Personal security reputation evaluation system

Abstract

The invention discloses a personal safety reputation evaluation system, and the system comprises a data collection module which is used for collecting personal behavior data, consumption data and credit data from different platforms; the data preprocessing module is used for preprocessing the data; the data analysis module is used for performing association analysis and quantification on the personal behavior data, the consumption data and the credit data through a graph neural network model to obtain an analysis result; the reputation calculation module is used for carrying out weighted fusion on the analysis result through an attention mechanism, generating a personal safety reputation score, and dividing personal reputation grades according to the personal safety reputation score; and the reward setting module is used for providing corresponding rewards according to the personal reputation level. By integrating multi-dimensional data, the personal real security reputation level can be comprehensively and dynamically reflected, resource scheduling can be optimized through reinforcement learning, differentiated security services can be provided according to the reputation level, and accurate allocation of security resources is achieved.

Description

Technical Field

[0001] The present invention relates to the field of big data analysis technology, and in particular to a personal security reputation evaluation system. Background Art

[0002] In the digital age, the rapid development of cutting-edge technologies such as big data and artificial intelligence has profoundly transformed social governance and business service models. Massive amounts of behavioral data on individuals in social activities are being collected and digitized, providing unprecedented technical possibilities for accurately profiling individuals and assessing comprehensive risks. At the same time, society's demand for refined security management is growing. Whether it's risk prevention and control in the public security sector or differentiated service provision in commercial scenarios, there is an urgent need to establish a scientific, objective, and dynamic personal security reputation evaluation system to optimize resource allocation and accurately manage risks. Against this backdrop, building a personal security reputation evaluation system based on big data analysis technologies has become a key approach to promoting the modernization of social governance and improving the efficiency of commercial services.

[0003] However, existing personal reputation evaluation systems have significant limitations. Traditional evaluation methods primarily rely on data from a single domain, lacking a comprehensive consideration of an individual's multi-dimensional safety behaviors, and thus struggle to fully reflect an individual's true safety awareness and risk level. For example, data closely related to public safety, such as traffic violation records and home safety maintenance behaviors, is not fully incorporated into the evaluation system. Existing technologies lack data processing and analysis capabilities, and are unable to effectively explore potential correlations between multi-source, heterogeneous data. For example, intelligent algorithms cannot reveal the inherent connection between individual consumption behavior and safety risks. Furthermore, existing evaluation systems generally lack dynamic update mechanisms, making it difficult to reflect changing trends in an individual's safety reputation in real time, resulting in evaluation results lagging behind actual behavioral performance. These shortcomings significantly reduce the effectiveness of existing reputation evaluation systems in guiding the allocation of security resources and providing differentiated services, and they are unable to meet the needs of modern society for refined safety management. Summary of the Invention

[0004] 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 and title of this application to avoid obscuring the purpose of this section, the abstract and the title of the invention, and such simplifications or omissions should not be used to limit the scope of the present invention.

[0005] In light of the above-mentioned existing problems, the present invention proposes a personal safety reputation evaluation system to address the problems of traditional grassland grazing livestock management, which relies on manual experience and lacks systematic application of Internet of Things technology, making it difficult to achieve precise grassland resource management, timely disease prevention and control, and reasonable grazing planning.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a personal security reputation evaluation system, comprising:

[0008] Data collection module, used to collect personal behavior data, consumption data and credit data from different platforms;

[0009] A data analysis module is used to perform correlation analysis and quantification on the personal behavior data, consumption data, and credit data through a graph neural network model to obtain analysis results;

[0010] The reputation calculation module is used to weightedly fuse the analysis results through an attention mechanism to generate a personal safety reputation score, and classify personal reputation levels according to the personal safety reputation score.

[0011] As a preferred solution of the personal safety reputation evaluation system of the present invention, it includes:

[0012] A data preprocessing module, used to preprocess the personal behavior data, consumption data and credit data;

[0013] The reward setting module is used to provide corresponding rewards based on personal credit level through the resource scheduling algorithm optimized by reinforcement learning.

[0014] As a preferred solution of the personal safety reputation evaluation system of the present invention, wherein: the personal behavior data includes traffic violation records, dispute handling records, safety training participation records and emergency response behavior data;

[0015] The consumption data includes safety protection equipment purchase records, safety service subscription records, insurance purchase records, and safety product repair and maintenance consumption data;

[0016] The credit data includes credit default records, contract performance, public safety commitment fulfillment records, and safety-related complaint handling results data.

[0017] As a preferred solution of the personal safety and reputation evaluation system of the present invention, the data preprocessing module is used to preprocess the personal behavior data, consumption data and credit data, including:

[0018] The data cleaning unit uses the isolation forest algorithm to identify and remove outliers, uses the K-nearest neighbor algorithm to interpolate missing values, and uses regular expressions to standardize unstructured text data;

[0019] The feature engineering unit converts text data into vector representations using word embedding technology, extracts periodic features of behavioral data using time series analysis, and discovers hidden patterns in consumer data through association rule mining;

[0020] The standardization processing unit performs Z-score normalization on numerical features, one-hot encoding on categorical features, and performs feature dimensionality reduction through principal component analysis to ensure the dimensional consistency and scale uniformity of the input data.

[0021] As a preferred solution of the personal security reputation evaluation system described in the present invention, the graph neural network model includes:

[0022] The behavior relationship layer constructs individual behavior data into a heterogeneous graph, where node types include behavior events, timestamps, and geographic locations, and edge types include causal relationships, temporal relationships, and spatial associations.

[0023] The consumption association layer constructs consumption data into a bipartite graph, where one type of node represents the user and the other type of node represents the security product or service. The edge weights represent the consumption frequency and amount.

[0024] The credit interaction layer constructs credit data into a directed graph, where nodes represent credit entities and edges represent credit relationships. Edge attributes include default probability and fulfillment time.

[0025] The multimodal fusion layer performs weighted fusion of each layer through the graph attention mechanism, extracts cross-domain correlation features, and generates a unified feature representation vector.

[0026] As a preferred solution of the personal safety and reputation evaluation system of the present invention, wherein: the personal behavior data, consumption data and credit data are subjected to correlation analysis and quantification by a graph neural network model to obtain analysis results, including:

[0027] Using a random walk algorithm to identify the heterogeneous graph, a typical sequence pattern of individual safety behavior is obtained;

[0028] Calculate recommendation scores between users and security products and services to obtain potential security needs;

[0029] Based on the feature representation vector, the mutual influence among the three dimensions of behavior, consumption, and credit is analyzed to obtain analysis results.

[0030] As a preferred embodiment of the personal safety reputation evaluation system of the present invention, the analysis results are weighted and integrated through an attention mechanism to generate a personal safety reputation score, and personal reputation levels are divided according to the personal safety reputation score, including:

[0031] Calculate the attention weights of different dimensions, and based on the attention weights, calculate the weighted personal credit score, which is expressed as:

[0032] F=α1h 行为 +α2h 消费 +α3h 信用

[0033] Among them, F is the personal credit score, α1, α2, α3 are the attention weights of different dimensions, and h 行为 、h 消费 、h 信用 Represents the analysis result vectors of the three dimensions of behavior, consumption, and credit respectively;

[0034] According to the personal credit score and classification criteria, a personal credit rating is obtained, and the credit rating is divided into five levels.

[0035] In a second aspect, the present invention provides a method for a personal security reputation evaluation system, comprising:

[0036] Collect personal behavior data, consumption data, and credit data from different platforms;

[0037] Preprocessing the personal behavior data, consumption data, and credit data;

[0038] Perform correlation analysis and quantification on the personal behavior data, consumption data, and credit data through a graph neural network model to obtain analysis results;

[0039] The analysis results are weighted and integrated through an attention mechanism to generate a personal safety reputation score, and the personal reputation level is divided according to the personal safety reputation score;

[0040] According to the individual's credit rating, corresponding rewards are provided through a resource scheduling algorithm optimized by reinforcement learning. In a third aspect, the present invention provides a computing device comprising:

[0041] memory and processor;

[0042] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the personal security reputation evaluation system are implemented.

[0043] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the personal security reputation evaluation system.

[0044] Compared with the existing technology, the beneficial effects of the present invention are as follows: the present invention integrates multi-dimensional data such as personal behavior, consumption, and credit through a multi-source data acquisition module, and uses graph neural networks and attention mechanisms to realize deep correlation analysis and precise quantification of data. It can not only comprehensively and dynamically reflect the individual's true security credibility level, but also effectively make up for the defects of single data and lagging analysis in the traditional evaluation system; it can also optimize resource scheduling through reinforcement learning, provide differentiated security services according to credibility levels, and realize precise allocation of security resources; at the same time, the application of data preprocessing and advanced algorithms has significantly improved data processing efficiency and the reliability of evaluation results, providing a scientific decision-making basis for social security management and commercial service optimization, and promoting the transformation of personal security credibility evaluation from extensive to intelligent and refined. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] 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. Those skilled in the art can also derive other drawings based on these drawings without inventive effort. Among them:

[0046] Figure 1 This is a schematic diagram of a personal security reputation evaluation system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0047] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is given in conjunction with the accompanying drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in this field without creative work should fall within the scope of protection of the present invention.

[0048] 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.

[0049] 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 refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0050] The present invention is described in detail with reference to schematic diagrams. For ease of illustration, cross-sectional views of device structures may be partially enlarged and not to scale when describing embodiments of the present invention. Furthermore, the schematic diagrams are merely illustrative and should not limit the scope of the present invention. Furthermore, in actual production, the three-dimensional dimensions of length, width, and depth should be included.

[0051] Furthermore, in the description of the present invention, it should be noted that the terms "upper, lower, inner, and outer" and other references to orientations or positional relationships are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the systems or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limitations on the present invention. Furthermore, the terms "first, second, or third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0052] In this disclosure, unless otherwise specified or limited, the terms "mounted," "connected," and "connected" should be interpreted broadly. For example, they may refer to fixed, removable, or integral connections. They may also refer to mechanical, electrical, or direct connections, indirect connections through an intermediary, or internal communication between two components. Those skilled in the art will understand the specific meanings of these terms in this disclosure.

[0053] Example 1

[0054] Reference Figure 1 , as one embodiment of the present invention, provides a personal security reputation evaluation system, including:

[0055] Data collection module, used to collect personal behavior data, consumption data and credit data from different platforms;

[0056] The data analysis module is used to perform correlation analysis and quantification on personal behavior data, consumption data, and credit data through a graph neural network model to obtain analysis results;

[0057] The reputation calculation module is used to generate a personal security reputation score by weighted fusion analysis results through the attention mechanism, and to classify personal reputation levels based on the personal security reputation score;

[0058] Data preprocessing module, used to preprocess personal behavior data, consumption data and credit data;

[0059] The reward setting module is used to provide corresponding rewards based on personal credit level through the resource scheduling algorithm optimized by reinforcement learning.

[0060] Furthermore, personal behavior data includes traffic violation records, dispute resolution records, safety training participation records, and emergency response behavior data;

[0061] Consumption data includes safety protection equipment purchase records, safety service subscription records, insurance purchase records, and safety product repair and maintenance consumption data;

[0062] Credit data includes credit default records, contract performance, public safety commitment fulfillment records, and safety-related complaint handling results data.

[0063] Furthermore, the data preprocessing module is used to preprocess personal behavior data, consumption data, and credit data, including:

[0064] The data cleaning unit uses the isolation forest algorithm to identify and remove outliers, uses the K-nearest neighbor algorithm to interpolate missing values, and uses regular expressions to standardize unstructured text data;

[0065] The feature engineering unit converts text data into vector representations using word embedding technology, extracts periodic features of behavioral data using time series analysis, and discovers hidden patterns in consumer data through association rule mining;

[0066] The standardization processing unit performs Z-score normalization on numerical features, one-hot encoding on categorical features, and performs feature dimensionality reduction through principal component analysis to ensure the dimensional consistency and scale uniformity of the input data.

[0067] Furthermore, the graph neural network model includes:

[0068] The behavior relationship layer constructs individual behavior data into a heterogeneous graph, where node types include behavior events, timestamps, and geographic locations, and edge types include causal relationships, temporal relationships, and spatial associations.

[0069] The consumption association layer constructs consumption data into a bipartite graph, where one type of node represents the user and the other type of node represents the security product or service. The edge weights represent the consumption frequency and amount.

[0070] The credit interaction layer constructs credit data into a directed graph, where nodes represent credit entities and edges represent credit relationships. Edge attributes include default probability and fulfillment time.

[0071] The multimodal fusion layer performs weighted fusion of each layer through the graph attention mechanism, extracts cross-domain correlation features, and generates a unified feature representation vector.

[0072] Furthermore, the graph neural network model is used to perform correlation analysis and quantification on personal behavior data, consumption data, and credit data to obtain analysis results, including:

[0073] Use the random walk algorithm to identify heterogeneous graphs and obtain typical sequence patterns of personal safety behaviors;

[0074] Calculate recommendation scores between users and security products and services to obtain potential security needs;

[0075] Based on the feature representation vector, the mutual influence between the three dimensions of behavior, consumption, and credit is analyzed to obtain the analysis results.

[0076] Specifically, the multi-source heterogeneous data is constructed into heterogeneous graph structures, such as behavioral relationship graphs, consumption relationship graphs, and credit interaction graphs. Nodes represent entities, such as users, behavioral events, and products, and edges represent relationships between entities, such as temporal correlations, consumption preferences, and credit lending. A graph convolutional network is then used to extract and propagate features across each graph structure. Dependencies between nodes are learned through a message passing mechanism. For example, a user's interest score for safety products is calculated in a consumption relationship graph. A graph attention mechanism is then used to weightedly fuse graph features from different dimensions, automatically assigning importance weights to each dimension and highlighting features strongly correlated with safety reputation. Finally, a three-dimensional correlation matrix is ​​constructed using tensor decomposition techniques to quantify the mutual influence between behavior, consumption, and credit. For example, this allows analyzing the impact of traffic safety behavior on insurance purchase decisions, thereby generating a feature vector representation that comprehensively reflects an individual's safety reputation. By transforming unstructured relational data, which is difficult for traditional machine learning to process, into a computable graph representation, this approach enables in-depth correlation analysis and precise quantification of multidimensional data.

[0077] Furthermore, the attention mechanism is used to weight the fusion analysis results to generate a personal security reputation score. Based on the personal security reputation score, personal reputation levels are divided, including:

[0078] Calculate the attention weights of different dimensions, and based on the attention weights, calculate the weighted personal credit score, which is expressed as:

[0079] F=α1h 行为 +α2h 消费 +α3h 信用

[0080] Among them, F is the personal credit score, α1, α2, α3 are the attention weights of different dimensions, and h 行为 、h 消费 、h 信用 Represents the analysis result vectors of the three dimensions of behavior, consumption, and credit respectively;

[0081] Based on the personal credit score and the classification criteria, the personal credit level is obtained, and the credit level is divided into five levels.

[0082] Specifically, a multi-layer perceptron network is constructed to calculate the attention weights of the analysis results of each dimension. The input is the feature vectors of the three dimensions of behavior, consumption, and credit. The output is normalized by the Softmax function to obtain the importance coefficient of each dimension. The personal security reputation score is then calculated using a weighted summation formula. This formula realizes the dynamic weighted fusion of data from different dimensions, avoiding the limitations of fixed weights in traditional methods.

[0083] The calculated continuous score is mapped to a preset five-level discrete grade system, and the K-means clustering algorithm is used to determine the optimal division threshold. For example, a score of ≥90 points is mapped to grade A, and a score of 80-90 points is mapped to grade B.

[0084] A dynamic adjustment mechanism has been established. When a user's rating fluctuates by more than 10% over three consecutive months, a re-rating process is triggered. Rules for the accumulation and decay of credit points are also introduced to ensure that ratings reflect changes in individual safety behaviors in real time. The attention mechanism enables adaptive optimization of the rating model, and dynamic ratings improve the timeliness and accuracy of evaluation results.

[0085] Furthermore, the reward setting module is used to provide corresponding rewards based on personal credit level through the resource scheduling algorithm optimized by reinforcement learning, including:

[0086] A Markov decision process model is constructed with reputation level as the state space, security service combination as the action space, and the weighted sum of service cost and user satisfaction as the reward function. The deep Q-network algorithm is used for iterative training to solve the optimal service resource allocation strategy under different reputation levels.

[0087] For Class A users, the system automatically allocates advanced security services, such as free installation of smart home security devices and annual inspections. Class B users enjoy priority registration for safety drills and discounts on emergency rescue services. The system also monitors user behavior in real time and dynamically adjusts reward strategies through reinforcement learning. When a user maintains a high credit rating for three consecutive months, reward services are automatically upgraded. Any security violations trigger a downgrade mechanism and reduce service benefits, ensuring precise and dynamic allocation of security resources.

[0088] It should be noted that the present invention integrates multi-dimensional data such as personal behavior, consumption, and credit through a multi-source data acquisition module, and uses graph neural networks and attention mechanisms to achieve deep correlation analysis and precise quantification of data. It can not only comprehensively and dynamically reflect the individual's true security credibility level, effectively make up for the defects of single data and delayed analysis in the traditional evaluation system; it can also optimize resource scheduling through reinforcement learning, provide differentiated security services according to credibility levels, and achieve precise allocation of security resources; at the same time, the application of data preprocessing and advanced algorithms has significantly improved data processing efficiency and the reliability of evaluation results, providing a scientific decision-making basis for social security management and commercial service optimization, and promoting the transformation of personal security credibility evaluation from extensive to intelligent and refined.

[0089] The personal security reputation evaluation method of big data analysis in this embodiment includes:

[0090] Collect personal behavior data, consumption data, and credit data from different platforms;

[0091] Pre-process personal behavior data, consumption data, and credit data;

[0092] Use graph neural network models to perform correlation analysis and quantification on personal behavior data, consumption data, and credit data to obtain analysis results;

[0093] The weighted fusion analysis results are combined through the attention mechanism to generate a personal security reputation score, and the personal reputation level is divided according to the personal security reputation score;

[0094] Based on the individual's credit level, corresponding rewards are provided through the resource scheduling algorithm optimized by reinforcement learning.

[0095] This embodiment further provides a computing device applicable to a personal security reputation evaluation system, including:

[0096] Memory and processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the personal security reputation evaluation system proposed in the above embodiment.

[0097] This embodiment further provides a storage medium on which a computer program is stored. When the program is executed by a processor, the personal security reputation evaluation system proposed in the above embodiment is implemented.

[0098] The storage medium proposed in this embodiment and the personal security reputation evaluation system proposed in the above embodiment belong to the same inventive concept. For technical details not fully described in this embodiment, please refer to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0099] Through the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented with the help of software and necessary general-purpose hardware, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as a computer's floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.

[0100] 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 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. Personal safety reputation evaluation system, characterized by: include: Data collection module, used to collect personal behavior data, consumption data and credit data from different platforms; A data analysis module is used to perform correlation analysis and quantification on the personal behavior data, consumption data, and credit data through a graph neural network model to obtain analysis results; The reputation calculation module is used to weightedly fuse the analysis results through an attention mechanism to generate a personal safety reputation score, and classify personal reputation levels according to the personal safety reputation score.

2. The personal safety reputation evaluation system according to claim 1, characterized in that: Also includes: A data preprocessing module, used to preprocess the personal behavior data, consumption data and credit data; The reward setting module is used to provide corresponding rewards based on personal credit level through the resource scheduling algorithm optimized by reinforcement learning.

3. The personal safety reputation evaluation system according to claim 1 or 2, characterized in that: The personal behavior data includes traffic violation records, dispute resolution records, safety training participation records, and emergency response behavior data; The consumption data includes safety protection equipment purchase records, safety service subscription records, insurance purchase records, and safety product repair and maintenance consumption data; The credit data includes credit default records, contract performance, public safety commitment fulfillment records, and safety-related complaint handling results data.

4. The personal safety reputation evaluation system according to claim 3, characterized in that: The data preprocessing module is used to preprocess the personal behavior data, consumption data and credit data, including: The data cleaning unit uses the isolation forest algorithm to identify and remove outliers, uses the K-nearest neighbor algorithm to interpolate missing values, and uses regular expressions to standardize unstructured text data; The feature engineering unit converts text data into vector representations using word embedding technology, extracts periodic features of behavioral data using time series analysis, and discovers hidden patterns in consumer data through association rule mining; The standardization processing unit performs Z-score normalization on numerical features, one-hot encoding on categorical features, and performs feature dimensionality reduction through principal component analysis to ensure the dimensional consistency and scale uniformity of the input data.

5. The personal safety reputation evaluation system according to claim 4, characterized in that: Graph neural network models include: The behavior relationship layer constructs individual behavior data into a heterogeneous graph, where node types include behavior events, timestamps, and geographic locations, and edge types include causal relationships, temporal relationships, and spatial associations. The consumption association layer constructs consumption data into a bipartite graph, where one type of node represents the user and the other type of node represents the security product or service. The edge weights represent the consumption frequency and amount. The credit interaction layer constructs credit data into a directed graph, where nodes represent credit entities and edges represent credit relationships. Edge attributes include default probability and fulfillment time. The multimodal fusion layer performs weighted fusion of each layer through the graph attention mechanism, extracts cross-domain correlation features, and generates a unified feature representation vector.

6. The personal safety reputation evaluation system according to claim 5, characterized in that: The personal behavior data, consumption data, and credit data are analyzed and quantified through a graph neural network model to obtain analysis results, including: Using a random walk algorithm to identify the heterogeneous graph, a typical sequence pattern of individual safety behavior is obtained; Calculate recommendation scores between users and security products and services to obtain potential security needs; Based on the feature representation vector, the mutual influence among the three dimensions of behavior, consumption, and credit is analyzed to obtain analysis results.

7. The personal safety reputation evaluation system according to claim 6, characterized in that: The analysis results are weighted and fused through an attention mechanism to generate a personal safety reputation score. Based on the personal safety reputation score, personal reputation levels are divided, including: Calculate the attention weights of different dimensions, and based on the attention weights, calculate the weighted personal credit score, which is expressed as: F=α1h 行为 +α2h 消费 +α3h 信用 Among them, F is the personal credit score, α1, α2, α3 are the attention weights of different dimensions, and h 行为 、h 消费 、h 信用 Represents the analysis result vectors of the three dimensions of behavior, consumption, and credit respectively; According to the personal credit score and classification criteria, a personal credit rating is obtained, and the credit rating is divided into five levels.

8. An electronic device comprising: memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the personal security reputation evaluation system according to any one of claims 1 to 7 are implemented.

9. A computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions, when executed by a processor, implement the steps of the personal security reputation evaluation system according to any one of claims 1 to 7.

Images