis sequence selected by the two parties to the transaction, and encrypting the transaction data stream using the dynamic quantum key to generate encrypted ciphertext includes: generating a dynamic quantum key from the polarization basis sequence selected by the two parties to the transaction through quantum polarization state transmission; hashing the dynamic quantum key to generate a hash parameter; and encrypting the transaction data stream using the hash parameter to generate encrypted ciphertext.

[0008] In the present application, based on the aforementioned scheme, the entity relationship network constructed based on the transaction data stream and the role information in the user information, dynamically desensitizes the user information and the transaction data stream to obtain desensitized data, including: parsing data semantics based on the transaction data stream to construct an entity relationship network involved in the transaction; identifying the data sensitivity level based on the entity relationship network and the context information of the transaction data stream; generating a desensitization intensity coefficient based on the role information in the user information and the data sensitivity level; and desensitizing the transaction data stream based on the comparison result between the desensitization intensity coefficient and the set threshold to generate desensitized data.

[0009] In the present application, based on the aforementioned scheme, the encrypted ciphertext and the desensitized data are trained through federated learning to generate a transaction risk identification model, including: decrypting the encrypted ciphertext to generate decrypted data; inputting the decrypted data and the desensitized data into a trusted execution environment, training a local model based on a loss function, and generating local model weights; submitting the hash value of the local model weight to the blockchain in the trusted execution environment, calling a verification function through the smart contract on the blockchain to verify the security of the local model weight and obtain a first weight that has passed the verification; aggregating the first weight to generate an aggregated weight, and constructing a global risk identification model based on the aggregated weight.

[0010] In the present application, based on the aforementioned scheme, the risk probability is determined based on the acquired real-time transaction features through the risk identification model, and a security management strategy is generated according to the risk probability, including: based on the acquired real-time transaction features, determining the data type corresponding to the real-time transaction features; sending the real-time transaction features to the corresponding model according to the data type, and obtaining the output result of the corresponding model; processing the output result through the dynamic attention mechanism to generate the risk probability; and generating a security management strategy according to the risk probability.

[0011] In the present application, based on the aforementioned scheme, the generation of a security control strategy according to the risk probability includes: obtaining an anomaly score of the transaction user's historical behavior; generating a comprehensive risk index based on the risk probability, the transaction amount in the transaction data stream, and the anomaly score, and determining a dynamic risk level; and selecting a security control strategy from a preset policy library based on the dynamic risk level.

[0012] According to one aspect of the present application, a big data transaction security management and control system is provided, comprising:

[0013] An acquisition unit, configured to acquire user information of both parties to a transaction and a transaction data stream, wherein the transaction data stream includes structured and unstructured transaction data;

[0014] an encryption unit, configured to generate a dynamic quantum key using the polarization base sequence selected by the transaction parties, and encrypt the transaction data stream using the dynamic quantum key to generate an encrypted ciphertext;

[0015] a desensitizing unit, configured to dynamically desensitize the user information and the transaction data stream based on the entity relationship network constructed by the transaction data stream and the role information in the user information, thereby obtaining desensitized data;

[0016] A training unit, configured to generate a transaction risk identification model by training through federated learning based on the encrypted ciphertext and the desensitized data;

[0017] The management and control unit is used to determine the risk probability through the risk identification model based on the acquired real-time transaction characteristics, and generate a security management and control strategy according to the risk probability.

[0018] According to one aspect of the present application, a computer-readable medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method for security management and control of big data transactions as described in the above embodiments is implemented.

[0019] According to one aspect of the present application, an electronic device is provided, comprising: one or more processors; a storage device for storing one or more programs, which, when executed by the one or more processors, enables the one or more processors to implement the big data transaction security management and control method as described in the above embodiments.

[0020] According to one aspect of the present application, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to implement the big data transaction security management and control methods provided in the various optional implementations described above.

[0021] In the technical solution of the present application, the user information and transaction data stream of the two parties to the transaction are obtained, a dynamic quantum key is generated by using the polarization base sequence selected by the two parties to the transaction, the transaction data stream is encrypted by the dynamic quantum key to generate an encrypted ciphertext; based on the entity relationship network constructed by the transaction data stream and the role information in the user information, the user information and transaction data stream are dynamically desensitized to obtain desensitized data; based on the encrypted ciphertext and the desensitized data, a transaction risk identification model is generated by training through federated learning; based on the acquired real-time transaction features, the risk probability is determined by the risk identification model, and a security management strategy is generated according to the risk probability. By utilizing quantum encryption technology to protect data security and privacy, and generating accurate security management strategies based on real-time transaction features, comprehensive, real-time, and accurate security management of transaction behaviors is achieved, which significantly improves the security, privacy, and reliability of transactions, and provides financial institutions with effective risk prevention measures.

[0022] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The accompanying drawings are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and those skilled in the art can derive other drawings based on these drawings without inventive effort.

[0024] Figure 1 The following schematically illustrates a flow chart of a method for secure management and control of big data transactions in one embodiment of the present application.

[0025] Figure 2 The flowchart of generating encrypted ciphertext in one embodiment of the present application is schematically shown.

[0026] Figure 3 The following schematically shows a schematic diagram of a big data transaction security management and control system in one embodiment of the present application.

[0027] Figure 4 A schematic diagram of the structure of a computer system suitable for implementing an electronic device according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0028] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this application will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art.

[0029] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner.In the following description, many specific details are provided so as to provide a full understanding of the embodiments of the present application. However, it will be appreciated by those skilled in the art that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps etc. can be adopted. In other cases, known methods, devices, implementations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0030] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically separate entities. That is, these functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0031] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, while others may be combined or partially combined. Therefore, the actual execution order may vary depending on the actual situation.

[0032] The implementation details of the technical solution of this application are described in detail below:

[0033] Figure 1 The flowchart of the method for controlling the security of big data transactions according to an embodiment of the present application is shown. Figure 1 As shown, the big data transaction security management method includes at least steps S110 to S150, which are described in detail as follows:

[0034] In step S110 , user information of both parties to the transaction and a transaction data stream are obtained, where the transaction data stream includes structured and unstructured transaction data.

[0035] In one embodiment of the present application, obtaining user information and transaction data streams of both parties to the transaction includes: obtaining user information and real-time transaction requests from a mobile application or a point of sales (POS) terminal; and obtaining transaction data streams from a payment platform.

[0036] In one embodiment of the present application, real-time transaction requests, including transaction amount, time, and geographic location, are obtained from mobile applications, web pages, or POS terminals. Transaction status is captured from connected banks or third-party payment platforms. Furthermore, historical user operations within the platform, such as login frequency, browsing paths, and password modification history, are recorded to build a user behavior profile.

[0037] Optionally, real-time transaction data is transmitted via a message queue to ensure millisecond-level response. Historical behavior data is regularly imported from the database in batches, and user profiles are updated simultaneously.

[0038] Optionally, user information and transaction data streams are obtained according to the authorization scope of the transaction user.

[0039] Optionally, the acquired data is cleaned and normalized to facilitate subsequent analysis. Obviously erroneous data, such as negative transaction amounts and incorrect timestamp formats, is automatically identified and removed. Standardized fields, such as unified currency units and time zone conversions, are stored in a distributed database.

[0040] Optionally, while acquiring user information and transaction data streams from both parties, a real-time data pipeline collects structured data, such as transaction amount, timestamp, user ID, and payment status. This data is then integrated with payment gateways, databases, and third-party transaction interfaces, using a stream processing engine to capture transaction events and ensure millisecond-level data transmission. This structured data is then optionally fed into a risk control engine, providing high-precision input for real-time risk scoring models. For example, by correlating historical user transaction records, anomalous spending patterns can be quickly identified.

[0041] A multimodal processing framework enables in-depth analysis of unstructured data, such as contract texts, customer service conversation recordings, and operation logs within transaction data streams. Specifically, natural language processing is used to extract key entities from text data, such as transaction terms and risk keywords within contracts. Speech recognition is used to convert speech data into text for semantic and sentiment analysis. Operation logs are then analyzed using optical character recognition (OCR) and visual modeling to identify screenshots of abnormal operations.

[0042] The acquired data is cleaned and stored in object storage. It is then converted into structured feature vectors through the feature extraction layer and fused with the original transaction data. For example, an "urgent transfer request" mentioned in a customer service conversation can be matched with the same user's transaction history to enhance contextual understanding for risk assessment. Ultimately, the multimodal data stream provides a rich input dimension for the hybrid model, supporting the accurate identification of complex risk scenarios.

[0043] This process captures user information and transaction data streams from multiple channels, including mobile apps, POS terminals, and payment platforms. This ensures comprehensiveness and diversity, providing a rich foundation for subsequent security management. By capturing transaction requests and data streams in real time, the security management system can rapidly respond to transactions and promptly identify and address potential risks.

[0044] In step S120, a dynamic quantum key is generated using the polarization basis sequence selected by the transaction parties, and the transaction data stream is encrypted using the dynamic quantum key to generate an encrypted ciphertext.

[0045] In one embodiment of the present application, based on the extracted identity information of the users of both parties to the transaction, such as user ID, device fingerprint and biometrics, this information will be used as the seed for generating the quantum key. Using the principles of quantum mechanics, through quantum polarization state transmission or quantum entanglement, a dynamic quantum key that is closely related to the user information of both parties to the transaction and is unique for each communication is generated. The key is then applied to the encryption process of the transaction data stream, and the original transaction data is converted into ciphertext form through an encryption algorithm, thereby ensuring the confidentiality, integrity and authenticity of the transaction data during transmission and storage, and effectively preventing the risk of data leakage and tampering.

[0046] like Figure 2 As shown, in one embodiment of the present application, a dynamic quantum key is generated using the polarization base sequence selected by the transaction parties, and the transaction data stream is encrypted using the dynamic quantum key to generate an encrypted ciphertext, including:

[0047] S210, generating a dynamic quantum key by transmitting the polarization base sequence selected by the transaction parties through quantum polarization state transmission;

[0048] S220, performing hash processing on the dynamic quantum key to generate a hash parameter;

[0049] S230: Encrypt the transaction data stream using the hash parameter to generate an encrypted ciphertext.

[0050] In one embodiment of the present application, the two parties to the transaction select a polarization base sequence, and based on the quantum key distribution method of the quantum key distribution protocol, the polarization base sequence is transmitted through the polarization state of the photon to generate an initial key. At the same time, a quantum random number generator is used to generate a true random number using quantum physical phenomena (such as the time of photon emission), and the timestamp t is bound to enhance the dynamicity to generate a dynamic quantum key K. Q for:

[0051]

[0052] Among them, S A ,S BThey represent the polarization base sequences selected by the two parties to the transaction, QKD(S A ,S B ) represents the quantum key distribution process based on the quantum key distribution protocol (such as the BB84 protocol), which generates the initial key through the transmission of photon polarization states; QRNG(t) represents the quantum random number generator, which uses quantum physical phenomena (such as the photon emission time) to generate true random numbers and binds the timestamp t to enhance the dynamics. Represents the exclusive OR operation.

[0053] In practical applications, traditional BB84 protocol keys are statically generated and vulnerable to replay attacks. This solution introduces real-time perturbations and time-varying factors in the quantum random number generator, combining the initial key generated by the quantum key distribution protocol with the random number generated by the quantum random number generator. This breaks the periodicity of the key, avoids the risk of reuse, and enables dynamic key updates, improving resistance to quantum computing attacks.

[0054] Afterwards, the dynamic quantum key is hashed to generate a hash parameter, and the transaction data stream is encrypted using the hash parameter to generate an encrypted ciphertext c i for:

[0055]

[0056] Among them, f(K Q ) represents the hash parameter generated by hashing the key, d i Represents transaction data flow, e i represents the error parameter that follows the Gaussian distribution, Indicates rounding operation, and mod indicates remainder operation.

[0057] The above process maps the data to a high-dimensional space and performs hash operations through dynamic quantum keys, so that the hash parameters change with the dynamic quantum keys, so that each transaction has a unique encryption key, breaking the fixed parameter attack surface, with extremely high security and unbreakable properties, effectively preventing data from being stolen or tampered with during transmission, and improving data security.

[0058] In step S130, based on the entity relationship network constructed by the transaction data stream and the role information in the user information, the user information and the transaction data stream are dynamically desensitized to obtain desensitized data.

[0059] In this embodiment, the transaction data stream is parsed, and the entities involved in the transaction and the relationships between them are identified and extracted, thereby constructing an entity relationship network. The entity relationship network reveals the relationship between the transaction parties and also reflects the flow path of data in the transaction process. At the same time, the role information in the user information, such as the user's identity, authority level, and role positioning in the transaction, is combined to comprehensively judge the data sensitivity and desensitization needs. Based on these analyses, desensitization intensity coefficients are dynamically generated, and these desensitization intensity coefficients are used to accurately desensitize sensitive information in the transaction data stream, such as replacing, masking or generalizing sensitive data, to ensure that while protecting user privacy and data security, the data availability and business logic consistency are maintained as much as possible, and ultimately desensitized data that is both secure and compliant is obtained.

[0060] In one embodiment of the present application, based on the entity relationship network constructed by the transaction data stream and the role information in the user information, the user information and the transaction data stream are dynamically desensitized to obtain desensitized data, including:

[0061] Parsing data semantics based on the transaction data stream to construct a relationship network of entities involved in the transaction;

[0062] identifying a data sensitivity level based on the entity relationship network and contextual information of the transaction data flow;

[0063] generating a desensitization intensity coefficient according to the role information in the user information and the data sensitivity level;

[0064] According to the comparison result between the desensitization intensity coefficient and the set threshold, the transaction data stream is desensitized to generate desensitized data.

[0065] In one embodiment of the present application, the transaction data stream contains a large amount of information about the transaction, such as the identities of the two parties to the transaction, the transaction amount, the transaction time, and the transaction items. Natural language processing technology and machine learning algorithms are used to deeply analyze this data to identify and understand the semantic information in the data. Through semantic analysis, it is possible to identify the various entities in the transaction, such as buyers, sellers, and goods, and further analyze the relationships between these entities, such as purchase relationships, payment relationships, etc. Ultimately, based on these relationships, an entity relationship network is constructed. This network can clearly display the various entities in the transaction and their relationships, providing a basis for subsequent data processing and analysis.

[0066] Second, data sensitivity levels are identified based on contextual information from the entity relationship network and transaction data streams. Data sensitivity is assessed by analyzing individual nodes and edges in the entity relationship network, as well as contextual information from the transaction data stream, such as transaction time, location, and frequency. For example, if a transaction involves a large transfer of funds or sensitive personal information, the data sensitivity level will be relatively high. Based on pre-set sensitivity assessment rules, this information is comprehensively evaluated to determine the sensitivity level of each transaction data.

[0067] Next, a desensitization intensity coefficient is generated based on the user's role information and the data sensitivity level. First, the user's role information is obtained, such as whether the user is an administrator, regular user, or guest. Different roles have different data access permissions and desensitization requirements. Then, combined with the previously determined data sensitivity level, a desensitization intensity coefficient is generated. This coefficient reflects the degree of data desensitization required to ensure data security and privacy.

[0068] Finally, based on the comparison between the desensitization intensity coefficient and the set threshold, the transaction data stream is desensitized to generate desensitized data. The desensitization intensity coefficient is compared with the preset threshold. If the desensitization intensity coefficient is higher than the threshold, the transaction data stream undergoes a higher degree of desensitization, such as replacing sensitive information or encrypting the data. If the desensitization intensity coefficient is lower than or equal to the threshold, a lower degree of desensitization is performed or no desensitization is performed. Ultimately, desensitized data is generated, which retains a certain level of information value while ensuring data security and privacy.

[0069] This process improves the accuracy of sensitive field identification through semantic understanding from knowledge graphs and natural language processing, combined with dynamic user role adjustment strategies. By building an entity relationship network and identifying data sensitivity levels, and generating a desensitization intensity coefficient based on user roles and data sensitivity levels, sensitive data can be accurately desensitized, making desensitization more flexible and personalized, protecting privacy while ensuring data availability.

[0070] In step S140, a transaction risk identification model is generated by training through federated learning based on the encrypted ciphertext and the desensitized data.

[0071] In one embodiment of the present application, encrypted ciphertext is securely decrypted to ensure that the original transaction data features are restored in a trusted execution environment, while the desensitized data is directly used to provide transaction behavior patterns in a non-sensitive context. These two types of data can both protect user privacy and retain sufficient information for model learning. Subsequently, within the framework of federated learning, multiple participants (such as different financial institutions or data holders) independently train local models within their respective data domains based on the same model architecture and loss function to generate local model weights. These local model weights are subject to strict privacy protection measures, such as differential privacy or secure multi-party computation, before being uploaded to a central server or a trusted third party such as a blockchain to prevent the leakage of sensitive information. The central server or blockchain is responsible for aggregating these local model weights to form a global aggregate weight, and based on these weights, constructing the final transaction risk identification model. The transaction risk identification model can learn the general laws and potential risk patterns of transaction behavior across different data sources, thereby achieving accurate identification and early warning of transaction risks.

[0072] In one embodiment of the present application, based on the encrypted ciphertext and the desensitized data, a transaction risk identification model is generated through training through federated learning, including:

[0073] Decrypting the encrypted ciphertext to generate decrypted data;

[0074] Inputting the decrypted data and the masked data into a trusted execution environment, training a local model based on a loss function, and generating a local model weight;

[0075] Submitting the hash value of the local model weight to the blockchain in the trusted execution environment, calling a verification function through the smart contract on the blockchain to verify the security of the local model weight, and obtaining a verified first weight;

[0076] The first weights are aggregated to generate aggregated weights, and a global risk identification model is constructed based on the aggregated weights.

[0077] In one embodiment of this application, the core goal of local Trusted Execution Environment (TEE) training is to ensure data security during decryption and model training. First, each participant (such as a bank or payment institution) must locally deploy a hardware-isolated TEE to create a protected secure enclave. This environment uses hardware encryption and memory isolation to ensure that the data and computing processes within the enclave are inaccessible to outsiders.

[0078] When encrypted or desensitized data is input, if the data is in ciphertext, it is decrypted within the enclave using the quantum key. This process occurs entirely within a secure environment, with neither the key nor the plaintext data exposed to the operating system or other applications, effectively preventing side-channel attacks such as information theft through power consumption or electromagnetic leakage. The decrypted data is directly used for local model training, and the model parameter optimization process is also completed within the enclave. During training, the model minimizes the loss function and incorporates regularization terms to prevent overfitting, ultimately generating local model weights.

[0079] In traditional federated learning, data decryption and training may occur in an unprotected environment, posing a risk of data leakage. However, TEE uses hardware-level protection to ensure that sensitive data is processed only within a secure enclave. Even if the host system is compromised, attackers cannot access the information within the enclave, significantly improving privacy and security.

[0080] The goal of blockchain-verified aggregation is to ensure that the global model is composed only of legitimate, untampered local model weights. First, after completing local training, each participant submits a hash of the model weights to the blockchain network. This hash is generated using an irreversible hash function and uniquely identifies the weights. Any modification of the weights will result in a change in the hash value.

[0081] The smart contract invokes pre-set verification logic, such as checking whether the hash value comes from a TEE-authenticated party. Weights that fail verification, such as those from unauthorized devices or tampered models, are rejected directly, and only legitimate weights are retained as the first weight.

[0082] The weights that have passed the verification then enter the aggregation phase. Based on the verification results of each participant, the first weights are aggregated to generate the aggregate weights of round t+1 iterations. for:

[0083]

[0084] Among them, k represents the identification of the participants participating in the model training, K represents the total number of participants, and represents, represents the local model weight of participant k in round t, Hash(·) represents hash operation, and Verify(·) represents verification security processing.

[0085] Unlike traditional federated learning methods, which simply average the weights of all nodes, this approach is vulnerable to malicious nodes submitting forged parameters, such as model poisoning attacks. The blockchain verification mechanism, on the other hand, ensures that only trusted weights are aggregated through hashing on-chain and automated screening with smart contracts. It also leverages the immutability of the blockchain to record the entire process, significantly improving the model's robustness and anti-attack capabilities.

[0086] In the above process, local TEE training protects data decryption and training through hardware-level isolation, addressing the data leakage risks associated with traditional federated learning. By training models in a trusted execution environment, data security and privacy are ensured, preventing data leakage. Federated learning enables joint training using data from multiple participants, improving model accuracy and generalization. Blockchain verification aggregation utilizes hashing on-chain and smart contracts to screen legitimate weights, verifying the security of local model weights and preventing malicious nodes from contaminating the global model, ensuring model reliability and security. This creates a closed loop of secure computing and trusted verification, enabling efficient and reliable distributed model collaboration while protecting privacy.

[0087] In step S150, based on the acquired real-time transaction features, the risk probability is determined by the risk identification model, and a security management and control strategy is generated according to the risk probability.

[0088] In one embodiment of the present application, a pre-trained risk identification model analyzes acquired real-time transaction features to determine the risk probability of each transaction. This process involves inputting real-time transaction features, such as transaction amount, transaction time, transaction location, user behavior patterns, and device information, into the risk identification model. Using machine learning algorithms, such as deep learning or ensemble learning, complex data processing and pattern recognition are performed on the input features to output a quantified risk probability value, which reflects the likelihood that the transaction presents a potential risk. Once the risk probability is determined, corresponding security control policies are automatically generated based on pre-set risk thresholds and decision rules. For low-risk transactions, only routine monitoring or recording is performed; for high-risk transactions, more stringent measures are implemented, such as triggering additional identity verification steps, limiting the transaction amount, suspending the transaction until manual review, or immediately blocking the transaction and notifying the relevant security team for further investigation. The development and implementation of these security control policies are designed to ensure transaction security and compliance while minimizing disruption to the normal transaction process, thereby providing users with a safe and convenient transaction experience.

[0089] In one embodiment of the present application, based on the acquired real-time transaction features, the risk probability is determined by the risk identification model, and a security management and control strategy is generated according to the risk probability, including:

[0090] Determining, based on the acquired real-time transaction features, a data type corresponding to the real-time transaction features;

[0091] Sending the real-time transaction features to a corresponding model according to the data type, and obtaining an output result of the corresponding model;

[0092] Processing the output results through a dynamic attention mechanism to generate a risk probability;

[0093] A security management and control strategy is generated based on the risk probability.

[0094] In one embodiment of the present application, in a hybrid model architecture, transaction risks are comprehensively analyzed by integrating three different machine learning models. First, structured data is processed by the distributed gradient boosting library model XGBoost. The distributed gradient boosting library is good at capturing complex interactions between structured features, such as discovering the possible risks of the combination of "high-value transactions and uncommon IP addresses." Second, unstructured text data is input into the deep learning model. The deep learning Transformer model extracts key information through semantic analysis, such as identifying high-risk clauses implied in contracts or descriptions of abnormal operations in logs. Third, time series data (such as transaction timestamps, user historical behavior) is processed by the Long Short-Term Memory (LSTM) model, which can capture continuous patterns in time series, such as the abnormal rhythm of frequent transactions in a short period of time.

[0095] The outputs of these three models are not simply added together, but intelligently integrated through a dynamic attention mechanism. The weights of each model are automatically adjusted based on the specific characteristics of the current transaction. For example, when it is detected that the transaction amount is greater than the preset first threshold, the weight of XGBoost increases, emphasizing the importance of structural features such as the amount and IP. If the complexity of the contract text involved in the transaction is greater than the preset second threshold, the weight of Transformer is enhanced to focus on the semantic analysis results. If the transaction time interval is less than the preset third threshold and conforms to the historical abnormal pattern, the weight of LSTM is increased. This adjustment based on the dynamic attention mechanism enables each model to flexibly adapt to different scenarios, avoid the limitations of a single modality, and ultimately output a comprehensive risk probability (between 0 and 1), with higher scores representing greater risks.

[0096] Furthermore, spatiotemporal correlation analysis is used to identify coordinated attack patterns across time and space. First, the similarity of features between transactions is calculated, for example, by comparing the similarity of transaction initiating devices, geographic locations, or operational behaviors. If two transactions originate from the same device and have highly consistent behavioral patterns, the similarity is high. Next, the system introduces spatiotemporal correlation parameters to evaluate the time intervals between transactions. For example, if the same network address initiates multiple transactions within 5 minutes, the time interval is small and the correlation is strong; however, transactions initiated several hours apart are considered less correlated.

[0097] Specifically, in order to identify cross-platform and cross-time coordinated attack patterns, the spatiotemporal correlation parameter A between transaction features is calculated. ij for:

[0098]

[0099] Among them, i, j and k represent the identifiers of transaction features, n represents the total number of transaction features, τ represents the hyperparameter controlling the attention distribution, λ represents the time decay factor, and t i , t j Represents the timestamps of transactions i and j. By calculating the spatiotemporal correlation parameters, the model's ability to capture spatiotemporal anomalies is enhanced. For example, if the same IP initiates multiple high-value transactions in a short period of time with a small time interval and high feature similarity, then A ij Large, triggering a risk warning.

[0100] By combining spatiotemporal correlation parameters and time decay, a correlation weight is generated for each pair of transactions. Transactions with high weights are flagged as potential coordinated attacks, such as when multiple accounts execute the same operation within a short period of time in a distributed denial of service attack. This process not only identifies anomalies in individual transactions but also captures cross-platform and cross-time attack chains that traditional methods miss. For example, an attacker may operate across multiple accounts at different times, but spatiotemporal correlation analysis can still identify implicit connections between these accounts in terms of characteristics and time, providing early warning.

[0101] Ultimately, the results of spatiotemporal analysis are combined with the risk probability of the hybrid model to further optimize the overall judgment. This multi-dimensional analysis significantly improves the detection rate of complex attack patterns, while also filtering out accidental correlations through a time decay mechanism, reducing the risk of false positives.

[0102] The above process determines risk probabilities based on real-time transaction characteristics, enabling the security management system to rapidly respond to changes in trading behavior. Processing the output results through a dynamic attention mechanism enables more precise assessment of transaction risks, improving the accuracy of risk identification. Generating security management strategies based on risk probabilities allows for more flexible and personalized strategy formulation, enabling the implementation of tailored control measures for different risk levels.

[0103] In one embodiment of the present application, generating a security management and control strategy based on the risk probability includes:

[0104] Obtain anomaly scores for historical transaction users' behaviors;

[0105] generating a comprehensive risk index based on the risk probability, the transaction amount in the transaction data stream, and the anomaly score, and determining a dynamic risk level;

[0106] According to the dynamic risk level, a security control strategy is selected from a preset strategy library.

[0107] In one embodiment of the present application, a dynamic assessment of transaction risk is first performed by integrating multi-dimensional information. The risk probability is derived from the output of a multimodal model and weighted by factors such as the specific characteristics of the current transaction (e.g., transaction amount), historical user behavior (e.g., anomaly scores of historical transaction user behavior), and real-time system status (e.g., server load) to generate a comprehensive risk index. For example, if a high-value transfer comes from an unfamiliar device and the user has a recent suspicious login history, the comprehensive risk index will be significantly increased.

[0108] After generating the comprehensive risk index, the current threshold is calculated and compared with the comprehensive risk index to determine the dynamic risk level. Specifically, in this embodiment, the risk threshold is dynamically adjusted according to the real-time environment. Specifically, the current threshold θ is determined as follows:

[0109]

[0110] Among them, θ base Indicates the preset basic threshold, L, L max They represent the real-time load and maximum load capacity of the system respectively; λ represents the attenuation coefficient, which decreases with the increase of attack frequency; T attack Indicates the recent attack frequency, such as the number of high-risk transactions in the past hour.

[0111] For example, during peak periods or when a recent increase in attack frequency is detected, the high-risk threshold is automatically lowered to more strictly block suspicious transactions. Conversely, during periods of low load, the threshold is appropriately relaxed to reduce interference with normal users. This dynamic adjustment ensures that the strategy can respond to sudden attacks while avoiding excessive defense.

[0112] Based on the risk level and transaction characteristics, specific control actions are matched from the predefined policy library. The policy library is constructed based on dimensions such as risk level, transaction type, and user credibility. For example: For high-risk cross-border payments, if the transaction amount is huge and comes from a high-risk area, the system will immediately freeze the account, trigger a manual review, and simultaneously notify the risk control team. For medium-risk large-value transfers, if the user's device is trustworthy but the transaction time is abnormal (such as in the early morning), the system will require secondary verification (such as facial recognition) and delay the transaction for 5 minutes to observe subsequent behavior. For low-risk small-value consumption, only logs are recorded and released normally.

[0113] In addition, for edge cases (such as when the risk score approaches a threshold), the system uses a progressive verification mechanism. For example, when a user logs in for the first time from a new device, they will first receive a verification code via SMS. If verification fails, the verification is upgraded to biometric verification, gradually increasing the verification strength to balance security and user experience.

[0114] Optionally, after each control action is executed, the system collects feedback data, such as whether the fraud was successfully intercepted and whether the user complained about the false alarm, and continuously optimizes the strategy based on this information. For example: when the interception is successful, the application weight of the current strategy in similar scenarios is strengthened; when a false alarm occurs, the sensitivity of the relevant risk parameters is reduced to avoid repeated misjudgments; when an underreporting occurs, the model threshold is adjusted or new risk features are added to improve detection sensitivity. Through the automated learning mechanism, the pattern changes in historical data (such as new attack methods) are regularly analyzed, and the strategy library is dynamically updated. For example, if a recent phishing attack against a specific payment channel is detected, the system will temporarily increase the risk weight of such transactions and add verification steps.

[0115] For example, an attacker manipulates multiple accounts to initiate small transfers within a short period of time. Through spatiotemporal correlation analysis, the system finds that these transactions originate from different devices but have concentrated IP addresses, with extremely short time intervals. This leads to a comprehensive assessment of the attack as a coordinated attack. Even if a single transaction has a medium risk score, the system will still block them in batches and trigger a global risk alert.

[0116] This process uses thresholds and policies to adjust in real time to the environment, avoiding false positives or missed alerts caused by a "one-size-fits-all" approach. Differentiated management actions are matched to different transaction types and user profiles, balancing security and efficiency. Through a closed-loop feedback loop, policies are continuously optimized to adapt to new attack methods and changes in user behavior. Translating abstract risk scores into specific, actionable security actions makes policy formulation more dynamic and flexible, adapting to different transaction scenarios and risk levels. This allows for precise risk control in complex and volatile trading environments while maximizing user experience.

[0117] The technical solution of this application obtains the user information and transaction data stream of both parties to the transaction, generates a dynamic quantum key with the polarization base sequence selected by the two parties to the transaction, encrypts the transaction data stream through the dynamic quantum key to generate an encrypted ciphertext; based on the entity relationship network constructed by the transaction data stream and the role information in the user information, dynamically desensitizes the user information and transaction data stream to obtain desensitized data; based on the encrypted ciphertext and the desensitized data, trains through federated learning to generate a transaction risk identification model; based on the acquired real-time transaction features, determines the risk probability through the risk identification model, and generates a security management strategy based on the risk probability. By using quantum encryption technology to protect data security and privacy, and generating accurate security management strategies based on real-time transaction features, comprehensive, real-time, and accurate security management of transaction behaviors is achieved, which significantly improves the security, privacy, and reliability of transactions, and provides financial institutions with an effective risk prevention measure.

[0118] The following describes an embodiment of the device of the present application, which can be used to execute the method for security control of big data transactions in the above-mentioned embodiment of the present application. It is understood that the device can be a computer program (including program code) running on a computer device, for example, the device is an application software; the device can be used to execute the corresponding steps in the method provided in the embodiment of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method for security control of big data transactions in the above-mentioned embodiment of the present application.

[0119] Figure 3 A block diagram of a big data transaction security management and control system according to an embodiment of the present application is shown.

[0120] Reference Figure 3 As shown, a big data transaction security management and control system according to an embodiment of the present application includes:

[0121] An acquisition unit 310 is configured to acquire user information of both parties to a transaction and a transaction data stream, wherein the transaction data stream includes structured and unstructured transaction data;

[0122] An encryption unit 320 is configured to generate a dynamic quantum key using the polarization base sequence selected by the transaction parties, and encrypt the transaction data stream using the dynamic quantum key to generate encrypted ciphertext;

[0123] A desensitizing unit 330 is configured to dynamically desensitize the user information and the transaction data stream based on the entity relationship network constructed by the transaction data stream and the role information in the user information to obtain desensitized data;

[0124] A training unit 340 is configured to generate a transaction risk identification model by performing training through federated learning based on the encrypted ciphertext and the desensitized data;

[0125] The management and control unit 350 is configured to determine the risk probability based on the acquired real-time transaction characteristics through the risk identification model, and generate a security management and control strategy according to the risk probability.

[0126] In this application, based on the aforementioned solution, obtaining the user information and transaction data stream of both parties to the transaction includes: obtaining the user information and real-time transaction requests of both parties to the transaction from a mobile application or POS terminal; and obtaining the transaction data stream from a payment platform. In this application, based on the aforementioned solution, generating a dynamic quantum key from the polarization basis sequence selected by the two parties to the transaction, and encrypting the transaction data stream using the dynamic quantum key to generate encrypted ciphertext includes: generating a dynamic quantum key from the polarization basis sequence selected by the two parties to the transaction through quantum polarization state transmission; hashing the dynamic quantum key to generate a hash parameter; and encrypting the transaction data stream using the hash parameter to generate encrypted ciphertext.

[0127] In the present application, based on the aforementioned scheme, the entity relationship network constructed based on the transaction data stream and the role information in the user information, dynamically desensitizes the user information and the transaction data stream to obtain desensitized data, including: parsing data semantics based on the transaction data stream to construct an entity relationship network involved in the transaction; identifying the data sensitivity level based on the entity relationship network and the context information of the transaction data stream; generating a desensitization intensity coefficient based on the role information in the user information and the data sensitivity level; and desensitizing the transaction data stream based on the comparison result between the desensitization intensity coefficient and the set threshold to generate desensitized data.

[0128] In the present application, based on the aforementioned scheme, the encrypted ciphertext and the desensitized data are trained through federated learning to generate a transaction risk identification model, including: decrypting the encrypted ciphertext to generate decrypted data; inputting the decrypted data and the desensitized data into a trusted execution environment, training a local model based on a loss function, and generating local model weights; submitting the hash value of the local model weight to the blockchain in the trusted execution environment, calling a verification function through the smart contract on the blockchain to verify the security of the local model weight and obtain a first weight that has passed the verification; aggregating the first weight to generate an aggregated weight, and constructing a global risk identification model based on the aggregated weight.

[0129] In the present application, based on the aforementioned scheme, the risk probability is determined based on the acquired real-time transaction features through the risk identification model, and a security management strategy is generated according to the risk probability, including: based on the acquired real-time transaction features, determining the data type corresponding to the real-time transaction features; sending the real-time transaction features to the corresponding model according to the data type, and obtaining the output result of the corresponding model; processing the output result through the dynamic attention mechanism to generate the risk probability; and generating a security management strategy according to the risk probability.

[0130] In the present application, based on the aforementioned scheme, the generation of a security control strategy according to the risk probability includes: obtaining an anomaly score of the transaction user's historical behavior; generating a comprehensive risk index based on the risk probability, the transaction amount in the transaction data stream, and the anomaly score, and determining a dynamic risk level; and selecting a security control strategy from a preset policy library based on the dynamic risk level.

[0131] The technical solution of this application obtains the user information and transaction data stream of both parties to the transaction, generates a dynamic quantum key with the polarization base sequence selected by the two parties to the transaction, encrypts the transaction data stream through the dynamic quantum key to generate an encrypted ciphertext; based on the entity relationship network constructed by the transaction data stream and the role information in the user information, dynamically desensitizes the user information and transaction data stream to obtain desensitized data; based on the encrypted ciphertext and the desensitized data, trains through federated learning to generate a transaction risk identification model; based on the acquired real-time transaction features, determines the risk probability through the risk identification model, and generates a security management strategy based on the risk probability. By using quantum encryption technology to protect data security and privacy, and generating accurate security management strategies based on real-time transaction features, comprehensive, real-time, and accurate security management of transaction behaviors is achieved, which significantly improves the security, privacy, and reliability of transactions, and provides financial institutions with an effective risk prevention measure.

[0132] Figure 4 A schematic diagram of the structure of a computer system suitable for implementing an electronic device according to an embodiment of the present application is shown.

[0133] It should be noted that the computer system of the electronic device in this embodiment is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0134] In this embodiment, the computer system includes a central processing unit (CPU) 401, which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 402 or programs loaded from a storage unit 408 into a random access memory (RAM) 403, such as executing the big data transaction security management and control method described in the above embodiment. The RAM 403 also stores various programs and data required for system operation. The CPU 401, ROM 402, and RAM 403 are interconnected via a bus 404. An input / output interface 405 is also connected to the bus 404.

[0135] The following components are connected to the input / output interface 405: an input section 406 including a keyboard, a mouse, and the like; an output section 407 including devices such as a cathode ray tube (CRT), a liquid crystal display (LCD), and a speaker; a storage section 408 including a hard disk and the like; and a communication section 409 including a network interface card such as a LAN (Local Area Network) card or a modem. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to the input / output interface 405 as needed. Removable media 411, such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, is installed in the drive 410 as needed, so that computer programs read therefrom can be installed into the storage section 408 as needed.

[0136] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product that includes a computer program carried on a computer-readable medium, the computer program including a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 409 and / or installed from a removable medium 411. When the computer program is executed by the central processing unit 401, the various functions defined in the system of the present application are performed.

[0137] It should be noted that the computer-readable medium shown in the embodiments of the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, which carries a computer-readable computer program. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. A computer program embodied on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wired, or any suitable combination thereof.

[0138] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. Among them, each box in the flowchart or block diagram can represent a module, program segment, or part of the code, and the above-mentioned module, program segment, or part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0139] The units involved in the embodiments described in this application may be implemented by software or hardware, and the units described may also be set in a processor. In some cases, the names of these units do not constitute limitations on the units themselves.

[0140] According to one aspect of the present application, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in the various optional implementations described above.

[0141] As another aspect, the present application also provides a computer-readable medium, which may be included in the electronic device described in the above embodiments, or may exist independently without being incorporated into the electronic device. The computer-readable medium carries one or more programs, and when the one or more programs are executed by the electronic device, the electronic device implements the big data transaction security management and control method described in the above embodiments.

[0142] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiment of the application, the features and functions of two or more modules or units described above can be concretized in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be concretized.

[0143] Through the description of the above embodiments, it is easy for those skilled in the art to understand that the example embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solution according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the embodiments of the present application.

[0144] Those skilled in the art will readily conceive of other embodiments of the present application after considering the specification and practicing the embodiments disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of this application and include common knowledge or customary techniques in the art that are not disclosed herein.

[0145] It should be understood that the present application is not limited to the exact structures described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A method for security management and control of big data transactions, characterized in that: include: Obtaining user information and transaction data streams of both parties to the transaction, wherein the transaction data streams include structured and unstructured transaction data; generating a dynamic quantum key using the polarization base sequence selected by the transaction parties, and encrypting the transaction data stream using the dynamic quantum key to generate an encrypted ciphertext; Based on the entity relationship network constructed by the transaction data stream and the role information in the user information, dynamically desensitizing the user information and the transaction data stream to obtain desensitized data; Based on the encrypted ciphertext and the desensitized data, training is performed through federated learning to generate a transaction risk identification model; Based on the acquired real-time transaction characteristics, the risk probability is determined through the risk identification model, and a security management and control strategy is generated according to the risk probability.

2. The big data transaction security management and control method according to claim 1 is characterized in that: The obtaining of user information and transaction data streams of both parties to the transaction includes: Obtain user information and real-time transaction requests from mobile applications or POS terminals; Get transaction data stream from the payment platform.

3. The big data transaction security management and control method according to claim 1 is characterized in that: The step of generating a dynamic quantum key using the polarization base sequence selected by the transaction parties, and encrypting the transaction data stream using the dynamic quantum key to generate encrypted ciphertext includes: The polarization base sequence selected by the transaction parties is used to generate a dynamic quantum key through quantum polarization state transmission; Performing hash processing on the dynamic quantum key to generate a hash parameter; The transaction data stream is encrypted using the hash parameter to generate an encrypted ciphertext.

4. The method for security management and control of big data transactions according to claim 1, characterized in that: The entity relationship network constructed based on the transaction data stream and the role information in the user information, dynamically desensitizing the user information and the transaction data stream to obtain desensitized data, includes: Parsing data semantics based on the transaction data stream to construct a relationship network of entities involved in the transaction; identifying a data sensitivity level based on the entity relationship network and contextual information of the transaction data flow; generating a desensitization intensity coefficient according to the role information in the user information and the data sensitivity level; According to the comparison result between the desensitization intensity coefficient and the set threshold, the transaction data stream is desensitized to generate desensitized data.

5. The method for security management and control of big data transactions according to claim 1, characterized in that: The generating of a transaction risk identification model by training based on the encrypted ciphertext and the desensitized data through federated learning includes: Decrypting the encrypted ciphertext to generate decrypted data; Inputting the decrypted data and the masked data into a trusted execution environment, training a local model based on a loss function, and generating a local model weight; Submitting the hash value of the local model weight to the blockchain in the trusted execution environment, calling a verification function through the smart contract on the blockchain to verify the security of the local model weight, and obtaining a verified first weight; The first weights are aggregated to generate aggregated weights, and a global risk identification model is constructed based on the aggregated weights.

6. The method for security management and control of big data transactions according to claim 1, characterized in that: The method of determining the risk probability based on the acquired real-time transaction characteristics through the risk identification model and generating a security management and control strategy according to the risk probability includes: Determining, based on the acquired real-time transaction features, a data type corresponding to the real-time transaction features; Sending the real-time transaction features to a corresponding model according to the data type, and obtaining an output result of the corresponding model; Processing the output results through a dynamic attention mechanism to generate a risk probability; A security management and control strategy is generated based on the risk probability.

7. The method for security management and control of big data transactions according to any one of claims 1 to 6, characterized in that: Generating a security control strategy according to the risk probability includes: Obtain anomaly scores for historical transaction users' behaviors; generating a comprehensive risk index based on the risk probability, the transaction amount in the transaction data stream, and the anomaly score to determine a dynamic risk level; According to the dynamic risk level, a security control strategy is selected from a preset strategy library.

8. A big data transaction security management and control system, characterized by: include: An acquisition unit, configured to acquire user information of both parties to a transaction and a transaction data stream, wherein the transaction data stream includes structured and unstructured transaction data; an encryption unit, configured to generate a dynamic quantum key using the polarization base sequence selected by the transaction parties, and encrypt the transaction data stream using the dynamic quantum key to generate an encrypted ciphertext; a desensitizing unit, configured to dynamically desensitize the user information and the transaction data stream based on the entity relationship network constructed by the transaction data stream and the role information in the user information, thereby obtaining desensitized data; A training unit, configured to generate a transaction risk identification model by training through federated learning based on the encrypted ciphertext and the desensitized data; The management and control unit is used to determine the risk probability through the risk identification model based on the acquired real-time transaction characteristics, and generate a security management and control strategy according to the risk probability.

9. The big data transaction security management and control system according to claim 8, characterized in that: The obtaining of user information and transaction data streams of both parties to the transaction includes: Obtain user information and real-time transaction requests from mobile applications or POS terminals; Get transaction data stream from the payment platform.

10. The big data transaction security management and control system according to claim 8, characterized in that: The step of generating a dynamic quantum key using the polarization base sequence selected by the transaction parties, and encrypting the transaction data stream using the dynamic quantum key to generate encrypted ciphertext includes: The polarization base sequence selected by the transaction parties is used to generate a dynamic quantum key through quantum polarization state transmission; Performing hash processing on the dynamic quantum key to generate a hash parameter; The transaction data stream is encrypted using the hash parameter to generate an encrypted ciphertext.

Images