Construction method of cross-platform real-name DID credit point system

Abstract

The invention discloses a construction method of a cross-platform real-name DID credit point system, and relates to the technical field of information security and credit management, and the method comprises the following steps: S1, constructing a DID identity code; s2, multi-platform behavior label mapping and unified structure tensor generation modeling are carried out; s3, constructing a cross-platform consistency verification model; s4, identifying an abnormal value; s5, constructing a credit factor function call matrix; s6, generating real-name credit points; by setting a behavior semantic difference tensor construction function and a consistency verification index function and based on a semantic behavior tensor model of a user DID index, structure-level comparison and consistency measurement are performed on label mapping results formed by a user in multiple platforms, so that the problem of distortion of platform behavior characteristics in a traditional weighted fusion mode is avoided, and the user experience is improved. The robustness and interpretability of behavior consistency calculation are effectively improved, and it is ensured that a behavior aggregation result has high consistency and verifiability.

Description

Technical Field

[0001] The present invention relates to the technical field of information security and credit management technology, and specifically to a method for constructing a cross-platform real-name DID credit points system. Background Art

[0002] With the continuous expansion and differentiation of internet application service platforms, users have generated a large number of heterogeneous and fragmented behavioral records across multiple platforms. These behavioral records involve credit performance in multiple dimensions, such as credit payments, content evaluations, and task completion rates. Due to the lack of unified identity identification and data structure specifications across platforms, user behavior records are difficult to effectively collect, model, and share. As a result, a cross-platform credit mutual recognition mechanism has long been unsuccessful, seriously hindering credit collaboration and co-construction across multiple platforms. Therefore, a technical solution is urgently needed that can achieve structured modeling and consistency verification of user cross-platform behavior based on a unified identity system while protecting user privacy, thereby building a decentralized credit management system.

[0003] In the existing technology, existing cross-platform behavioral data fusion solutions mostly use weighted aggregation or simple label alignment methods, lacking structural verification methods for the semantic consistency of multi-platform behaviors and unable to measure the consistency of user behavior logic in a unified semantic tensor space. As a result, cross-platform credit identification results are easily affected by platform behavior heterogeneity and lack verifiability and credibility. Summary of the Invention

[0004] In response to the shortcomings of the existing technology, the present invention provides a method for constructing a cross-platform real-name DID credit points system to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] In a first aspect, an embodiment of the present invention provides a method for constructing a cross-platform real-name DID credit score system, comprising the following steps:

[0007] S1. Use the cross-platform user identity authentication mechanism to construct the DID identity code and obtain the user's unique cross-platform identity;

[0008] S2. Use the DID identity code to perform multi-platform behavior label mapping and unified structure tensor generation modeling to obtain the user semantic behavior tensor model;

[0009] S3. Use the user semantic behavior tensor to build a cross-platform consistency verification model and obtain a multi-source behavior consistency score matrix;

[0010] S4. Use the multi-source behavior consistency score matrix to identify outliers and obtain a cross-platform behavior anomaly score vector;

[0011] S5. Use the anomaly score vector to construct a credit factor function call matrix to obtain an encapsulated factor interface mapping structure;

[0012] S6. Use the credit factor structure tensor to generate real-name credit points and obtain a cross-platform unified real-name DID credit point result.

[0013] To further optimize this technical solution, step S1 generates a user identity code DID by integrating the user's real-name registration information on multiple platforms and using a decentralized identity authentication mechanism u .

[0014] To further optimize this technical solution, the unified structure tensor generation model in step S2 is:

[0015]

[0016] in, The set of behavioral labels for user u on the mth platform, m∈{1,2,...,M};

[0017] Cross-platform tag semantic mapping function, mapping each platform tag to a unified tag ontology set;

[0018] Φ(·, DID u ): Structural tensor generation function, which performs multi-dimensional structural embedding based on a unified tag ontology set and DID user identity;

[0019] The final generated semantic behavior tensor model, where M is the number of access platforms, N is the total number of standardized tag ontologies, and K is the measure / intensity of the tag's behavioral response to the user in the corresponding platform.

[0020] To further optimize this technical solution, the unified structure tensor generation model in step S2 includes the following process when used:

[0021] Platform tag aggregation: based on DID u The primary key indexes the user’s tag set on M platforms, and we get

[0022] Unified tag ontology mapping: through function Align platform native tags to a unified tag ontology set to resolve semantic inconsistencies across multiple platforms;

[0023] Label semantic embedding and tensorization: Using Φ(·, DID u ), embed the unified tag ontology into the user behavior response model on each platform, and construct a tensor based on the three dimensions of platform-tag-response intensity;

[0024] Tensor generation and storage: Outputs consistent structure, unified semantics, and traceable behavioral response tensors

[0025] To further optimize this technical solution, step S3 uses a multi-platform semantic label behavior slice comparison model to measure the consistency of behavior between different platforms. The model formula is:

[0026]

[0027] in, Behavior sequence under label dimension m and behavior type k;

[0028] Ωp ( ) : The label behavior feature encoding function of the p-th platform, used to extract features from the time series;

[0029] Λ(·): cross-platform behavioral feature consistency calculation function, used to evaluate the consistency of multi-platform feature results;

[0030] The final consistency score under label dimension m and behavior type k.

[0031] To further optimize this technical solution, the platform behavior feature encoding function in step S3 is:

[0032]

[0033] in, Represents the time series of behaviors under the platform;

[0034] The second-order difference of the time series is used to characterize behavioral volatility;

[0035] Γ(·): compression function.

[0036] To further optimize this technical solution, the consistency measurement function in step S3 is:

[0037]

[0038] Among them, x p : is the behavioral characteristic value on platform p;

[0039] ||·||1: L1 norm, used to measure the difference between features;

[0040] The function output value range is [0,1], and the closer to 1, the higher the consistency.

[0041] To further optimize this technical solution, the multi-platform semantic label behavior slice comparison model in step S3 includes:

[0042] Input unified tensor structure Slice by label dimension m and behavior type k to obtain the behavior time series;

[0043] Using the subfunction Ω p (·) Extract volatility features from time series of different platforms;

[0044] The consistency measurement function Λ(·) is used to summarize the differences in behavioral characteristics of all platforms and obtain the consistency score under the label-behavior type dimension;

[0045] Repeat the above process, traverse M label dimensions and L behavior types, and output the consistency matrix

[0046] Further optimize this technical solution, the step S4 outputs the cross-platform behavior anomaly score vector summarized by user dimension The output process includes:

[0047] Consistency score matrix aggregation:

[0048] Aggregate the behavior type dimension k to obtain the consistency average or stability index under each label dimension m:

[0049]

[0050] Anomaly score conversion:

[0051] Combined with the mean μ of all user behavior consistency m and standard deviation σ m , the anomaly score is calculated using the Z-score standardization method:

[0052]

[0053] μ m : The mean consistency score of all users under the label dimension m;

[0054] σ m : The standard deviation of the consistency scores of all users under the label dimension m;

[0055] ∈: A small constant to prevent division by zero, set to 10 -6 .

[0056] To further optimize this technical solution, the credit factor function mapping matrix is constructed in step S5:

[0057]

[0058] in, The semantic tensor strength of the user under the label dimension m, behavior type k, and platform p;

[0059] Behavior abnormality score, the larger the value, the less trustworthy it is;

[0060] α∈[0,1]: Abnormal suppression factor, which controls the degree to which abnormal behavior reduces the credit factor weight;

[0061] φ(·): nonlinear factor mapping function:

[0062] φ(x)=log(1+max(x,0));

[0063] In step S5, a packaged credit interface structure tensor is constructed:

[0064]

[0065] in, The mth type of credit factor interface provided by platform p;

[0066] Call trigger threshold, the package call is only made when the factor value exceeds the set value;

[0067] The result structure of the factor interface call corresponding to the user in the mth label dimension and platform p.

[0068] In a second aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of a method for constructing a cross-platform real-name DID credit points system as described in the first aspect of the present invention are implemented.

[0069] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of a method for constructing a cross-platform real-name DID credit points system as described in the first aspect of the present invention are implemented.

[0070] Compared with the existing technology, this invention provides a method for constructing a cross-platform real-name DID credit score system, which has the following beneficial effects:

[0071] This method for constructing a cross-platform real-name DID credit points system sets a behavior semantic difference tensor constructor and a consistency verification index function. Based on the semantic behavior tensor model of the user's DID index, it performs structural-level comparison and consistency measurement on the label mapping results formed by users on multiple platforms. This avoids the problem of distortion of platform behavior characteristics by traditional weighted fusion methods and effectively improves the robustness and interpretability of behavior consistency calculations. This method does not rely on a weight model, but instead accurately reflects semantic structure deviations through a difference tensor function, and combines a consistency scoring function to achieve quantitative analysis of the consistency of multi-source behavior structures, ensuring that the behavior aggregation results are highly consistent and verifiable, providing a basic guarantee for the credibility of subsequent credit factor encapsulation and calling. This method is particularly suitable for scenarios where there are semantic tensor differences in user behavior structures between multiple platforms but unified verification is required, enhancing the rigor and credibility of cross-platform credit modeling. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of 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 paying any creative work.

[0073] Figure 1 This is a structural diagram of a method for constructing a cross-platform real-name DID credit points system proposed by the present invention;

[0074] Figure 2 This is a schematic diagram of the unified structure tensor generation process for the construction method of a cross-platform real-name DID credit points system proposed by the present invention;

[0075] Figure 3 This is a schematic diagram of the multi-platform semantic label behavior slice comparison process of the method for constructing a cross-platform real-name DID credit points system proposed in the present invention. DETAILED DESCRIPTION

[0076] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0077] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. 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.

[0078] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it constitute a separate or selective embodiment that is mutually exclusive with other embodiments.

[0079] Example 1:

[0080] Reference Figures 1 to 3 , which is the first embodiment of the present invention, provides a method for constructing a cross-platform real-name DID credit score system, including the following steps:

[0081] S1: Use the cross-platform user identity authentication mechanism to construct the DID identity code and obtain the user's unique cross-platform identity;

[0082] The purpose of step S1 is to establish a unified user identity to provide a unified reference basis for the collection and modeling of cross-platform credit data. By integrating the user's real-name registration information on multiple platforms, a decentralized identity authentication mechanism is used to generate a user identity code DID. u ;

[0083] The identity code is constructed through the following process:

[0084] Information standardization: Process the original identity data into a unified format to remove platform identifiers and noise attributes;

[0085] Encrypted identity hash mapping: Introducing an irreversible identity summary algorithm based on hash chains;

[0086] Multi-dimensional identity vector construction: Encode different identity fields into structure vectors and perform associated nesting processing;

[0087] Generate DID identity code:

[0088] DID u =H(ID1||ID2||…||ID n ||Salt);

[0089] Among them, ID k Represents the real-name fields collected on different platforms. Salt is the random seed derived from the key.

[0090] The output of this step is the user's unique identifier DID u As the unique index key for subsequent multi-platform behavioral data aggregation and semantic modeling, this identifier is not directly bound to the platform account to ensure privacy isolation.

[0091] S2: Use DID identity codes to perform multi-platform behavior label mapping and unified structure tensor generation modeling to obtain the user semantic behavior tensor model;

[0092] Step S2 is based on the user's unique DID identity code DID generated in step S1. u Through this unique identifier, we can retrieve user-related behavior tag data from multiple independent platforms, map the original platform tags to the unified behavior tag ontology space, and further construct a three-dimensional tensor with structural consistency. It serves as the core basic data structure relied upon by subsequent behavioral consistency analysis and behavioral factor abstraction.

[0093] The unified structure tensor generation model in step S2:

[0094]

[0095] in, The set of behavioral labels for user u on the mth platform, m∈{1,2,...,M};

[0096] Cross-platform tag semantic mapping function, mapping each platform tag to a unified tag ontology set;

[0097] Φ(·, DID u ): Structural tensor generation function, which performs multi-dimensional structural embedding based on a unified tag ontology set and DID user identity;

[0098] The final generated semantic behavior tensor model, where M is the number of access platforms, N is the total number of standardized tag ontologies, and K is the measure / intensity of the tag's behavioral response to the user in the corresponding platform.

[0099] Cross-platform tag semantic mapping function in the formula:

[0100]

[0101] Among them, the mapping relationship is represented by the sparse mapping matrix express:

[0102]

[0103] is the semantic embedding matrix of platform labels, obtained by general embedding techniques;

[0104] The structure tensor generating function in the formula:

[0105]

[0106] in, Embed unified tags under platform p;

[0107] Indicates user DID u The activation vector of the tag behavior under the platform pp, that is, whether it is associated with the tag;

[0108] ⊙ indicates broadcast multiplication by row;

[0109] Finally, the behavior label semantic matrices of M platforms are concatenated to generate a tensor

[0110] The unified structure tensor generation model in step S2 includes the following process when used:

[0111] Platform tag aggregation: based on DID u The primary key indexes the user’s tag set on M platforms, and we get

[0112] Unified tag ontology mapping: through function Align platform native tags to a unified tag ontology set to resolve semantic inconsistencies across multiple platforms;

[0113] Label semantic embedding and tensorization: Using Φ(·, DID u ), embed the unified tag ontology into the user behavior response model on each platform, and construct a tensor based on the three dimensions of platform-tag-response intensity;

[0114] Tensor generation and storage: Outputs consistent structure, unified semantics, and traceable behavioral response tensors For use in step S3.

[0115] Compared with the existing technology that focuses on single platform features or directly splices multi-platform data, this step proposes a unified modeling mechanism of "semantic mapping + multi-dimensional tensor structure", which realizes the standardization of the structural expression of user behavior and solves the problem of label semantic conflicts between heterogeneous platforms. Finally, the user's semantic behavior tensor model is obtained through the model formula Serves as the basic input for subsequent behavioral consistency verification.

[0116] S3: Use user semantic behavior tensors to build a cross-platform consistency verification model and obtain a multi-source behavior consistency score matrix;

[0117] Step S3 unifies the behavior semantic tensor Based on this, we model the performance consistency of each platform, each label dimension, and each behavior type, and construct a multi-source behavior consistency score matrix.

[0118] Step S3 uses a multi-platform semantic label behavior slice comparison model to measure the consistency of behavior across different platforms. The model formula is:

[0119]

[0120] in, Behavior sequence under label dimension m and behavior type k;

[0121] Ω p (·): the label behavior feature encoding function of the p-th platform, used to extract features from the time series;

[0122] Λ(·): cross-platform behavioral feature consistency calculation function, used to evaluate the consistency of multi-platform feature results;

[0123] The final consistency score under label dimension m and behavior type k.

[0124] The platform behavior feature encoding function in step S3 is:

[0125]

[0126] in, Represents the time series of behaviors under the platform;

[0127] The second-order difference of the time series is used to characterize behavioral volatility;

[0128] Γ(·): compression function;

[0129] The consistency measurement function in step S3 is:

[0130]

[0131] Among them, x p : is the behavioral characteristic value on platform p;

[0132] ||·||1: L1 norm, used to measure the difference between features;

[0133] The function output value range is [0, 1], and the closer to 1, the higher the consistency.

[0134] The multi-platform semantic label behavior slice comparison model includes:

[0135] Input unified tensor structure Slice by label dimension m and behavior type k to obtain the behavior time series;

[0136] Using the subfunction Ω p(·) Extract volatility features from time series of different platforms;

[0137] The consistency measurement function Λ(·) is used to summarize the differences in behavioral characteristics of all platforms and obtain the consistency score under the label-behavior type dimension;

[0138] Repeat the above process, traverse M label dimensions and K behavior types, and output the consistency matrix

[0139] The multi-platform semantic label behavior slice comparison model is different from the traditional vector weighting or feature fusion method. Through structural tensor, platform feature extraction and consistency comparison, it avoids the dimension compression problem under the assumption of feature space uniformity, while fully retaining the dynamic change characteristics of label behavior under each platform, and completes cross-platform consistency modeling through hypothetical deduction and slice comparison, with higher semantic explanatory power and personalized behavior analysis capabilities.

[0140] S4: Use the multi-source behavior consistency score matrix to identify outliers and obtain a cross-platform behavior anomaly score vector;

[0141] Step S4 uses the output of step S3 As input, it identifies the user's consistent abnormal behavior under different label dimensions and behavior types, and outputs a cross-platform behavior anomaly score vector summarized by user dimensions. That is, the degree of behavioral consistency abnormality corresponding to each label dimension. The output process includes:

[0142] Consistency score matrix aggregation:

[0143] Aggregate the behavior type dimension k to obtain the consistency average or stability index under each label dimension m:

[0144]

[0145] Anomaly score conversion:

[0146] Combined with the mean μ of all user behavior consistency m and standard deviation σ m , the anomaly score is calculated using the Z-score standardization method:

[0147]

[0148] μm: the mean consistency score of all users under label dimension m;

[0149] σ m : The standard deviation of the consistency scores of all users under the label dimension m;

[0150] ∈: A small constant to prevent division by zero, set to 10-6 .

[0151] The generated anomaly score vector It will serve as the basic input for behavioral confidence evaluation and offset modeling, and provide indicator support for subsequent credit structure calculation and label credibility correction.

[0152] S5: Use the anomaly score vector to construct a credit factor function call matrix to obtain an encapsulated factor interface mapping structure;

[0153] Step S5 is based on the behavior anomaly score vector generated in step S4. As the core input for measuring the credibility of each type of behavior label in the cross-platform dimension. At the same time, combined with the unified semantic label tensor constructed in step S2 Identify the structural mapping relationship of each type of label on each platform;

[0154] Step S5 includes constructing a credit factor function mapping matrix:

[0155]

[0156] in, The semantic tensor strength of the user under the label dimension m, behavior type k, and platform p;

[0157] Behavior abnormality score, the larger the value, the less trustworthy it is;

[0158] α∈[0,1]: Abnormal suppression factor, which controls the degree to which abnormal behavior reduces the credit factor weight;

[0159] φ(·): nonlinear factor mapping function:

[0160] φ(x)=log(1+max(x,0));

[0161] In step S5, a packaged credit interface structure tensor is constructed:

[0162]

[0163] in, The mth type of credit factor interface provided by platform p;

[0164] τ: call trigger threshold, the package call is only made when the factor value exceeds the set value;

[0165] The result structure of the factor interface call corresponding to the user in the mth label dimension and platform p.

[0166] In step S5, constructing the credit factor function mapping matrix and constructing the encapsulated credit interface structure tensor includes the following process when used:

[0167] Input combination: Collection step S2 semantic label tensor Behavioral abnormality scoring as in step S4

[0168] Inhibition Modeling: Constructing a Credit Weight Tensor Adjusted for Behavioral Abnormality

[0169] Nonlinear compression: Perform nonlinear processing on the synthetic tensor to prevent abnormal amplification;

[0170] Interface mapping judgment: Based on the set threshold τ, determine whether the encapsulated API is allowed;

[0171] Output package structure: Construct structured credit factor interface tensor

[0172] The key difference between step S5 and existing mature technologies is that it integrates semantic behavior tensors and anomaly scoring vectors, realizes multi-dimensional modeling of cross-platform user behavior, and introduces dynamic inhibition factors and nonlinear mapping functions to respond to and regulate abnormal behaviors, forming an interpretable and encapsulated credit factor function structure tensor. Different from the traditional API call mechanism based only on scoring or static rules, this method uses the structure tensor to It realizes the unification of controllability and schedulability of interface encapsulation, and has behavioral dimension adjustability, exception suppression capability and encapsulation logic determinism, reflecting an integrated innovation path from behavioral semantic modeling to interface output structure.

[0173] S6: Use the credit factor structure tensor to generate real-name credit points, and obtain a cross-platform unified real-name DID credit score result;

[0174] In step S6, the credit score normalization function and weighted scoring logic are called to process the multi-dimensional credit factors in the tensor, thereby generating a real-name credit score result that can be used in practice.

[0175] The model formula corresponding to the credit score normalization function and weighted scoring logic is as follows:

[0176]

[0177] in, Indicates the output value of the ppth credit factor in the mmth behavioral dimension;

[0178] ω p It is the factor importance weight preset in the system;

[0179] It is the final generated real-name credit score value;

[0180] Step S6 is an extension of step S5. It maps the complex credit factor structure tensor result into a cross-platform credit score value that can be read, verified, and shared by other systems, completing the closed-loop output of the entire DID credit system. It can also be called by future business interfaces and inter-platform mutual trust mechanisms, ensuring the system's operability and versatility in terms of authenticity, privacy protection, and credit sharing.

[0181] Example 2:

[0182] This embodiment also provides a computer device suitable for a method for constructing a cross-platform real-name DID credit points system, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement a method for constructing a cross-platform real-name DID credit points system as proposed in the above embodiment.

[0183] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, it implements a method for constructing a cross-platform real-name DID credit points system as proposed in the above embodiment.

[0184] The computer device may be a terminal, comprising a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0185] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0186] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0187] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0188] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0189] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. 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 method for constructing a cross-platform real-name DID credit score system, characterized in that: The following steps are involved: S1. Use the cross-platform user identity authentication mechanism to construct the DID identity code and obtain the user's unique cross-platform identity; S2. Use the DID identity code to perform multi-platform behavior label mapping and unified structure tensor generation modeling to obtain the user semantic behavior tensor model; S3. Use the user semantic behavior tensor to build a cross-platform consistency verification model and obtain a multi-source behavior consistency score matrix; S4. Use the multi-source behavior consistency score matrix to identify outliers and obtain a cross-platform behavior anomaly score vector; S5. Use the anomaly score vector to construct a credit factor function call matrix to obtain an encapsulated factor interface mapping structure; S6. Use the credit factor structure tensor to generate real-name credit points and obtain a cross-platform unified real-name DID credit point result.

2. The method for constructing a cross-platform real-name DID credit score system according to claim 1, characterized in that: Step S1 generates a user identity code DID by integrating the user's real-name registration information on multiple platforms and using a decentralized identity authentication mechanism. u .

3. The method for constructing a cross-platform real-name DID credit score system according to claim 1, characterized in that: The unified structure tensor generation model in step S2 is: in, The set of behavioral labels for user u on the mth platform, m∈{1, 2, ..., M}; Cross-platform tag semantic mapping function, mapping each platform tag to a unified tag ontology set; Φ(·, DID u ): Structural tensor generation function, which performs multi-dimensional structural embedding based on a unified tag ontology set and DID user identity; The final generated semantic behavior tensor model, where M is the number of access platforms, N is the total number of standardized tag ontologies, and K is the measure / intensity of the tag's behavioral response to the user in the corresponding platform.

4. The method for constructing a cross-platform real-name DID credit score system according to claim 3, characterized in that: The unified structure tensor generation model in step S2 includes the following process when used: Platform tag aggregation: based on DID u The primary key indexes the user’s tag set on M platforms, and we get Unified tag ontology mapping: through function Align platform native tags to a unified tag ontology set to resolve semantic inconsistencies across multiple platforms; Label semantic embedding and tensorization: Using Φ(·, DID u ), embed the unified tag ontology into the user behavior response model on each platform, and construct a tensor based on the three dimensions of platform-tag-response intensity; Tensor generation and storage: Outputs consistent structure, unified semantics, and traceable behavioral response tensors 5. The method for constructing a cross-platform real-name DID credit score system according to claim 1, characterized in that: Step S3 uses a multi-platform semantic label behavior slice comparison model to measure the consistency of behaviors across different platforms. The model formula is: in, Behavior sequence under label dimension m and behavior type k; Ω p (·): the label behavior feature encoding function of the p-th platform, used to extract features from the time series; Λ(·): cross-platform behavioral feature consistency calculation function, used to evaluate the consistency of multi-platform feature results; The final consistency score under label dimension m and behavior type k.

6. A method for constructing a cross-platform real-name DID credit score system according to claim 5, characterized in that: The platform behavior feature encoding function in step S3 is: in, Represents the time series of behaviors under the platform; The second-order difference of the time series is used to characterize behavioral volatility; Γ(·): compression function.

7. The method for constructing a cross-platform real-name DID credit score system according to claim 5, characterized in that: The consistency metric function in step S3 is: Among them, x p : is the behavioral characteristic value on platform p; ||·||1: L1 norm, used to measure the difference between features; Function output value range The closer to 1, the higher the consistency.

8. The method for constructing a cross-platform real-name DID credit score system according to claim 5, characterized in that: The multi-platform semantic label behavior slice comparison model in step S3 includes: Input unified tensor structure Slice by label dimension m and behavior type k to obtain the behavior time series; Using the subfunction Ω p (·) Extract volatility features from time series of different platforms; The consistency measurement function Λ(·) is used to summarize the differences in behavioral characteristics of all platforms and obtain the consistency score under the label-behavior type dimension; Repeat the above process, traverse M label dimensions and K behavior types, and output the consistency matrix 9. The method for constructing a cross-platform real-name DID credit score system according to claim 1, characterized in that: Step S4 outputs a cross-platform behavior anomaly score vector summarized by user dimension The output process includes: Consistency score matrix aggregation: Aggregate the behavior type dimension k to obtain the consistency average or stability index under each label dimension m: Anomaly Score Conversion: Combined with the mean μ of all user behavior consistency m and standard deviation σ m , the anomaly score is calculated using the Z-score standardization method: μ m : The mean consistency score of all users under the label dimension m; o m : The standard deviation of the consistency scores of all users under the label dimension m; ∈: A small constant to prevent division by zero, set to 10 -6 .

10. The method for constructing a cross-platform real-name DID credit score system according to claim 1, characterized in that: In step S5, a credit factor function mapping matrix is constructed: in, The semantic tensor strength of the user under the label dimension m, behavior type k, and platform p; Behavior abnormality score, the larger the value, the less trustworthy it is; α∈[0,1]: Abnormal suppression factor, which controls the degree to which abnormal behavior reduces the credit factor weight; φ(·): nonlinear factor mapping function: φ(x)=log(1+max(x,0)); In step S5, a packaged credit interface structure tensor is constructed: in, The mth type of credit factor interface provided by platform p; T: Call trigger threshold, the package call is only made when the factor value exceeds the set value; The result structure of the factor interface call corresponding to the user in the mth label dimension and platform p.

Images