Customer information intelligent management platform and method based on cloud computing

Abstract

The invention discloses a client information intelligent management platform and method based on cloud computing, and the platform comprises an identity verification and dynamic authority control module which is used for carrying out the safe and efficient identity verification of a user, and dynamically adjusting the authority level of the user in real time according to the behavior and context information of the user; the encryption storage module adopts an encryption algorithm based on a lattice password and is used for performing encryption storage on customer information and enhancing the capability of resisting quantum computing attack; the data recording module is used for recording data changes by utilizing a block chain technology in the operation process of the customer information; the intelligent federation learning module is used for cooperatively training a machine learning model among the modules; the artificial intelligence safety monitoring module is used for monitoring abnormal behaviors in the system in real time; and the user interface and interaction module is used for providing a friendly user interface based on augmented reality. The client information management platform has the beneficial effect of improving the safety, reliability and intelligent degree of the client information management platform.

Description

Technical Field

[0001] The present invention relates to the technical field of cloud computing and information security management, and in particular to a customer information intelligent management platform and method based on cloud computing. Background Art

[0002] With the rapid development of cloud computing technology, customer information management platforms are gradually migrating to the cloud, realizing centralized storage and efficient processing of data. In order to ensure the security and integrity of large amounts of sensitive data, advanced cryptographic technologies continue to emerge.

[0003] However, existing customer information management platforms still have many shortcomings when integrating the above-mentioned advanced technologies. Traditional identity authentication and permission control mechanisms lack dynamic perception of user context and cannot adjust permission levels in real time according to user behavior and environment, which may lead to security risks or reduce user experience.

[0004] Faced with the potential threat of quantum computing, traditional encryption algorithms such as RSA and ECC may be hacked, and the confidentiality and integrity of data are challenged. Although homomorphic encryption technology can perform calculations on ciphertext, its computing efficiency is low and it is difficult to meet high concurrency and real-time business needs; the application of blockchain technology in data recording also faces the problems of low efficiency and insufficient scalability of consensus algorithms, which affects the performance and availability of the system.

[0005] The existing anomaly detection models do not fully utilize advanced algorithms such as GAN, and the detection accuracy and response speed of abnormal behaviors still need to be improved. These problems limit the further improvement of the customer information management platform in terms of security, efficiency and intelligence. Summary of the invention

[0006] The purpose of this section is to summarize some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the abstract of the specification and the title of the invention of this application to avoid blurring the purpose of this section, the abstract of the specification and the title of the invention, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0007] In view of the above existing problems, the present invention is proposed.

[0008] Therefore, the technical problem solved by the present invention is that the existing customer information intelligent management technology has the problems of insufficient security, inflexible authority control, low data processing efficiency and limited anomaly detection capability.

[0009] In order to solve the above technical problems, the present invention provides the following technical solutions: an identity authentication and dynamic authority control module, which is used to perform safe and efficient identity authentication on users, and dynamically adjust their authority levels in real time according to the user's behavior and context information, to ensure the security and flexibility of the system;

[0010] The encrypted storage module uses a lattice-based encryption algorithm to encrypt and store customer information, enhance the ability to resist quantum computing attacks, and ensure the confidentiality and integrity of data;

[0011] A data recording module is used to record data changes using blockchain technology during the operation of the customer information, and adopt a hybrid consensus algorithm to improve the scalability and consistency of the system;

[0012] Intelligent federated learning module, which is used to collaboratively train machine learning models between modules, improve model performance without leaving the local data, and protect data privacy;

[0013] The AI ​​security monitoring module is used to monitor abnormal behavior within the system in real time. It uses a generative adversarial network-based model to detect and respond to potential security threats;

[0014] The user interface and interaction module is used to provide a friendly user interface based on augmented reality, support multi-terminal access, and improve user experience and operating efficiency.

[0015] As a preferred solution of the cloud computing-based customer information intelligent management platform of the present invention, it includes:

[0016] The identity authentication and dynamic permission control module interacts with the user interface and interaction module, accepts the user's login request, links with the artificial intelligence security monitoring module, adjusts the user's permissions in real time, and connects with the data access module to control the user's access rights to the data;

[0017] The encryption storage module interacts with the data processing module to provide data encryption and decryption services, and is connected to the data consistency verification module to ensure the consistency of distributed storage data;

[0018] The data recording module is connected to the identity authentication and dynamic authority control module to obtain user operation information, and is connected to the storage module to save blockchain data.

[0019] As a preferred solution of the cloud computing-based customer information intelligent management platform of the present invention, it also includes:

[0020] The intelligent federated learning module is connected to the local data storage of each module and interacts with the central server to update and distribute the model;

[0021] The artificial intelligence security monitoring module interacts with the identity authentication and dynamic permission control module to adjust user permissions and connect to the log system to record abnormal behavior;

[0022] The user interface and interaction module is connected with all modules and is responsible for displaying information and transmitting user operations.

[0023] As a preferred solution of the cloud computing-based customer information intelligent management method of the present invention, the management method specifically includes the following steps:

[0024] User authentication and dynamic allocation of permissions;

[0025] Access and processing of encrypted data;

[0026] Records and consensus on data operations;

[0027] Training and updating of intelligent federated learning models;

[0028] Security monitoring and response;

[0029] Real-time dynamic adjustment of user permissions.

[0030] As a preferred solution of the method for intelligent management of customer information based on cloud computing described in the present invention, the user identity authentication and dynamic authority allocation include:

[0031] The user opens the login interface of the management platform and enters the user name ID;

[0032] The management platform generates a random challenge value and sends it to the user;

[0033] The user generates a zero-knowledge proof using the Schnorr protocol using the private key and the challenge value and sends it to the management platform;

[0034] The management platform verifies the validity of the zero-knowledge proof, collects context information of the user, and calculates the permission level of the user through a policy function;

[0035] The management platform allocates corresponding access rights according to the authority level of the user.

[0036] As a preferred solution of the method for intelligent management of customer information based on cloud computing described in the present invention, the access and processing of encrypted data includes:

[0037] When a user requests to access the encrypted customer information ciphertext, the management platform verifies the user's authority level and confirms that the user has the right to access the requested data;

[0038] The management platform executes the calculation function f on the ciphertext in the homomorphic encryption domain to obtain the result ciphertext;

[0039] The management platform returns the encrypted result to the user;

[0040] The user uses a private key to decrypt the ciphertext to obtain a plaintext result.

[0041] As a preferred solution of the method for intelligent management of customer information based on cloud computing described in the present invention, the record and consensus of data operations include:

[0042] When a user views or edits customer information, the management platform generates a transaction block containing the operation details and verifies the legitimacy of the transaction block through a hybrid consensus mechanism.

[0043] As a preferred solution of the cloud computing-based customer information intelligent management method of the present invention, the training and updating of the intelligent federated learning model includes:

[0044] Each module trains local model parameters on a local dataset, calculates the model update amount, adds differential privacy noise, and then sends it to the central server;

[0045] The central server aggregates the model update amounts of each module, obtains new global model parameters, and sends them to each module to enter the next round of training.

[0046] As a preferred solution of the method for intelligent management of customer information based on cloud computing described in the present invention, the security monitoring and response includes:

[0047] The management platform collects user behavior data and management platform logs in real time to form a feature set;

[0048] Analyzing the feature set using an anomaly detection model based on a generative adversarial network;

[0049] If abnormal behavior is detected, the management platform triggers a security warning mechanism.

[0050] As a preferred solution of the method for intelligent management of customer information based on cloud computing described in the present invention, the real-time dynamic adjustment of user rights includes:

[0051] The management platform continuously monitors the context information of the user, and when a change in the context information is detected, the management platform recalculates the permission level of the user;

[0052] The management platform dynamically adjusts the access rights of the user according to the new permission level.

[0053] Beneficial effects of the present invention:

[0054] 1. Enhanced the security of identity authentication to prevent unauthorized access;

[0055] 2. Implemented dynamic permission control based on user context, improving the flexibility and security of the system;

[0056] 3. Adopting homomorphic encryption and differential privacy technology to protect the confidentiality and privacy of data;

[0057] 4. Using blockchain and hybrid consensus mechanism to ensure the immutability and traceability of data operations;

[0058] 5. Through federated learning and GAN anomaly detection model, the system's intelligence level and security monitoring capabilities are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0060] Figure 1 It is a flow chart of the method for intelligent management of customer information based on cloud computing shown in the present invention;

[0061] Figure 2 It is a schematic diagram of module distribution association of the customer information intelligent management platform based on cloud computing shown in the present invention;

[0062] Figure 3 This is a schematic diagram of anomaly detection based on GAN shown in the present invention. DETAILED DESCRIPTION

[0063] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0064] Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without making any creative work should fall within the scope of protection of the present invention.

[0065] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0066] According to an embodiment of the present invention, Figure 1 The flowchart shown is a method for intelligent management of customer information based on cloud computing, which specifically includes the following steps:

[0067] (1) User identity authentication and permission allocation

[0068] The user opens the management platform login interface through a terminal device (such as a computer or mobile phone) and enters the user name ID;

[0069] In order to verify the user's identity, the management platform generates a random challenge value c and sends it to the user;

[0070] User generates zero-knowledge proof:

[0071] Select a random number: The user selects a random number r;

[0072] Calculate the commitment value: Calculate X = g r modp;

[0073] Calculation proof: Calculate y = r + c·s modq;

[0074] Send proof: The user sends (X, y) to the management platform;

[0075] Management platform verification certificate:

[0076] Calculate the verification value: The management platform calculates X′=g y ·v -c modp;

[0077] Verification result: If X′=X, the proof is valid and the user identity verification is successful; otherwise, access is denied;

[0078] Among them, g is a generator, which is a number published by the management platform and used for key calculation. p and q are large prime numbers, satisfying that q is a factor of p-1. s is the user's private key, which is only known to the user himself. v is the user's public key. v=g s The modp calculation shows that c is the random challenge value generated by the management platform to prevent replay attacks, r is the random number selected by the user to ensure the uniqueness of each proof, and X and y are zero-knowledge proofs generated by the user;

[0079] Collect context information:

[0080] After verification, the management platform collects the user's context information U, including but not limited to:

[0081] Time t: the time when the user logs in, to determine whether it is within normal working hours;

[0082] Judgment rule: judge whether the user login time is within the preset normal working hours, such as 9:00 to 18:00 every day;

[0083] Judgment method: Get the user login timestamp t login , determine whether 9:00≤t login ≤18:00;

[0084] Judgment result: If it is within normal working hours, then t = 1, otherwise, t = 0;

[0085] Location l: The user's geographic location information, to determine whether they are in a trusted location;

[0086] Judgment rules: determine whether the user's geographic location is within the scope of trusted locations, such as the company's office area;

[0087] Judgment method: Obtain user location through IP address and GPS login , determine whether it belongs to the preset trusted location set L trusted ;

[0088] Judgment result: If it is in a trusted location, then l = 1, otherwise, l = 0;

[0089] Device information d: the type of device, operating management platform, and browser used by the user;

[0090] Judgment rule: judge whether the device used by the user is in the trusted device list;

[0091] Judgment method: Get device identifier (such as MAC address, device ID) login , determine whether it belongs to the trusted device set D trusted ;

[0092] Judgment result: If it is a trusted device, d = 1, otherwise, d = 0;

[0093] Calculation permission level:

[0094] Policy function L = f(U): The management platform uses the policy function to calculate the user's permission level L;

[0095] Strategy function implementation:

[0096] Rule-based policy: Define a set of permission rules, for example: if t is within working hours and l is within the company, then L = high, if d is an unknown device, then L = low;

[0097] Policy function f(U):

[0098] L=f(U)=w t ·t+w ll+w d ·d+b

[0099] Among them, LLL is the user's authority level. The higher the value, the higher the authority. U = [t, l, d] is the user's context information feature vector, w t ,w l ,w d is the weight coefficient of the corresponding feature, reflecting the influence of each factor on the authority level, and b is the bias term;

[0100] Machine learning model: Use algorithms such as random forest to learn permission allocation strategies from historical data;

[0101] Data collection: Collect a large amount of historical login records, including feature U and the corresponding actual permission level L;

[0102] Model training: Use the random forest algorithm, with U as the input feature and L as the target variable, to train the regression model;

[0103] The model automatically learns the importance of each feature and obtains the weight w t ,w l ,w d and bias b;

[0104] Calculation process example: Assume that the weight obtained by training is w t =0.5, w l =0.3, w d =0.2, bias b=0, if a user's U=[1,0,1], then L=0.5×1+0.3×0+0.2×1+0=0.7;

[0105] Allocate access rights: The management platform allocates corresponding access rights to users based on the calculated permission level L, updates the user's permission configuration, and loads the corresponding permission policy;

[0106] According to the value range of L, the permission level is divided into different levels and their corresponding permissions:

[0107] Advanced permissions (Level 3): L≥1.5, allowing access to all data and functions;

[0108] Intermediate authority (Level 2): ​​1.0≤L<1.5, allowing access to most data, restricting some sensitive functions;

[0109] Low-level permissions (Level 1): L<1.0, only allowing access to public data and basic functions.

[0110] It should be noted that this step uses a zero-knowledge proof protocol to ensure that users complete identity authentication without disclosing private keys, thereby enhancing security, and utilizing dynamic permission allocation to adjust permissions based on real-time contextual information to balance security and user experience.

[0111] (2) Access and processing of encrypted data

[0112] After passing identity authentication, the user requests the management platform to access the encrypted customer information ciphertext C;

[0113] Permission verification: The management platform checks the user's permission level L to confirm whether the user has the right to access the requested data. If the permission is insufficient, access is denied and a log is recorded;

[0114] Homomorphic encryption computing:

[0115] Allows specific operations to be performed directly on ciphertext, and the decrypted result is the same as the result of the operation on the plaintext;

[0116] The management platform executes the calculation function f:

[0117] Execute the calculation function f on the ciphertext C to obtain the result ciphertext C result ;

[0118] Supported calculation functions include:

[0119] Data statistics: such as summing f sum , mean value f avg ;

[0120] Data screening: such as conditional query f filter ;

[0121] Data conversion: such as format conversion transform ;

[0122] The result is:

[0123] The management platform will cipher the result C result Return to the user, ensuring that the data remains encrypted during transmission;

[0124] User decryption result:

[0125] The user uses the private key s to encrypt the resulting ciphertext C result Decrypt and get the plaintext result M result .

[0126] As an example, the embodiment of the present invention uses the LWE algorithm as an example for encryption, and the encryption process is as follows:

[0127] Plain text data: M

[0128] encryption:

[0129] C=A T s+e+Mmodq

[0130] Among them, A T is the public key matrix, s is the private key, e is the error vector, and C is the ciphertext;

[0131] Execute the computation function f on the ciphertext:

[0132] Given two ciphertexts C1 and C2, the corresponding plaintexts are M1 and M2;

[0133] Ciphertext addition:

[0134] C sum =C1+C2modq

[0135] After decryption, we get:

[0136] M sum =M1+M2modq

[0137] It should be noted that the LWE algorithm supports limited homomorphic multiplication, and its specific implementation is relatively complex. In this embodiment, it is simplified, and the user uses the private key s for decryption:

[0138] M′=C-ATsmodq

[0139] Among them, M′ is the plaintext result after decryption, which is equivalent to the original plaintext M.

[0140] Preferably, this step supports addition and multiplication operations on ciphertext through a homomorphic encryption algorithm, namely a lattice-based encryption algorithm (LWE), thereby protecting data confidentiality.

[0141] (3) Records and consensus of data operations

[0142] When users view and edit customer information, the management platform generates a transaction block B, which includes:

[0143] Operation type: view, modify, delete;

[0144] Data summary: hash value of the data involved in the operation to ensure integrity;

[0145] Timestamp: the exact time when the operation occurred;

[0146] User ID: the identity of the operator who performs the operation;

[0147] Hybrid consensus mechanism verification:

[0148] Consensus algorithm: The management platform adopts a hybrid consensus mechanism of Proof of Stake (PoS) and Practical Byzantine Fault Tolerance (PBFT);

[0149] Verification process:

[0150] Node verification: The verification nodes in the network verify the transaction block B and check its legitimacy and integrity;

[0151] Voting: Nodes reach consensus through voting, which must satisfy n≥3f+1, where n is the total number of nodes and f is the maximum number of malicious nodes that can be tolerated;

[0152] Reaching consensus: If two-thirds of the nodes agree, transaction block B is confirmed;

[0153] Blocks are added to the blockchain:

[0154] The confirmed transaction block B is added to the blockchain, forming an immutable and traceable chain of records.

[0155] As an example, the legitimacy check includes digital signature verification and transaction format verification, such as verifying whether the signature of transaction block BBB is valid, ensuring that the data has not been tampered with, and checking whether the structure and fields of the block meet predefined specifications.

[0156] As an example, integrity checking is implemented through hash verification, such as calculating the hash value of the block and comparing it with the hash value recorded in the block header.

[0157] As an example, the consensus process:

[0158] Pre-preparation: The master node broadcasts transaction block B to all backup nodes;

[0159] Prepare: The backup node verifies the legitimacy and integrity of B and sends a "prepare" message;

[0160] Commit: If at least two-thirds of the "Prepare" messages are received, the node sends a "Commit" message;

[0161] Confirmation: When the node receives enough "commit" messages, it considers that B has reached a consensus and records it in the blockchain.

[0162] (4) Model training and updating

[0163] Each medical institution has a local dataset D i The training model parameters W i , using a unified model structure and training algorithm (such as neural network and Adam optimizer);

[0164] Calculate the model update and add differential privacy noise:

[0165] Calculate the update amount: ΔW i =W i -W global , where Wglobal is a global model parameter;

[0166] Add differential privacy noise:

[0167] Noise generation: From a normal distribution N(0,δ 2 ) in the sampling noise σ;

[0168] Update amount protection: ΔW′ i =ΔW i +σ,δ, where δ is the noise scale, controlling the privacy protection strength;

[0169] Send updates to the central server:

[0170] Each mechanism will add the update amount of noise ΔW′ i Send securely to a central server;

[0171] Central server aggregates model updates:

[0172] Aggregate calculation: Where N is the number of participating institutions;

[0173] Update global model parameters: get new W global ;

[0174] Send global model parameters: The central server sends the updated global model parameters W global Distribute to each institution (independent data holder) to start the next round of training;

[0175] Training stop conditions:

[0176] Convergence condition: The change in the loss function of the global model is less than the preset threshold;

[0177] Maximum number of iterations: reaches the preset maximum number of training rounds (such as 100);

[0178] Performance indicator: The accuracy of the model on the validation set reaches the target value.

[0179] The above preset thresholds and target values ​​can be set according to actual application scenarios, and are not limited to them in this embodiment.

[0180] It should be noted that this step uses federated learning to achieve joint modeling of multiple institutions without sharing the original data, protects data privacy, and prevents attackers from inferring the original data through the model update amount by adding noise.

[0181] (5) Security monitoring and response

[0182] The management platform collects user behavior data and management platform logs in real time to form a feature set X, including:

[0183] Login frequency: the number of times a user logs in within a certain period of time (such as 24 hours);

[0184] Operation type: the type and number of operations performed by the user;

[0185] Abnormal events: management platform errors, permission denial;

[0186] Anomaly detection model analysis:

[0187] GAN model composition:

[0188] Generator G: learns normal behavior patterns and generates fake normal data;

[0189] Discriminator D: distinguishes between real data and generated data;

[0190] Model training: Use normal behavior data (such as the user's normal login records, access behavior, and operation sequences) to train G and D, making the data generated by G difficult to be distinguished by D;

[0191] Generator G:

[0192] Input random noise z and output forged normal behavior data G(z);

[0193] Discriminator D:

[0194] Input behavioral data x, and output the probability D(X) of judging it as true data;

[0195] The objective function is as follows:

[0196]

[0197] in, is the real data distribution, is the noise distribution;

[0198] Real-time detection:

[0199] Input the real-time feature set X into D and calculate the output D(X), whose value is between [0,1];

[0200] Threshold setting: According to the performance of the model on the validation set, set the threshold θ, such as θ = 0.4;

[0201] Judging criteria:

[0202] If D(X)<θ, it is considered that abnormal behavior is detected and the safety warning mechanism is triggered;

[0203] If D(X)≥θ, the discriminator considers X to be a normal behavior;

[0204] Safety early warning mechanism:

[0205] Limit user rights: temporarily reduce or freeze user rights to prevent the spread of potential risks;

[0206] Notify security administrators: Send alert information with detailed data of abnormal behavior;

[0207] Record logs: Save the results of anomaly detection and related data for later analysis.

[0208] As an example, a batch of normal behavior data and abnormal behavior data are simulated, and the discriminator is used to evaluate the real-time feature set, and whether there is abnormal behavior is determined based on the threshold. Normal behavior data: simulates the characteristics of the user's login frequency and operation times under normal circumstances, and the discriminator outputs a higher D(X) value, close to 1; abnormal behavior data: simulates the characteristics under abnormal circumstances, such as abnormally high login frequency, and the discriminator outputs a lower D(X) value, close to 0; threshold setting: set the threshold θ = 0.4, when D(X) < θ, it is considered that abnormal behavior is detected;

[0209] Reference Figure 3 , horizontal axis (sample number): represents the sample number, from 1 to 100, a total of 100 sample data are simulated; vertical axis (discriminator output D(X): represents the output value of the discriminator for each sample, ranging from [0,1]; solid line: represents the output D(X) of the discriminator for all samples; dotted line: represents the threshold value θ = 0.4, when D(X) is lower than this value, it is judged as abnormal behavior; black circle mark: represents the detected abnormal behavior point;

[0210] like Figure 3 As shown, at sample numbers 30, 31, 32, 70, 71, and 72, the discriminator outputs are significantly lower than the threshold θ and are marked as abnormal behaviors. The discriminator outputs of the remaining samples are all higher than the threshold and are considered normal behaviors.

[0211] (6) Dynamic Adjustment of User Rights

[0212] The management platform continuously obtains the user's context information U, such as location, device, and network environment;

[0213] Detecting context changes:

[0214] Change judgment: compare the current context U current With the previous context U previous Is it consistent?

[0215] Change type: location change, equipment replacement, network switch;

[0216] Recalculate permission levels:

[0217] Strategy function calculation: L = f(U current );

[0218] Model application: Use the policy function trained in step (1) to calculate the new permission level based on the new context information;

[0219] Dynamically adjust access permissions:

[0220] Permission update: According to the new permission level L, adjust the user's operation permissions in the management platform. The corresponding relationship between permission level L and operation permissions is shown in the following table:

[0221] Table 1. Permission mapping table

[0222] Clearance Level L Executable operations L≥1.5 View, edit, delete, export data and other operations 1.0≤L<1.5 View, edit, delete and export prohibited L<1.0 View only, no editing, deleting or exporting

[0223] According to the new level permission L, the permission mapping table is searched to determine the set of operations that the user can perform;

[0224] Update the user's permission configuration in the system and enable or disable the corresponding functional modules;

[0225] Notify users: Choose to notify users of permission changes to improve transparency.

[0226] As an example, functional module control: restricting user access to specific functions through the front-end interface and the back-end interface; data access control: verifying the user's permission level when requesting data and rejecting operations that exceed the permission range.

[0227] Preferably, the embodiments of the present invention improve the security, flexibility and intelligence level of the system by introducing zero-knowledge proof identity authentication, dynamic permission control based on user context, quantum computing-resistant encrypted storage, blockchain-based hybrid consensus mechanism, federated learning and GAN-based anomaly detection model, and solve the problems of insufficient security, inflexible permission control, low data processing efficiency and limited anomaly detection capability of existing customer information intelligent management technologies.

[0228] In the application of the above embodiments, other aspects disclosed in the embodiments of the present invention also propose a customer information intelligent management platform based on cloud computing, referring to Figure 2 The management platform includes an identity authentication and dynamic permission control module, an encryption storage module, a data recording module, an intelligent federated learning module, an artificial intelligence security monitoring module and a user interface and interaction module.

[0229] As an example, the identity authentication and dynamic permission control module is used to perform secure and efficient identity authentication on users and dynamically adjust their permission levels in real time according to the user's behavior and context information to ensure the security and flexibility of the system;

[0230] The identity authentication and dynamic permission control module interacts with the user interface and interaction module, accepts the user's login request, links with the artificial intelligence security monitoring module, adjusts the user's permissions in real time, and connects with the data access module to control the user's access rights to the data;

[0231] in:

[0232] Improved zero-knowledge proof protocol: The user enters the user name ID, the system generates a random challenge value, and the user uses the private key and the challenge value to generate a zero-knowledge proof, such as using the improved Schnorr protocol to ensure that the user's identity is verified without leaking the private key;

[0233] Dynamic permission allocation: The system collects the user's contextual information (such as time, location, and device information) and calculates the user's permission level through a policy function. The policy function is trained based on a machine learning model (such as random forest) and can dynamically adjust permissions based on different contexts.

[0234] As an example, an encrypted storage module, which uses an encryption algorithm based on lattice cryptography to encrypt and store customer information, enhance the ability to resist quantum computing attacks, and ensure the confidentiality and integrity of data;

[0235] The encryption storage module interacts with the data processing module to provide data encryption and decryption services, and connects with the data consistency verification module to ensure the consistency of distributed storage data;

[0236] in:

[0237] Lattice-based encryption algorithm: Use the learning and error (LWE) algorithm to encrypt plaintext data and generate ciphertext;

[0238] Distributed hybrid cloud storage: Encrypted data is stored in a distributed hybrid cloud architecture consisting of private cloud and public cloud to improve data availability and security.

[0239] As an example, a data recording module is used to record data changes using blockchain technology during the operation of the customer information, and adopt a hybrid consensus algorithm to improve the scalability and consistency of the system;

[0240] The data recording module is connected with the identity authentication and dynamic permission control module to obtain user operation information, and is connected with the storage module to save blockchain data;

[0241] in:

[0242] Hybrid consensus algorithm: Combining Proof of Stake (PoS) and Practical Byzantine Fault Tolerance (PBFT) algorithms to ensure that the network can still reach consensus even in the presence of malicious nodes;

[0243] Data records: The user's operation records are packaged into blocks and added to the blockchain through a consensus mechanism to ensure the data's immutability and traceability.

[0244] As an example, the intelligent federated learning module is used to collaboratively train machine learning models between modules, improve model performance without leaving the local data, and protect data privacy;

[0245] The intelligent federated learning module is connected to the local data storage of each module and interacts with the central server to update and distribute the model;

[0246] in:

[0247] Local model training: Each institution trains model parameters on a local dataset;

[0248] Model update and aggregation: The local model update is added with differential privacy noise and sent to the central server. The central server aggregates the update to obtain new global model parameters.

[0249] As an example, the AI ​​security monitoring module is used to monitor abnormal behaviors within the system in real time. It uses a model based on generative adversarial networks to detect and respond to potential security threats;

[0250] The AI ​​security monitoring module interacts with the identity authentication and dynamic permission control module to adjust user permissions and connect to the log system to record abnormal behavior;

[0251] in:

[0252] Anomaly detection model: Using the GAN model, the generator learns normal behavior patterns and the discriminator detects abnormal behavior;

[0253] Security response: When abnormal behavior is detected, the security warning mechanism is triggered, user permissions are restricted, and the security administrator is notified.

[0254] As an example, the user interface and interaction module is used to provide a friendly user interface based on augmented reality, support multi-terminal access, and improve user experience and operational efficiency.

[0255] The user interface and interaction module is connected with all modules and is responsible for the display of information and the transmission of user operations;

[0256] in:

[0257] Multi-terminal support: compatible with Web, mobile and wearable devices, providing a consistent user experience;

[0258] Interaction method: Supports multiple interaction methods such as gestures and voice, interacts with various background modules in real time, and provides feedback on operation results.

[0259] The management platform also includes: one or more processors and a memory.

[0260] The memory is used to store operable instructions, which, when executed by the one or more processors, cause the one or more processors to perform operations, including the process of the cloud computing-based customer information intelligent management method of the aforementioned embodiment, especially Figure 1 The process of the method shown.

[0261] The management platform also includes a computer-readable medium storing software, the software including instructions that can be executed by one or more computers, and the instructions, through such execution, enable one or more computers to perform operations, the operations including the process of the customer information intelligent management method based on cloud computing of the aforementioned embodiment, especially Figure 1 The process of the method shown.

[0262] It should be appreciated that embodiments of the present invention can be implemented or carried out by computer hardware, a combination of hardware and software, or by computer instructions stored in a non-transitory computer-readable memory.

[0263] The method may be implemented in a computer program using standard programming techniques, including a non-transitory computer-readable storage medium configured with a computer program, wherein the storage medium so configured causes a computer to operate in a specific and predefined manner.

[0264] Each program can be implemented in a high-level procedural or object-oriented programming language to communicate with a computer system, however, the program can be implemented in assembly or machine language if desired.

[0265] In any case, the language may be a compiled or interpreted language.

[0266] Furthermore, the program can be run on an application specific integrated circuit programmed for this purpose.

[0267] The processes described herein (or variations and / or combinations thereof) may be performed under the control of one or more computer systems configured with executable instructions, and may be implemented as code (e.g., executable instructions, one or more computer programs, or one or more applications) that is executed collectively on one or more processors, by hardware, or a combination thereof. The computer program includes a plurality of instructions that can be executed by one or more processors.

[0268] Further, the method may be implemented in any type of computing platform operably connected to a suitable computer, including but not limited to a personal computer, a minicomputer, a mainframe, a workstation, a network or distributed computing environment, a separate or integrated computer platform, or in communication with a charged particle tool or other imaging device.

[0269] Various aspects of the present invention may be implemented in machine-readable code stored on a non-transitory storage medium or device, whether removable or integrated into a computing platform, such as a hard disk, optical read and / or write storage media, RAM, ROM, etc., so that it can be read by a programmable computer, which, when the storage medium or device is read by the computer, can be used to configure and operate the computer to perform the processes described herein.

[0270] Additionally, the machine-readable code, or portions thereof, can be transmitted over a wired or wireless network.

[0271] The invention described herein includes these and other various types of non-transitory computer-readable storage media when such media include instructions or programs that implement the steps described above in conjunction with a microprocessor or other data processor.

[0272] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A customer information intelligent management platform based on cloud computing, characterized in that: include: The identity authentication and dynamic permission control module is used to perform safe and efficient identity authentication on users and dynamically adjust their permission levels in real time according to the user's behavior and context information to ensure the security and flexibility of the system; The encrypted storage module uses a lattice-based encryption algorithm to encrypt and store customer information, enhance the ability to resist quantum computing attacks, and ensure the confidentiality and integrity of data; A data recording module is used to record data changes using blockchain technology during the operation of the customer information, and adopt a hybrid consensus algorithm to improve the scalability and consistency of the system; Intelligent federated learning module, which is used to collaboratively train machine learning models between modules, improve model performance without leaving the local data, and protect data privacy; The AI ​​security monitoring module is used to monitor abnormal behavior within the system in real time. It uses a generative adversarial network-based model to detect and respond to potential security threats; The user interface and interaction module is used to provide a friendly user interface based on augmented reality, support multi-terminal access, and improve user experience and operating efficiency.

2. The cloud computing-based customer information intelligent management platform according to claim 1, characterized in that: include: The identity authentication and dynamic permission control module interacts with the user interface and interaction module, accepts the user's login request, links with the artificial intelligence security monitoring module, adjusts the user's permissions in real time, and connects with the data access module to control the user's access rights to data; The encryption storage module interacts with the data processing module to provide data encryption and decryption services, and is connected to the data consistency verification module to ensure the consistency of distributed storage data; The data recording module is connected to the identity authentication and dynamic authority control module to obtain user operation information, and is connected to the storage module to save blockchain data.

3. The cloud computing-based customer information intelligent management platform according to claim 1, characterized in that: Also includes: The intelligent federated learning module is connected to the local data storage of each module and interacts with the central server to update and distribute the model; The artificial intelligence security monitoring module interacts with the identity authentication and dynamic permission control module to adjust user permissions and connect to the log system to record abnormal behavior; The user interface and interaction module is connected with all modules and is responsible for displaying information and transmitting user operations.

4. The customer information intelligent management platform according to claim 1, characterized in that: Each module is implemented by applying the management method of the customer information intelligent management platform, and the management method specifically includes the following steps: User authentication and dynamic allocation of permissions; Access and processing of encrypted data; Records and consensus on data operations; Training and updating of intelligent federated learning models; Security monitoring and response; Real-time dynamic adjustment of user permissions.

5. The cloud computing-based customer information intelligent management platform according to claim 4 is characterized in that: The user identity authentication and dynamic authority allocation include: The user opens the login interface of the management platform and enters the user name ID; The management platform generates a random challenge value and sends it to the user; The user generates a zero-knowledge proof using the Schnorr protocol using the private key and the challenge value and sends it to the management platform; The management platform verifies the validity of the zero-knowledge proof, collects context information of the user, and calculates the permission level of the user through a policy function; The management platform allocates corresponding access rights according to the authority level of the user.

6. The method for intelligent management of customer information based on cloud computing according to claim 4, characterized in that: The access and processing of the encrypted data includes: When a user requests to access the encrypted customer information ciphertext, the management platform verifies the user's authority level and confirms that the user has the right to access the requested data; The management platform executes the calculation function f on the ciphertext in the homomorphic encryption domain to obtain the result ciphertext; The management platform returns the encrypted result to the user; The user uses a private key to decrypt the ciphertext to obtain a plaintext result.

7. The method for intelligent management of customer information based on cloud computing according to claim 4, characterized in that: The records and consensus of the data operations include: When a user views or edits customer information, the management platform generates a transaction block containing the operation details and verifies the legitimacy of the transaction block through a hybrid consensus mechanism.

8. The method for intelligent management of customer information based on cloud computing according to claim 4, characterized in that: The training and updating of the intelligent federated learning model includes: Each module trains local model parameters on a local dataset, calculates the model update amount, adds differential privacy noise, and then sends it to the central server; The central server aggregates the model update amounts of each module, obtains new global model parameters, and sends them to each module to enter the next round of training.

9. The method for intelligent management of customer information based on cloud computing according to claim 4, characterized in that: The security monitoring and response includes: The management platform collects user behavior data and management platform logs in real time to form a feature set; Analyzing the feature set using an anomaly detection model based on a generative adversarial network; If abnormal behavior is detected, the management platform triggers a security warning mechanism.

10. The method for intelligent management of customer information based on cloud computing according to claim 4, characterized in that: The real-time dynamic adjustment of the user rights includes: The management platform continuously monitors the context information of the user, and when a change in the context information is detected, the management platform recalculates the permission level of the user; The management platform dynamically adjusts the access rights of the user according to the new permission level.

Images