Data evaluation method and system based on tense constraint

Abstract

The invention belongs to the technical field of data evaluation, and discloses a tense constraint-based data evaluation method and system, and the method comprises the steps: obtaining the perception data of a plurality of target objects; constructing a dynamic object relation graph based on the target object relation data, and determining an object relation evaluation value based on the dynamic object relation graph; analyzing the data of the target object based on a pre-constructed time sequence analysis model to obtain an object demand evaluation value; performing multi-source fusion on the external environment data of the target object to obtain fused data, and evaluating the fused data based on a neural network algorithm and an empowerment algorithm to obtain an external environment evaluation value; and performing weighted fusion on the object relationship evaluation value, the object demand evaluation value and the external environment evaluation value to obtain a comprehensive evaluation score of the perception data, the comprehensive evaluation score of the perception data being used for representing the importance of the perception data. According to the method, data with relatively high value can be mined, and the utilization rate of the data is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of data evaluation, and in particular relates to a data evaluation method and system based on temporal constraints. Background Art

[0002] Data itself is a kind of encoding of the information system. Therefore, from the first principles of data, the importance of data is determined by three parts: the relationship between data and the relationship between data and users, and the relationship between data storage and usage environment. The relationship between data may include data correlation, complementarity and redundancy.

[0003] The essence of the importance of data is the reflection of the utility of resources across periods. Its core characteristics are: it is non-storable and cannot be stored statically like physical assets, and its utility changes dynamically with the application scenario.

[0004] However, the existing data space is unable to dynamically capture the instantaneous value of data, and data with higher importance cannot be fully mined, resulting in inefficient data utilization. Summary of the invention

[0005] The purpose of the present invention is to provide a data evaluation method and system based on temporal constraints to solve the problem that the prior art cannot dynamically capture the instantaneous value of data, and data with higher importance cannot be fully mined, resulting in low data utilization efficiency.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions: In a first aspect, the present invention provides a data evaluation method based on temporal constraints, the method comprising: Acquire perception data of a plurality of target objects, wherein the perception data includes: target object relationship data, target object self data, and target object external environment data; Based on the target object relationship data, a dynamic object relationship graph is constructed, and based on the dynamic object relationship graph, an object relationship evaluation value is determined; Analyze the target object's own data based on the pre-built time series analysis model to obtain the object demand assessment value; Perform multi-source fusion on the external environment data of the target object to obtain fused data, evaluate the fused data based on the neural network algorithm and the weighting algorithm to obtain the external environment evaluation value; The object relationship evaluation value, the object demand evaluation value and the external environment evaluation value are weighted and fused to obtain a comprehensive evaluation score of the perception data, and the comprehensive evaluation score of the perception data is used to characterize the importance of the perception data.

[0007] Preferably, the dynamic object relationship graph includes a plurality of nodes, and determining the object relationship evaluation value based on the dynamic object relationship graph includes: Based on the preset algorithm, calculate the centrality index value of each node in the dynamic object relationship graph; The object relationship evaluation value is determined based on the centrality index value of each node in the dynamic object relationship graph.

[0008] Preferably, the calculation expression of the object relationship evaluation value is: ; In the formula, Evaluate a value for an object relationship, i The first i node, I is the total number of nodes in the dynamic object relationship graph, For the i The centrality index value of the node, e is a natural constant, is the time decay factor, t For the current moment, t i For the i The timestamp when the node's relationship was created.

[0009] Preferably, the target object's own data is analyzed based on a pre-built time series analysis model to obtain the object demand assessment value, including: Extract the behavior of the target object from its own data to obtain several behavior features; Input each behavior feature into the time series analysis model to predict demand and obtain the future demand intensity of the target object; Based on the attention mechanism, calculate the contribution of each behavior feature to the demand; The contribution of each behavior feature to the demand is weighted and integrated to obtain the total contribution score of the target object's behavior; Determine the target object's demand assessment value based on the target object's total behavioral contribution score and the target object's future demand intensity.

[0010] Preferably, the calculation expression of the object demand evaluation value is: ; In the formula, is the object demand evaluation value, k is the kth behavior feature, t is the current moment, K is the total number of behavior features, is the weight of the kth behavior feature, is the contribution of the kth behavior feature to the demand, The total contribution score of the target object’s behavior. is the future demand intensity of the target object, is the coefficient of future demand intensity.

[0011] Preferably, the neural network algorithm is an LSTM algorithm, the weighting algorithm is an entropy weight method, the fused data includes several environmental factors, and the fused data is evaluated based on the neural network algorithm and the weighting algorithm to obtain an external environment evaluation value, including: Based on the LSTM algorithm, the long-term dependency of each environmental factor in the fusion data is captured to obtain the environmental relationship status score; The contribution weight of each environmental factor in the fusion data is calculated based on the entropy weight method; The contribution weights of each environmental factor in the fusion data are weighted to obtain the total environmental contribution score; Determine the external environment assessment value based on the total environmental contribution score and the environmental relationship status score.

[0012] Preferably, the calculation expression of the external environment evaluation value is: ; In the formula, is the external environment evaluation value, t is the current moment, J is the total number of environmental factors, is the contribution weight of the jth environmental factor, is the jth environmental factor, is the sensitivity coefficient, is the environmental relationship status score, Total contribution to the environment.

[0013] Preferably, the function expression of the comprehensive evaluation score of the perception data is: ; In the formula, is the comprehensive evaluation score of the perception data, is the relationship weight, is the demand weight, is the environmental weight, t is the current moment, is the initial score, is the object requirement evaluation value, Evaluate a value for an object relationship, It is the external environment assessment value.

[0014] Preferably, the method further comprises: generating an initial score, comprising: Based on the Bayesian-Nash equilibrium network, an inter-period game model is constructed; Predict the trend of perception data scores based on the inter-period game model; Construct an initial score function based on the score change trend of the perception data; Based on the multi-agent deep reinforcement learning algorithm, the initial score function is simulated and learned to obtain the optimal score, which is used as the initial score.

[0015] Preferably, after obtaining the comprehensive evaluation score of the perception data, the method further comprises: performing a credibility evaluation on the perception data of the target object, including: Verify the authenticity of the sensed data based on dynamic zero-knowledge proof driven by quantum random numbers; The authenticity of the perceived data is quantified based on a pre-constructed three-dimensional credibility assessment matrix to obtain credibility.

[0016] In a second aspect, the present invention provides a data evaluation system based on temporal constraints, which is used to implement the above-mentioned data evaluation method based on temporal constraints, and the system comprises: A data acquisition module is used to acquire the perception data of a plurality of target objects, wherein the perception data includes: target object relationship data, target object's own data and target object's external environment data; A relationship evaluation module is used to construct a dynamic object relationship graph based on the target object relationship data, and determine the object relationship evaluation value based on the dynamic object relationship graph; The demand assessment module is used to analyze the target object's own data based on the pre-built time series analysis model to obtain the object demand assessment value; The environmental assessment module is used to perform multi-source fusion of the external environmental data of the target object to obtain fused data, and evaluate the fused data based on the neural network algorithm and the weighting algorithm to obtain the external environment assessment value; The comprehensive evaluation module is used to perform weighted fusion on the object relationship evaluation value, the object demand evaluation value and the external environment evaluation value to obtain a comprehensive evaluation score of the perception data, and the comprehensive evaluation score of the perception data is used to characterize the importance of the perception data.

[0017] Beneficial effects: The present invention evaluates and analyzes the perception data through three dimensions, namely object relationship, external environment and own needs, and can accurately analyze the perception data with higher importance, facilitate the mining of data with higher value, and improve the utilization rate of data. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings: Figure 1 is a flow chart of a data evaluation method based on temporal constraints provided by an embodiment of the present invention; Figure 2 It is a block diagram of a data evaluation system based on temporal constraints provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in combination with the drawings and the description of the embodiments or the prior art. Obviously, the following description of the structure of the drawings is only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention.

[0020] Embodiment 1 Figure 1 FIG. 1 is a flow chart of a data evaluation method based on temporal constraints provided by an embodiment of the present invention. Figure 1 As shown, this embodiment provides a data evaluation method based on temporal constraints, the method comprising: Step S10: Acquire the perception data of several target objects, the perception data including: target object relationship data, target object's own data and target object's external environment data; in this embodiment, the target object may be a user, enterprise or other object. For the user, the user's perception data mainly includes: the user's basic information data (for example, name, age, interests and hobbies), interaction data (for example, browsing history, shopping history, click history, etc.), external environment data (for example, seasonal data, holiday status, etc.); and for the enterprise, taking the enterprise's industrial Internet platform as an example, the perception data mainly refers to the real-time collection of equipment operation data, compliance parameters, equipment relationships, equipment environment and other data by the sensor network.

[0021] Step S20: construct a dynamic object relationship graph based on the target object relationship data, and determine the object relationship evaluation value based on the dynamic object relationship graph.

[0022] In this embodiment, the dynamic object relationship graph includes a number of nodes and edges between the nodes. If there is an edge between any two nodes, it indicates that there is a relationship between the two nodes. Therefore, based on the dynamic object relationship graph, determining the object relationship evaluation value includes: Step S201: Based on a preset algorithm, calculate the centrality index value of each node in the dynamic object relationship graph; wherein the preset algorithm may be an improved PageRank algorithm, which is an algorithm for calculating the importance of web pages. It determines the importance of web pages based on the link relationship between web pages. The basic idea of ​​the algorithm is that the more times a web page is linked by other web pages, the more important the web page is.

[0023] Step S202: Determine the object relationship evaluation value based on the centrality index value of each node in the dynamic object relationship graph.

[0024] Specifically, the calculation expression of the object relationship evaluation value is: (1); In formula (1), Evaluate a value for an object relationship, i The first i node, I is the total number of nodes in the dynamic object relationship graph, For the i The centrality index value of the node, e is a natural constant, is the time decay factor, t For the current moment, t i For the i The timestamp when the node's relationship was created.

[0025] In this embodiment, when the dynamic object relationship graph is updated according to real-time data, a streaming data processing framework (such as Apache Kafka + Flink, where Apache Kafka is a distributed stream processing platform and Apache Flink is an open source stream processing framework for processing unbounded and bounded data streams) can be used to capture the interaction events of real-time data in real time to update the dynamic object relationship graph.

[0026] Step S30: Analyze the target object's own data based on the pre-built time series analysis model to obtain the object demand assessment value.

[0027] Specifically, the target object's own data is analyzed based on the pre-built time series analysis model to obtain the object demand assessment value, including: Step S301: extract behaviors from the target object's own data to obtain several behavioral features. This embodiment takes the user as the target object as an example. The user's own data contains several behaviors, such as query, download, subscription, etc. These behaviors can be extracted using data analysis tools (such as SQL, Python, R, etc.), or machine learning algorithms can be used to extract them, such as decision trees, random forests, support vector machines, and Bayesian algorithms.

[0028] Step S302: Input each behavior feature into the time series analysis model to perform demand forecasting to obtain the future demand intensity of the target object; the time series analysis model of this embodiment adopts the Transformer-XL time series model, and when the target object is a user, the user's behavior features (ie, query, download, and subscription) are analyzed.

[0029] Step S303: Based on the attention mechanism, calculate the contribution of each behavior feature to the demand; this embodiment introduces the attention mechanism to distinguish the contribution of different behaviors to the demand; for example, the weight of recent click behavior is higher than that of historical browsing records.

[0030] Step S304: weighted fusion is performed on the contribution of each behavior feature to the demand to obtain the total behavior contribution score of the target object.

[0031] Step S305: Determine the object demand assessment value based on the total contribution score of the target object's behavior and the future demand intensity of the target object.

[0032] Specifically, the calculation expression of the object demand evaluation value is: (2); In formula (2), is the object demand evaluation value, k is the kth behavior feature, t is the current moment, K is the total number of behavior features, is the weight of the kth behavior feature, is the contribution of the kth behavior feature to the demand, The total contribution score of the target object’s behavior. is the future demand intensity of the target object, is the coefficient of future demand intensity.

[0033] Step S40: Perform multi-source fusion on the external environment data of the target object to obtain fused data, evaluate the fused data based on the neural network algorithm and the weighting algorithm to obtain an external environment evaluation value.

[0034] In this embodiment, an enterprise is used as an example. The external environment data includes IoT sensor data (such as device temperature and humidity), market index (such as stock volatility), and policy text (such as compliance requirements). These data are then fused to obtain fused data. The vector representation of the fused data is: .

[0035] In this embodiment, the neural network algorithm is an LSTM (Long Short-Term Memory Network) algorithm, and the weighting algorithm is an entropy weight method. Therefore, the fusion data is evaluated based on the neural network algorithm and the weighting algorithm to obtain the external environment evaluation value, including: Step S401: Capture the long-term dependency of each environmental factor in the fused data based on the LSTM algorithm to obtain the environmental relationship status score.

[0036] Step S402: Calculate the contribution weight of each environmental factor in the fusion data based on the entropy weight method.

[0037] Step S403: Perform weighted processing on the contribution weights of each environmental factor in the fused data to obtain a total environmental contribution score.

[0038] Step S404: Determine the external environment evaluation value based on the total environmental contribution score and the environmental relationship status score.

[0039] Specifically, the calculation expression of the external environment evaluation value is: (3); In formula (3), is the external environment evaluation value, t is the current moment, J is the total number of environmental factors, is the contribution weight of the jth environmental factor, is the jth environmental factor, is the sensitivity coefficient, is the environmental relationship status score, Total contribution to the environment.

[0040] In this embodiment, the sensitivity coefficient may be dynamically adjusted using the ESARCA algorithm (Environmental Sensitivity Adaptive Robust Control Algorithm).

[0041] Step S50: weighted fusion of the object relationship evaluation value, the object demand evaluation value and the external environment evaluation value to obtain a comprehensive evaluation score of the perception data, where the comprehensive evaluation score of the perception data is used to characterize the importance of the perception data.

[0042] Specifically, the function expression of the comprehensive evaluation score of the perception data is: (4); In formula (4), is the comprehensive evaluation score of the perception data, is the relationship weight, is the demand weight, is the environmental weight, t is the current moment, is the initial score, is the object requirement evaluation value, Evaluate a value for an object relationship, It is the external environment assessment value.

[0043] Therefore, this embodiment evaluates and analyzes the perception data through three dimensions, namely object relationship, external environment and own needs, to obtain the weighted sum of the object demand evaluation value, the object relationship evaluation value and the external environment evaluation value, and uses the weighted result as the correction coefficient of the initial score to correct the initial score to obtain the final comprehensive evaluation score; the present invention can accurately analyze perception data with higher importance, facilitate the mining of data with higher value, and improve data utilization.

[0044] As a further optimization of this embodiment, the method further includes: generating an initial score, including: Step a10: Based on the Bayesian-Nash equilibrium network, construct an inter-period game model; the Bayesian-Nash equilibrium network is an extension of the concept of "Bayesian Nash equilibrium" in game theory in a network environment. It is mainly used to describe static games with incomplete information, where participants in the network structure choose strategies that maximize their expected returns based on the strategy choices of other participants and the probability distribution of their types.

[0045] Step a20: Predict the score change trend of the perception data based on the inter-period game model.

[0046] Step a30: constructing an initial score function based on the score change trend of the perception data; Step a40: Based on the multi-agent deep reinforcement learning algorithm, simulate learning of the initial score function to obtain the optimal score, and use the optimal score as the initial score.

[0047] In this embodiment, the expression of the initial score function is: (5); In formula (5), is the historical attenuation factor, is the sensitivity coefficient, To perceive the trend of the score change of the data, It is a multi-agent deep reinforcement learning algorithm.

[0048] In the process of simulating the learning of the initial score function, the cross-period decision consistency is evaluated in the process of game simulation using the multi-agent deep reinforcement learning algorithm. If it is inconsistent, the historical decay factor is adjusted until it is consistent. After the cross-period decision is consistent, the optimal score is calculated using formula (5).

[0049] As a further optimization of this embodiment, after obtaining the comprehensive evaluation score of the perception data, the method further includes: performing a credibility evaluation on the perception data of the target object. By performing a credibility evaluation on the perception data of the target object, the credibility of the perception data can be improved. The specific steps of the credibility evaluation include: Step b10: Verify the authenticity of the sensed data based on dynamic zero-knowledge proof driven by quantum random numbers.

[0050] Specifically, unpredictable true random numbers are generated based on quantum physics phenomena (such as photon polarization and vacuum fluctuations) to ensure the security of the source of randomness. The steps of dynamic zero-knowledge proof are: Initialization: The prover and the verifier share the quantum random number seed (through a secure channel); Proof generation: The prover uses QRNG to generate random parameters, construct zero-knowledge proofs (such as generating proof keys for zk-SNARKs), and embed dynamic parameters (such as timestamps and environment hashes) into the proof process; Verification execution: The verifier reconstructs the challenge based on the same quantum random number to verify the validity of the proof; if the verification passes, the authenticity of the data is confirmed; otherwise, an alarm is triggered.

[0051] In this embodiment, an anti-attack strategy is deployed during the verification process, for example: Anti-replay attack: embed timestamps and quantum random numbers in the challenge to ensure that each session is unique; Quantum-resistant computing: Use NIST post-quantum standard algorithms (such as CRYSTALS-Kyber) to encrypt communication channels.

[0052] Step b20: quantify the authenticity of the perception data based on the pre-constructed three-dimensional credibility evaluation matrix to obtain credibility.

[0053] Specifically, a three-dimensional credibility evaluation index system is first constructed, as shown in the following table:

[0054] All evaluation indicators are normalized and mapped to the interval [0,1]. The final calculation expression of credibility is: (6); In formula (6), For credibility, Score the credibility of the data source, Score the security of the transmission path, Score the reliability of storage nodes, is the weight corresponding to the credibility score of the data source, is the weight corresponding to the transmission path security score, The weight corresponding to the reliability score of the storage node. Take the calculation of the transmission path security score as an example: (7).

[0055] Therefore, the present invention performs credibility assessment on the perception data of the target object to determine the credibility of the perception data, thereby facilitating the mining of authentic and high-value data and further improving the utilization rate of the data.

[0056] Embodiment 2 Figure 2 FIG. 1 is a block diagram of a data evaluation system based on temporal constraints provided by an embodiment of the present invention. Figure 2 As shown, this embodiment provides a data evaluation system based on temporal constraints, which is used to implement the data evaluation method based on temporal constraints in embodiment 1. The system includes: A data acquisition module is used to acquire the perception data of a plurality of target objects, wherein the perception data includes: target object relationship data, target object's own data and target object's external environment data; A relationship evaluation module is used to construct a dynamic object relationship graph based on the target object relationship data, and determine the object relationship evaluation value based on the dynamic object relationship graph; The demand assessment module is used to analyze the target object's own data based on the pre-built time series analysis model to obtain the object demand assessment value; The environmental assessment module is used to perform multi-source fusion of the external environmental data of the target object to obtain fused data, and evaluate the fused data based on the neural network algorithm and the weighting algorithm to obtain the external environment assessment value; The comprehensive evaluation module is used to perform weighted fusion on the object relationship evaluation value, the object demand evaluation value and the external environment evaluation value to obtain a comprehensive evaluation score of the perception data, and the comprehensive evaluation score of the perception data is used to characterize the importance of the perception data.

[0057] This embodiment also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the data evaluation method based on temporal constraints in the first embodiment when executing the computer program.

[0058] This embodiment further provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the data evaluation method based on temporal constraints in the first embodiment is implemented.

[0059] The present invention evaluates and analyzes the perception data through three dimensions, namely object relationship, external environment and own needs, and can accurately analyze the perception data with higher importance, facilitate the mining of data with higher value, and improve the utilization rate of data.

[0060] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0061] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A system that specifies the functions of a box or multiple boxes.

[0062] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A data evaluation method based on temporal constraints, characterized in that: The method comprises: Acquire perception data of a plurality of target objects, wherein the perception data includes: target object relationship data, target object self data, and target object external environment data; Based on the target object relationship data, a dynamic object relationship graph is constructed, and based on the dynamic object relationship graph, an object relationship evaluation value is determined; Analyze the target object's own data based on the pre-built time series analysis model to obtain the object demand assessment value; Perform multi-source fusion on the external environment data of the target object to obtain fused data, evaluate the fused data based on the neural network algorithm and the weighting algorithm to obtain the external environment evaluation value; The object relationship evaluation value, the object demand evaluation value and the external environment evaluation value are weighted and fused to obtain a comprehensive evaluation score of the perception data, and the comprehensive evaluation score of the perception data is used to characterize the importance of the perception data.

2. The data evaluation method based on temporal constraints according to claim 1 is characterized in that: The dynamic object relationship graph includes a plurality of nodes. Based on the dynamic object relationship graph, an object relationship evaluation value is determined, including: Based on the preset algorithm, calculate the centrality index value of each node in the dynamic object relationship graph; The object relationship evaluation value is determined based on the centrality index value of each node in the dynamic object relationship graph.

3. The data evaluation method based on temporal constraints according to claim 2, characterized in that: The calculation expression of the object relationship evaluation value is: ； In the formula, Evaluate a value for an object relationship, i The first i node, I is the total number of nodes in the dynamic object relationship graph, For the i The centrality index value of the node, e is a natural constant, is the time decay factor, t For the current moment, t i For the i The timestamp when the node's relationship was created.

4. The data evaluation method based on temporal constraints according to claim 1, characterized in that: Analyze the target object's own data based on the pre-built time series analysis model to obtain the object demand assessment value, including: Extract the behavior of the target object from its own data and obtain several behavior features; Input each behavior feature into the time series analysis model to predict demand and obtain the future demand intensity of the target object; Based on the attention mechanism, calculate the contribution of each behavior feature to the demand; The contribution of each behavior feature to the demand is weighted and integrated to obtain the total contribution score of the target object's behavior; Determine the target object's demand assessment value based on the target object's total behavioral contribution score and the target object's future demand intensity.

5. The data evaluation method based on temporal constraints according to claim 4 is characterized in that: The calculation expression of the object demand evaluation value is: ； In the formula, is the object demand evaluation value, k is the kth behavior feature, t is the current moment, K is the total number of behavior features, is the weight of the kth behavior feature, is the contribution of the kth behavior feature to the demand, The total contribution score of the target object’s behavior. is the future demand intensity of the target object, is the coefficient of future demand intensity.

6. The data evaluation method based on temporal constraints according to claim 5, characterized in that: The neural network algorithm is an LSTM algorithm, the weighting algorithm is an entropy weight method, the fused data includes several environmental factors, and the fused data is evaluated based on the neural network algorithm and the weighting algorithm to obtain an external environment evaluation value, including: Based on the LSTM algorithm, the long-term dependency of each environmental factor in the fusion data is captured to obtain the environmental relationship status score; The contribution weight of each environmental factor in the fusion data is calculated based on the entropy weight method; The contribution weights of each environmental factor in the fusion data are weighted to obtain the total environmental contribution score; Determine the external environment assessment value based on the total environmental contribution score and the environmental relationship status score; The calculation expression of the external environment evaluation value is: ； In the formula, is the external environment evaluation value, t is the current moment, J is the total number of environmental factors, is the contribution weight of the jth environmental factor, is the jth environmental factor, is the sensitivity coefficient, is the environmental relationship status score, Total contribution to the environment.

7. The data evaluation method based on temporal constraints according to any one of claims 1 to 6, characterized in that: The function expression of the comprehensive evaluation score of the perception data is: ； In the formula, is the comprehensive evaluation score of the perception data, is the relationship weight, is the demand weight, is the environmental weight, t is the current moment, is the initial score, is the object requirement evaluation value, Evaluate a value for an object relationship, It is the external environment assessment value.

8. The data evaluation method based on temporal constraints according to claim 7, characterized in that: The method further includes: generating an initial score, including: Based on the Bayesian-Nash equilibrium network, an inter-period game model is constructed; Predict the trend of perception data scores based on the inter-period game model; Construct an initial score function based on the score change trend of the perception data; Based on the multi-agent deep reinforcement learning algorithm, the initial score function is simulated and learned to obtain the optimal score, which is used as the initial score.

9. The data evaluation method based on temporal constraints according to claim 7, characterized in that: After obtaining the comprehensive evaluation score of the perception data, the method further includes: performing a credibility evaluation on the perception data of the target object, including: Verify the authenticity of the sensed data based on dynamic zero-knowledge proof driven by quantum random numbers; The authenticity of the perceived data is quantified based on a pre-constructed three-dimensional credibility assessment matrix to obtain credibility.

10. A data evaluation system based on temporal constraints, the system being used to implement the data evaluation method based on temporal constraints according to any one of claims 1 to 9, characterized in that: The system comprises: A data acquisition module is used to acquire the perception data of a plurality of target objects, wherein the perception data includes: target object relationship data, target object's own data and target object's external environment data; A relationship evaluation module is used to construct a dynamic object relationship graph based on the target object relationship data, and determine the object relationship evaluation value based on the dynamic object relationship graph; The demand assessment module is used to analyze the target object's own data based on the pre-built time series analysis model to obtain the object demand assessment value; The environmental assessment module is used to perform multi-source fusion of the external environmental data of the target object to obtain fused data, and evaluate the fused data based on the neural network algorithm and the weighting algorithm to obtain the external environment assessment value; The comprehensive evaluation module is used to perform weighted fusion on the object relationship evaluation value, the object demand evaluation value and the external environment evaluation value to obtain a comprehensive evaluation score of the perception data, and the comprehensive evaluation score of the perception data is used to characterize the importance of the perception data.

Images