140 results about "Belief propagation" patented technology

The invention discloses a construction site digital-to-analog conversion and collaborative management method based on a distributed computing architecture. The method comprises the following steps: S1, collecting microscopic level data of a construction site; s2, obtaining mid-level data of the construction site; s3, acquiring macroscopic hierarchical data, and constructing a multi-scale spatio-temporal data set; s4, decomposing and processing the multi-scale spatio-temporal data set by using a spatio-temporal diagram convolutional network, and constructing a dynamic digital twin model; s5, designing a competition and cooperation mechanism between agents by using a game theory model, solving resource allocation conflicts through Nash equilibrium, and dynamically optimizing resource scheduling; s6, optimizing a scheduling strategy of the intelligent agent through a distributed algorithm based on near-end strategy optimization; s7, detecting errors in digital-to-analog conversion and scheduling processes by using a belief propagation algorithm; and S8, optimizing a scheduling strategy through a distributed adaptive feedback control mechanism. According to the invention, a distributed computing architecture, space-time diagram convolution, an intelligent scheduling technology and the like are adopted, and digital-to-analog conversion and resource management of the construction site are optimized.

The invention provides an intelligent memory leak prediction and tracking method and system for a micro-service architecture, and relates to the technical field of memory leak prediction, and the method comprises the steps: collecting the runtime data of a service node, constructing a service calling topological graph, extracting the time sequence and topological features, and carrying out the feature fusion. And constructing a deep learning prediction model by combining a bidirectional long-short term memory network with a multi-head attention mechanism and a graph attention network, and predicting a memory use trend and a risk score of a service node. Potential memory leak nodes are identified according to the risk score and the abnormal degree predicted value, a key propagation path is identified through a service calling topological graph and a minimum cut algorithm, and finally a root service of memory leak is positioned by using a belief propagation algorithm, so that accurate prediction and rapid positioning of the memory leak in the micro-service system are realized. And the system stability and reliability are improved.

The invention relates to a tunnel lining leakage infrared-millimeter wave fusion detection method and system, and belongs to the technical field of civil engineering detection, and the method comprises the steps: obtaining high-density millimeter wave radar data and thermal infrared imager data; performing data preprocessing to realize space alignment and time synchronization; topological features are extracted, manifold learning mapping is carried out on the preprocessed millimeter wave radar data, and a topological signature matrix is constructed; multi-phase spectrum analysis is carried out, nonlinear phase space reconstruction and multi-scale entropy feature analysis are carried out on the preprocessed infrared thermal imager data, and an entropy feature map is obtained; probability fusion and progressive reasoning are executed, and a leakage area probability graph is generated through a conditional random field model and a progressive belief propagation algorithm; the method overcomes the limitation of a single sensor detection method, achieves the high-precision detection and quantitative evaluation of the leakage of the tunnel lining, and has the technical effects of high detection accuracy, high environmental adaptability, high automation degree and the like.

The invention discloses a high-performance low-complexity LDPC (Low Density Parity Check) decoding method for satellite communication, and belongs to the field of digital signal processing. And performing iterative decoding on the received channel information based on a low density parity check (LDPC) code in the 5G standard. The implementation method comprises the following steps of: dynamically setting an erasing range by utilizing the amplitude of current check node information when a variable node is updated; the amplitude of check node information is gradually increased along with the increase of the number of iterations, the whole iteration process is divided into two parts, and the erasing range is set by adopting different parameters. And outputting the decoding result when the decoding result is correct or reaches the maximum number of iterations. According to the method, the reliability of variable node information is improved through the two points, reliable variable information is prevented from being erased, unreliable variable node information is prevented from being spread, and excellent decoding performance is achieved. Non-linear operation in a log-likelihood ratio belief propagation algorithm LLR-BPA does not exist in the decoding algorithm, and the decoding algorithm has the advantages of being high in decoding performance and low in decoding complexity.

The invention discloses a power distribution network topology state estimation method, electronic equipment, a medium and a product. The method comprises the steps of obtaining a topological structure and measurement data of a power system; constructing variable nodes and factor nodes according to the topological structure and the measurement data, and constructing a state-topological joint factor graph model according to the variable nodes, the factor nodes and the topological structure; based on the state-topology joint factor graph model, performing state estimation through a belief propagation algorithm to obtain an estimated value of state variable correction; if the estimated value of the state variable correction meets the convergence condition, correcting the topological state of the switch branch according to the active power and reactive power of the head end of the switch branch in the estimated value of the state variable correction; and outputting an estimation result of the topological state of the switch branch until the on-off state of the switch branch obtained according to the state variable is consistent with the original topological state of the switch branch. According to the method, asynchronous real-time updating of the topological state of the power system can be realized.

The invention relates to the technical field of meteorological short-term and imminent forecasting, in particular to a meteorological short-term and imminent forecasting method based on multi-source data fusion and real-time modeling, and the meteorological short-term and imminent forecasting method based on multi-source data fusion and real-time modeling performs spatio-temporal interpolation through a time sliding filling method in combination with a DEM-containing trust propagation algorithm. And complementing the missing data and expanding meteorological data of other non-automatic station monitoring areas in the target area at the same time so as to obtain complete high-precision ground three-dimensional meteorological data (10 minutes per 10 minutes, 1 kilometer-level forecast precision) used for representing meteorological data in a three-dimensional space of the target area, and updating the UNet model in real time through an incremental learning strategy so as to realize real-time prediction of the meteorological data in the three-dimensional space of the target area. And secondly, a physical constraint module of an output layer of the UNet model ensures that a forecast result accords with a meteorological dynamics rule, and SIMD instruction optimization enables the fusion efficiency to be improved by 8 times, so that the real-time service requirement is met.

The invention discloses an SVG valve hall cooling efficiency evaluation method and system based on a probabilistic graph model, and the method comprises the steps: obtaining original time sequence data which is obtained through the collection of a cooling system multi-parameter monitoring sensor group disposed in an SVG valve hall in continuous T sampling periods; preprocessing the original time series data to obtain a credible time series data set; constructing a Bayesian network topological structure comprising three-level nodes of an environment layer, a component layer and an efficiency layer and causal dependence edges, and optimizing parameters of the Bayesian network topological structure by adopting a maximum likelihood estimation method to form a dynamic Bayesian network model after parameter calibration; and the credible time sequence data set is used as an evidence variable to be input into the Bayesian network model after parameter calibration, calculation is carried out through a belief propagation reasoning algorithm, a final control instruction set is generated through probability weighted scoring processing, the final control instruction set is fed back to a valve group monitoring system, and early warning and automatic load reduction are achieved. The problems of large evaluation deviation and early warning lag in the prior art are solved.

The invention relates to the technical field of logistics robot navigation, in particular to an intelligent navigation method suitable for a goods shelf channel and a logistics area environment. The method comprises the following steps: acquiring point cloud data acquired by AGV vehicle operation, and after noise suppression of the point cloud data, constructing an accumulator based on Manhattan orthogonal constraint of a plant to carry out orthogonal accumulation weighted statistics on a search neighborhood of the point cloud data so as to determine source landmark point cloud features of an AGV vehicle extraction shelf channel; real-time sensing data of a multi-modal sensor is obtained, a real-time multi-sensor fusion model is constructed based on a hierarchical factor graph and belief propagation calculation, correction registration optimization is performed on source landmark point cloud features by using the real-time multi-sensor fusion model and the optical flow distortion amplitude of dynamic constraint, and positioning navigation is provided for the AGV; according to the invention, powerful environment sensing, positioning and navigation capabilities can be provided for the AGV vehicle, so that efficient warehousing operation and logistics task execution are realized.

The invention discloses a long video content information acquisition method based on an OCR and voice recognition technology. The method comprises the following steps: S1, carrying out preprocessing n on input long video data to extract an image frame sequence and an audio stream; s2, inputting the image frame sequence into an OCR recognition module, inputting the audio stream into an ASR recognition module, and obtaining a preliminary recognition result; s3, constructing a multi-target fitness function, and optimizing an OCR and ASR parameter combination by using a Kanglizard optimization algorithm; s4, respectively applying the optimal parameter group to an OCR identification module and an ASR identification module to obtain an optimized identification result; s5, constructing a fusion factor graph, executing edge message passing by adopting a belief propagation algorithm, and generating a multi-modal semantic block set; and S6, processing the multi-modal semantic block set to generate a unified multi-modal content information set. According to the invention, through fusion of the horny lizard optimization algorithm and the belief propagation mechanism, high-precision recognition and multi-modal semantic consistency extraction of the image text and the voice information in the long video are realized.

The invention provides a belief propagation LDPC decoding method and device based on deep learning, a medium and a program product. The method comprises the following steps: acquiring an LDPC code training sample set corresponding to a check matrix H; according to the check matrix H, determining a weight matrix and an activation function of each layer of a deep learning decoding model; based on the weight matrix and the activation function, using a belief propagation algorithm to establish a deep learning decoding model; training the deep learning decoding model by using the training sample set, and storing the trained deep learning decoding model as a decoder; and sending the LDPC code soft information which is corresponding to the check matrix H and needs to be decoded into the decoder to complete LDPC decoding. According to the method, iterative operation is completed by using deep learning in the error correction decoding process, the number of decoding iterations is reduced, and the sequence sent by the sending end is restored from the sequence containing noise and interference, so that the problem of decoding error codes caused by noise and other interference factors is solved.

The invention discloses a micro-service deployment and task unloading optimization method for tasks with complex dependency relationships. The method comprises the following steps: a system architecture covers a plurality of base stations, edge servers and user equipment; an application program generated by the user equipment comprises a plurality of micro-service tasks with a complex dependency relationship, and the tasks are unloaded to an edge server to be processed; in a task unloading process, optimizing a task unloading decision in each time slot, and managing a micro-service layer cache by adopting a long-time scale updating strategy; through joint optimization of a micro-service layer caching strategy and a task unloading decision, a system performance problem is constructed into an optimization model; firstly, an unloading path of a task is planned by applying a belief propagation algorithm, and then a caching strategy is trained by means of a deep reinforcement learning model; the problem of complicated task unloading and cache updating in different time scale scenes is solved by utilizing an alternative optimization algorithm, and finally, a trained model is deployed in an edge server, so that the long-term average time delay of the system is minimized, and meanwhile, the load balancing of the server is realized. According to the method, the utilization efficiency of system resources can be remarkably improved, the time delay is reduced, and various application requirements sensitive to the delay are fully met.

The invention relates to the technical field of enterprise service digitalization and intelligent management, in particular to an enterprise management method and management platform based on dynamic portraits. The method comprises the following steps: S1, accessing and standardizing multi-source heterogeneous data, calculating a source-level quality score and generating an evidence chain; s2, constructing a portrait map and a time axis taking an enterprise as a main node; s3, the rule engine and the LLM-RAG output numbers in parallel; s4, consistent arbitration and belief propagation; s5, event-driven dependency subgraph increment recalculation is carried out; s6, carrying out portrait-strategy matching and task linkage; s7, carrying out closed-loop write-back and auditable playback; and S8, multi-tenant fine-grained permission. According to the scheme, the problems of data fragmentation and insufficient timeliness and traceability are solved, and the linkage of portrait credibility, interpretability and real-time decision is realized.

The invention provides a backtracking-based spatial coupling low-density parity check code decoding method. The backtracking-based spatial coupling low-density parity check code decoding method comprises the following steps: acquiring a code word sequence of a spatial coupling low-density parity check code to be decoded according to set parameters; performing side information initialization on the decoding window, executing belief propagation decoding in the decoding window, and performing judgment when the number of iterations reaches the set minimum number of iterations; and determining the decoding position and the window size of a decoding window in the next step according to a judgment result until all target symbols in the code word sequence are decoded. According to the invention, the error propagation phenomenon in the space coupling LDPC code sliding window decoding process can be effectively improved, and the sliding window decoding performance is improved.

The invention discloses a defect repeated alarm screening method and system based on a twin network and topology analysis, and relates to the technical field of intelligent substations. The method comprises the steps of extracting image pair features through a twin network and generating matching potential energy; constructing an anti-environmental interference dynamic reference curved surface by using historical data; constructing a weighted coupling graph based on the potential energy residual error and the spatio-temporal context, and screening potential repeated alarm clusters through belief propagation; performing time sequence alignment and multi-dimensional consistency evaluation on intra-cluster alarms, and eliminating low-contribution noise nodes in combination with Shapley value game analysis; and finally, identifying a stable defect chain through topology persistent coherence analysis, selecting main alarms and screening out repeated alarms. According to the method, the problem of misjudgment caused by illumination change, equipment aging and complex interference is effectively solved, the recognition accuracy and the automation level are greatly improved, and the operation and maintenance rechecking workload is remarkably reduced.

The invention provides an equipment exception work order intelligent management method and device, equipment and a storage medium, and the method comprises the steps: carrying out the multi-source fusion collection and semantic standardization processing of equipment exception data, and obtaining the standardized exception work order data; performing space-time correlation analysis on the abnormal work order data, and obtaining an equipment abnormal correlation network and a propagation path by calculating space-time correlation intensity and applying a causal inference algorithm; performing state prediction on the equipment anomaly association network and the propagation path by applying a graph neural network model, and obtaining work order state probability distribution and a change trend by adopting a graph convolutional network, an attention mechanism and a belief propagation algorithm; and executing dynamic priority ranking and resource allocation on the work order state probability distribution and the change trend, and obtaining an optimal work order processing sequence and a resource allocation scheme according to the network influence index and the resource matching degree. According to the invention, through abnormal correlation analysis and state probability prediction, intelligent management of the abnormal work order is realized.

The invention discloses a typhoid theory teaching method based on a knowledge graph, and belongs to the technical field of typhoid theory teaching. The method comprises the steps of multi-version data acquisition and standardization, ancient Chinese context semantic unit construction, term semantic disambiguation based on comparative learning, semantic conflict detection based on consistency constraint, multi-source information fusion and map optimization and teaching path generation. Through ancient Chinese context semantic modeling and comparative learning, precise disambiguation of traditional Chinese medicine ancient book terms is realized, and knowledge errors caused by polysemy of one word are reduced; through an automatic detection mechanism of logic consistency constraint, the quality and reliability of the knowledge graph are improved; a version traceability and belief propagation optimization strategy is adopted, so that the finally constructed knowledge graph not only can integrate multi-version essence, but also can clearly trace knowledge sources and evaluate knowledge credibility; the teaching path chain generated based on the atlas can dynamically simulate the whole process of traditional Chinese medicine syndrome differentiation treatment, and the clinical thinking training effect is remarkably improved.

The invention relates to the technical field of mine rescue detection, and discloses a mine rescue multivariate information life detection system and method.The mine rescue multivariate information life detection method comprises the steps that a heterogeneous sensor network is deployed, a belief propagation algorithm is applied to construct a self-organizing network, and dynamic topological optimization is achieved; an environment adaptability calibration system is constructed, and dynamic calibration of the sensor in complex environment changes is achieved; the sensor performance is quantified through a functional space model, degradation evaluation is carried out, and intelligent redistribution of sensing tasks is achieved; constructing a cooperative detection framework by using a belief propagation algorithm, and fusing multi-sensor data through a distributed decision algorithm; performing multi-dimensional feature extraction and analysis on the life signal, performing reliability evaluation on a result, and outputting a high-reliability detection result; according to the invention, the technical problems of sensor performance drift, function degradation and insufficient topological structure adaptability of a traditional mine rescue detection system in a complex environment are solved.

The invention provides a digital twin accelerator beam modulation method based on Gaussian integrated belief propagation. The method comprises the following steps: initializing parameters; constructing a prior factor graph; generating a sample set and constructing a condition graph; variable nodes and factor nodes in the condition graph are converted, and a posterior graph is constructed; calculating an accelerator operation parameter estimation value; and acquiring an accelerator beam modulation result. According to the method, efficient calculation is carried out by using a near-low-order Gaussian matrix generated by a sample set, conversion can be carried out among set statistics, various parameterization and effective operation on the parameters, sampling is carried out from the generation priori, the set statistics are converted into a standard form, belief propagation is carried out, and a set sample matched with an updated belief is recovered; and low-rank local information is transmitted on the condition graph to update integration, so that complex dependent structures and high-dimensional states and parameters can be processed, and the efficiency of the beam modulation process of the accelerator is improved.

The invention discloses a new unit equipment risk management and control and spare part demand optimization method and system, and relates to the technical field of intelligent operation and maintenance, and the method comprises the steps: collecting operation condition signals in real time, constructing a digital twin model of each part of unit equipment, and forming a unified discrete state vector; performing feature extraction on the operation condition signal according to a fixed window, and constructing a Bayesian network of a hierarchical causal structure in combination with process attributes; performing posterior reasoning on the Bayesian network through belief propagation to obtain a fault posterior probability of a causal node, and calculating a risk score according to a weight and a consequence cost; and life parameter estimation is carried out, Monte Carlo simulation is used to predict the demand quantity, and the optimal spare part order quantity is calculated in combination with inventory constraints. According to the method, potential risks can be found in time, non-planned shutdown is reduced, inventory redundancy and capital occupation are reduced, and the safety, reliability and economical efficiency of operation of a new unit are improved.

The invention provides a hydraulic floating bridge ontology-free data reliability analysis method based on a time sequence decoupling network and a Bayesian network, and relates to the technical field of hydraulic floating bridges. Comprising the steps of determining source domain equipment according to a bill of material and a structure diagram of a target hydraulic floating bridge; determining a standardized time sequence monitoring data set of the source domain equipment; training a time sequence decoupling network by using the standardized time sequence monitoring data set to obtain a health probability data set of each key component of the target hydraulic floating bridge; inputting the health probability data set into a four-layer Bayesian network, and recursively calculating the posterior failure probability of each layer from bottom to top by adopting a belief propagation algorithm to obtain failure probability data and posterior probability distribution data; and performing risk assessment according to the failure probability data and the posterior probability distribution data to obtain a risk assessment result. The technical problems that in the prior art, data scarcity and model island exist in operation and maintenance of a hydraulic floating bridge and similar equipment, and reliability evaluation cannot be carried out are solved.

The present disclosure relates to an apparatus and method for decoding a quantum low-density parity-check code. An aspect of the present disclosure is to provide a technique of performing iterative decoding when error correction fails, in a process of decoding a quantum low-density parity-check code based on a belief propagation algorithm.

The invention relates to the technical field of computer artificial intelligence, in particular to a mobile edge network collaborative reasoning method based on DRL and BP, and provides an edge-terminal collaborative reasoning scheme in a large language model, which fully utilizes the computing power of terminal equipment and an edge server, protects the data privacy of a user to a certain extent, and improves the user experience. Under the condition of limited resources, the problems of reasoning task unloading and model caching are solved by utilizing a'learning-to-learning 'normal form in a combined manner. A lightweight and high-popularity model is selected and cached to an edge server by using deep reinforcement learning, and a related reasoning task unloading problem is solved by adopting a distributed belief propagation technology. A numerical result proves the effectiveness and superiority of the proposed scheme. By means of edge terminal cooperation, the communication and computing resource utilization rate in LLM reasoning can be improved. In other words, diversified reasoning requests can be effectively completed through effective small model caching and reasoning task offloading.

A method of processing wireless communication signals for joint communication and environmental awareness in a region of interest (ROI) is presented. The method alternately uses a first iterative process and a second iterative process for determining an initial representation of a voxelized environment of the ROI and for determining the transmitted signal. The first iterative process and the second iterative process may implement a Gaussian approximation belief propagation process.

The invention discloses a situational medical care emergency training method based on an intelligent decision tree, and the method comprises the following steps: S1, collecting and preprocessing medical care emergency historical case data, and extracting an expert path; s2, constructing a medical emergency situation virtual environment based on the structured data and initializing an intelligent decision tree; s3, performing evolution operation on the plurality of decision paths by adopting a genetic algorithm to generate a decision path set; s4, fusing the decision path and the expert path by adopting a belief propagation algorithm to generate an emergency decision path; s5, emergency training is carried out in the virtual environment, recording and evaluation are carried out, and an adjusted path is formed; and S6, redeploying the adjusted path to the decision tree, and carrying out targeted drilling. According to the method, the intelligent decision tree integrating genetic optimization and a belief propagation mechanism is constructed, so that accurate and efficient training and path optimization of medical staff in a situational emergency environment are realized.

This invention discloses a code-assisted baud rate sampling clock recovery method and system, applicable to the field of optical fiber communication technology. The method includes the following steps: performing matched filtering on the received signal; using the M&M algorithm to complete primary clock recovery on data from an oscilloscope to obtain baud rate sampling data; performing equalization processing on the baud rate sampling data; using a belief propagation algorithm to decode the equalized baud rate sampling data and output reliability estimation information; modulating the reliability estimation information of each baud rate sampling data into a PAM4 auxiliary symbol according to the PAM4 mapping rule; using the PAM4 auxiliary symbol to replace the decision symbol in the M&M algorithm to re-estimate the clock error and perform high-precision clock recovery. This invention, by combining primary and auxiliary clock recovery stages, significantly reduces the bit error rate and provides more stable clock synchronization in high data rate scenarios.

The application relates to a weak underwater target azimuth detection pre-tracking method based on belief propagation, which comprises the following steps: converting a weak target tracking problem into solving a joint posterior probability density function (pdf) based on a Bayesian rule; factorizing the joint posterior pdf, constructing a corresponding factor graph through a graph model; solving the transmitted information in the factor graph by using a belief propagation (BP) algorithm, and converting multiple integrals into ordinary integrals by using a Goodman variable substitution principle; constructing a likelihood function by using a complex Wishart distribution to describe the relationship between original sonar data and target states, and simplifying the remaining pdf by using a Gaussian distribution to obtain a confidence approximation of a target state edge posterior pdf; and calculating the number of weak targets and corresponding states by using a minimum mean square error (MMSE) estimator. The application can simultaneously track multiple weak targets without a complex clustering algorithm, has fewer preset parameters, and can accurately and efficiently track target state information including an azimuth, an azimuth angular velocity and target intensity.

A method of decoding information encoded in respective activation modes of an activated subset of antennas of a MIMO transmitter transmitting orthogonally modulated signal components, the method comprising: decoupling a real part and an imaginary part of a signal component of a received signal vector, respective channel matrices of real and virtual channels are received, determined, or estimated, and real and virtual signal components are represented as respective products of transmitted symbol components and activation unit vectors. The respective activation unit vectors of the decoupled real and imaginary parts are then represented with corresponding random variable vectors that are dependent on a probability quality function of the activation pattern available at the transmitter. Then, a Gaussian belief propagation-based messaging process is applied to the random variables to jointly estimate activation unit vectors of the respective real and imaginary parts of the received signal vector, and finally the activation unit vectors are decoded to obtain the information.

The invention discloses a target fusion method based on average field belief propagation under an unknown time-varying parameter condition, and belongs to the technical field of radar target fusion tracking. The method aims at solving the problem that an existing target fusion algorithm cannot achieve target fusion under the condition of unknown time-varying parameters. Comprising the steps of calculating a marginal probability of a target existence probability based on a Markov law; setting an associated data set and determining constraint conditions that the associated data of each sensor comes from a target and clutter; calculating prior probability distribution; calculating a joint posterior probability density function based on the conditional probability; establishing a factor graph model based on average field-belief propagation, reconstructing a joint posterior probability density function and calculating a belief value; solving a target motion state, a target existence probability, a measurement noise covariance and a correlation variable; and iteratively solving until the difference value of the target motion states in two adjacent calculations is smaller than a preset threshold value, and taking the current target motion state as a final target fusion state. The method is used for target fusion under the unknown time-varying parameter condition.

The present disclosure provides an iterative receiving method of a polar coded continuous phase modulation signal and a related device. The method comprises: receiving, by a demodulator, a continuous phase modulation signal from a channel, and determining an initial accuracy parameter, a forward metric and a backward metric of the continuous phase modulation signal; performing, by the demodulator, an operation on the initial accuracy parameter to obtain a target accuracy parameter; performing, by the demodulator, a recursive operation on the forward metric and the backward metric based on the target accuracy parameter to obtain a multi-ary code word probability, and obtaining an estimated information stream at a receiving end based on the multi-ary code word probability and converting the estimated information stream into a decoder prior information stream; performing, by a decoder, a decoding process on the decoder prior information stream based on a serial cancellation list decoding algorithm to obtain a candidate code word sequence, performing a decoding process on the candidate code word sequence based on a belief propagation decoding algorithm, and judging whether the decoding process meets a preset termination condition, and outputting a source information bit or a target information stream at the receiving end according to a result of the judgment.

The invention discloses a construction and evaluation method for an applicant drug clinical test quality management system, and aims to solve the problems that traceable evidence chains are lacked among applicant clinical test quality system elements, evidence materials and evaluation conclusions, evaluation is difficult to quantify, and closed-loop improvement based on evaluation results is difficult. According to the method, quality system element data, evidence material data and evidence source data are obtained and standardized, and a time heterogeneous evidence graph comprising element nodes, evidence nodes, version nodes, responsibility nodes and evidence source nodes is constructed; executing graph neural network representation learning on the evidence graph to generate a relationship likelihood parameter and an evidence credibility likelihood parameter; constructing a Bayesian factor graph containing conflict factors, and executing belief propagation to obtain element compliance posterior distribution and evidence credibility posterior distribution; and further calculating a compliance score, an uncertainty score and a coverage rate score to form a three-dimensional score result, and extracting an evidence chain to generate traceable output, thereby achieving the technical effects of automatic establishment of the evidence chain, compliance quantitative evaluation, conflict evidence suppression and evaluation result driven closed-loop improvement.