386 results about "Prediction score" patented technology

The invention relates to an article sequence recommendation method and system based on collaborative filtering and an LLM perspective, and the method comprises the steps: obtaining a user historical data set and an article data set, and carrying out the preprocessing, and obtaining an article title similarity matrix, a user behavior sequence, and historical collaborative filtering interaction information; historical collaborative filtering interaction information is enhanced through a large language model; constructing a sequence recommendation model, inputting the user behavior sequence into the sequence recommendation model for training, and correcting the deviation of the user behavior sequence through comparative learning according to the item title similarity matrix and the enhanced historical collaborative filtering interaction information to obtain a trained sequence recommendation model; and inputting the user behavior sequence of the user into the trained sequence recommendation model for prediction, calculating prediction scores, and generating a recommendation list according to score sorting, thereby completing article sequence recommendation. According to the method, the performance of the sequence recommendation system is greatly improved by solving the cold start problem.

The invention relates to an intelligent lighting energy-saving method and system based on multi-modal data fusion. The method comprises the following steps: periodically obtaining a decision trajectory data set; combining the experience pool of the previous period with the decision trajectory data set newly collected in the current period according to a set proportion, and screening according to priority weight; training to obtain candidate strategies; evaluating the candidate strategies in a test scene set; adding the strategy comprehensive score into a strategy library when the strategy comprehensive score meets an access condition; when the strategy capacity in the strategy library reaches a preset value, eliminating the strategy with the lowest comprehensive strategy score based on the Pareto frontier; constructing a fitness function of the strategy and establishing a feedback-weight mapping model used for obtaining a prediction score; and based on the feedback-weight mapping model, establishing a loss function gradient, and based on the loss function gradient, performing iterated weight calculation to obtain iterated weights after the iterated calculation. According to the invention, the long-term energy-saving efficiency and the user experience of the lighting system are improved.

The invention relates to the technical field of electric power system communication, in particular to an electric power 5G network service quality monitoring method and system and a storage medium, and the method comprises the steps: responding to a protocol data unit session establishment request signaling, and obtaining signaling plane data and data plane performance data through a signaling plane probe and a data plane probe; associating the signaling plane data with the data plane performance data through the service quality flow identifier and the timestamp to generate a multi-dimensional spatial-temporal characteristic matrix; based on the multi-dimensional spatio-temporal characteristic matrix, a network quality prediction score is obtained through dynamic spatio-temporal causality graph network prediction; and when the network quality prediction score exceeds a preset score threshold, establishing a network resource configuration optimization model to obtain an optimal network resource configuration strategy. According to the method, the signaling plane data and the data plane performance data are subjected to correlation analysis, and the network quality is predicted by using the dynamic space-time causal graph network, so that the network resource configuration is optimized, and the service quality of the power 5G network is ensured.

The invention relates to the technical field of artificial intelligence, in particular to a process reward model training method and system. The method comprises the following steps: calculating a confidence score of a question text in each sample; based on the correctness score, the confidence score and the tolerance distance hyper-parameter of each reasoning step of the problem text in the labeled sample, obtaining a target correctness score of each reasoning step of the problem text in the labeled sample; and training the process reward model based on binary cross entropy loss between the correctness prediction score of each reasoning step of the problem text in the labeled sample and the target correctness score, and taking the trained process reward model as a target process reward model. According to the method, the accuracy and reliability of process reward model training are improved, and the text generation precision of a large language model is enhanced.

The invention provides a risk control cross-domain risk prediction method and system in combination with transfer learning, and the method comprises the steps: carrying out the domain difference measurement of a source domain and target domain risk control scene, generating a domain difference measurement matrix, building a hierarchical knowledge transfer channel based on the domain difference measurement matrix, and carrying out the prediction of the risk control cross-domain risk. And selectively migrating the source domain risk prediction model parameters, generating a target domain initialization risk prediction model, and carrying out dynamic adaptive training on the target domain initialization risk prediction model by adopting a target domain risk sample to obtain a cross-domain risk prediction model. And inputting the to-be-evaluated sample of the target domain into the cross-domain risk prediction model to generate a preliminary risk prediction score, and finally performing cross-domain calibration on the preliminary score based on the domain difference metric matrix to generate a final risk prediction result, thereby improving the accuracy of risk control cross-domain risk prediction.

The invention discloses an N-sugar chain marker combination for predicting the curative effect state of IgA and IgG type MM and a prediction scoring system and application of the N-sugar chain marker combination. The N-sugar chain marker combination is composed of the following 11 N-sugar chains: NGA2, NGA2F, NG1A2F-1, NG1A2F-2, NA2, NA2F, NA2FB, NA3, NA3Fb, NA4 and NA4Fb. The method has the advantages that (1) the relevance between the N-carbohydrate chain and the curative effect in the blood of the IgA and IgG type MM patients in different curative effect states is found for the first time, and the N-carbohydrate chain can be used as an index for predicting the curative effect states of the IgA and IgG type MM patients for the first time; and (2) the N-carbohydrate chain curative effect prediction scoring system provided by the invention is used for predicting the curative effect states of IgA and IgG type MM patients, and a new technical means is provided for predicting the curative effects of the IgA and IgG type MM patients.

The invention discloses a psychological counseling intelligent recommendation method and system based on artificial intelligence, and relates to the field of psychological counseling cross technology, and the method comprises the steps: calculating a three-dimensional emotion vector according to an emotion axis space and a syntactic structure, constructing a psychological semantic tensor by using a tensor outer product, generating a user intention tensor by using a prototype weight, and carrying out a flattening operation. Obtaining a final intention vector; combining the structural potential energy deviation and the spectral distance to construct a matching factor tensor; constructing a factorization machine model, predicting a matching score, generating a prediction score matrix, screening by using a Top-K selection method, generating recommended content, and constructing a visual interface to display the recommended content. According to the method, the fine granularity and accuracy of emotional representation are improved through emotional axis space and syntactic dependency analysis, the matching precision of intention and resources is improved through comprehensive calculation of semantic gravitation intensity and spectral distance, and the expression ability of matching factor tensor is enhanced through fusion of structural potential energy deviation and spectral distance.

The invention provides a human behavior recognition method based on fusion of multi-modal data acquired by an unmanned aerial vehicle, and the method comprises the following steps: splicing a joint Tokens with a space CLS Tokens of the same frame after feature extraction is completed by using a space Transform, and obtaining a feature fusion module; performing structure coding on the joint point sequence based on a human anatomy structure, and constructing a human topological structure association module; the method comprises the following steps of: constructing a time sequence cross Transform module; the space Transform, the feature fusion module, the human body topological structure association module and the time sequence crossing Transform module are stacked to obtain a trunk feature extraction network; a 10-layer trunk feature extraction network is adopted to perform deep feature extraction on input early-stage fusion data and feature fusion data, and classification is performed by using a classification head; the features output by the backbone network are mapped to the dimensions with the same number as the behavior categories, a prediction score is output for each category, and a human body behavior recognition result is obtained according to the prediction scores. According to the method, the human body behavior in the multi-modal data acquired by the unmanned aerial vehicle can be effectively identified.

The invention relates to the technical field of computers, and discloses an account risk prediction method and system based on machine learning, and the method comprises the steps: obtaining transaction details, login behaviors and terminal environment information of a target account in different time periods, and constructing an account dynamic behavior feature data set; performing multi-layer nested feature extraction on the account dynamic behavior feature data set, and constructing an account risk behavior evolution model by adopting a time sequence convolutional network and attention mechanism fusion algorithm; judging whether the risk sensitivity of the account risk behavior evolution model in the current training period is stable or not based on the trend change of the response offset; aiming at the corrected behavior sequence, applying a multi-stage decision network and an anti-factual reasoning mechanism to generate a prediction score matrix of the potential risk account; and performing multi-dimensional fusion evaluation on the current state of the account according to a misjudgment boundary correction result in combination with a preset risk control knowledge base. The method has the advantage of reducing the misjudgment rate.

The invention discloses a doctor-patient communication risk assessment and early warning system and method based on an internet hospital, and the system comprises a basic feature calculation module which is used for calculating the basic feature score of the early warning according to the collected online inquiry data of a doctor and a patient, and calculating the weight ratio; the dynamic feature analysis module is used for taking the dynamic feature factors as input of the constructed patient end prediction model and the constructed doctor end prediction model and respectively calculating to obtain patient dynamic prediction scores and doctor dynamic prediction scores; the risk fusion prediction module is used for performing feature fusion and final risk prediction by taking the patient and doctor dynamic prediction scores, the number of times of previous complaints of the patient, the number of times of previous complaints of the doctor and the basic feature scores as input of a constructed fusion prediction model, and performing calculation to obtain a doctor-patient communication risk score; and the early warning triggering module is used for triggering an intelligent early warning mechanism when the doctor-patient communication risk score exceeds a set threshold value. According to the invention, efficient and accurate doctor-patient communication risk assessment and intelligent early warning can be realized.

The invention discloses a news detection method and system based on a dual-model collaborative optimization framework, and a medium. The method comprises the following steps: screening news materials and news descriptions from a news material library to construct a training data set; respectively inputting the training data set into a student model and a teacher network model to obtain a first image feature and a second image feature, and aligning the first image feature to a space of the second image feature to obtain a third image feature, inputting the third image feature into a decoder of a teacher network model to perform target query optimization so as to obtain a fourth image feature, taking a prediction score obtained by the teacher network model as a soft label, supervising training of a student model until a preset condition is met, and determining a trained target detection model; and inputting the news material obtained in real time into the target detection model, and determining the news category and the sensitive element, so that the news detection accuracy and efficiency can be improved.

The invention discloses a method for establishing a rough-arrangement search model based on LTR, and the method comprises the following steps: S1, collecting candidate total data, and completing the preprocessing operation; s2, the sorting score of each piece of data is calculated and arranged in an ascending order, and the data are divided into equivalent data buckets; s3, randomly extracting data to form a training sample pair, generating a Pair-wise data set, and marking a sorting relationship; s4, executing rough-arrangement search model training and calculating a prediction score; s5, constructing a loss function by adopting a Lambda gradient signal, feeding back a sorting error and iteratively updating model parameters; s6, monitoring a training performance index, stopping training after a stopping condition is met, and deploying the model to an online environment; and S7, storing the trained model file, and defining an input format, a sorting structure and an output interface. According to the method, the sorting accuracy and fine sorting consistency of the rough sorting search model are improved, and the generalization ability of the model is remarkably enhanced.

The present invention relates to a system and method for the enhanced detection and quantification of respiratory events, such as apneas and hypopneas, particularly useful in diagnosing and managing sleep-disordered breathing conditions. This system integrates sophisticated machine learning algorithms, specifically one-dimensional convolutional neural networks (1D CNNs), with advanced signal processing techniques to analyze physiological signals. It focuses on the detecting respiratory event with a localized portion of the input segment, a novel approach that increases specificity in event detection. The system segments physiological signals into overlapping segments, each analyzed by the machine learning model to generate a prediction score. A unique aspect of this invention is the application of a dual-threshold mechanism: a ‘Model threshold’ for initial event identification and a ‘Vote threshold’ for confirming events through an aggregate voting process of overlapping segment predictions. This innovative approach ensures high accuracy and reliability in event detection and counting.

The invention discloses a sports market demand prediction method and device based on deep learning and a medium, and relates to the technical field of market demand prediction, and the method comprises the steps: collecting sports demand data, and carrying out the preprocessing; performing relation mining and graph structure learning on the preprocessed sports demand data through a graph attention space-time network to generate a macroscopic demand potential energy graph; performing potential area identification on the macroscopic demand potential energy diagram by adopting a pre-trained sports market simulation model, and outputting local demand prediction data; performing weighted fusion and error correction on the macroscopic demand potential energy map and the local demand prediction data, and outputting a sports demand prediction score; and making a sports market demand strategy according to the sports demand prediction score and the multi-granularity demand prediction report, and transmitting the sports market demand strategy to an enterprise manager through an enterprise decision support interface. According to the method, multi-level accurate prediction and decision support of sports market demands are realized through dual-mechanism cooperation of the graph attention space-time network and the space-time convolution.

The invention discloses a curriculum recommendation method for two-channel comparative learning and heterogeneous knowledge graph propagation, and relates to the technical field of machine learning and recommendation systems, and the method comprises the steps: 1, constructing a curriculum heterogeneous knowledge graph and an initial entity set; step 2, cooperative propagation of the curriculum heterogeneous knowledge graph and combined generation of a ripple set; step 3, designing a fine-grained convergence attention network; step 4, performing two-channel contrast learning; 5, performing multi-hop path reasoning and multi-view fusion; step 6, prediction score calculation and model optimization; according to the course recommendation method based on dual-channel comparative learning and heterogeneous knowledge graph propagation, the technical problems of data sparsity, insufficient coverage of long-tail courses, poor interpretability of recommendation results and the like in the prior art are solved.

The invention discloses a method for detecting the content of salidroside in a rhodiola rosea extracting solution based on artificial intelligence. The method comprises the following steps: S1, performing spectral scanning of different wavebands on an extracting solution sample in a plurality of near-infrared wavelength ranges; s2, performing multi-dimensional interference suppression processing and signal purification; s3, carrying out spectral feature compression calculation to obtain a salidroside feature intensity factor; s4, carrying out salidroside characteristic signal-to-noise ratio enhancement and purity index calculation; s5, predicting the content of salidroside and outputting a content prediction score; and S6, mapping the content prediction score into the final salidroside content. According to the method, spectral signal purification, nonlinear feature extraction, dynamic noise enhancement, artificial intelligence prediction modeling and a secondary recheck mechanism are utilized to realize adaptive analysis of extracting solutions from different sources, so that the detection accuracy and stability are remarkably improved, and rapid and traceable salidroside content detection can be realized under the condition that a large precise instrument is not needed.

The invention discloses a pi RNA and disease association prediction method, device and equipment based on comparative learning. The method comprises the following steps: step S1, constructing a p RNA-disease heterogeneous graph network; s2, fusing the multi-level semantic information; step S3, according to the node representation, introducing a Transform model, and enhancing the global association modeling capability of the piRNA and the disease node; s4, constructing a topological graph and a semantic graph; step S5, obtaining an accurate p iRNA-disease association prediction score; the prediction device comprises a heterogeneous graph construction unit, a node embedding learning unit and a node embedding learning unit. An input and output unit, a storage unit, a communication unit, an RAM unit, an ROM unit and a GPU of the electronic equipment are connected with one another through a bus, and the requirements for complex calculation and data interaction of a p-RNA and disease associated prediction task are met. The method has the characteristic of high prediction accuracy.

The invention belongs to the technical field of open source software supply chain security, and particularly relates to a security patch variant positioning method based on two-dimensional feature design. By designing the two-dimensional features and applying the rule model, comprehensive, accurate and automatic positioning of the security patch variants in the specified open source software code warehouse is realized. The method comprises the following specific steps: data preprocessing: extracting basic information in a submission record and carrying out normalization processing; feature generation: designing and calculating similar features and characterization features of submission record pairs; and variant prediction: predicting the extracted features by applying a random forest model, and judging whether the submitted record is the security patch variant according to a prediction score. According to the method, by widely mining similar features and characterization features between patch variants and combining with a rule-based model, the comprehensiveness, accuracy and automation degree of patch variant positioning are improved, and an effective patch variant positioning tool is provided for software security analysts.

The invention relates to the technical field of data analysis, and discloses a corrugated board quality control method and system based on data analysis and a medium. The method comprises the following steps: acquiring body paper performance parameters and environment and process parameters through a sensor, and performing data processing and verification to obtain an original data set; the data are grouped through standardization processing and a clustering algorithm, and factor groups related to the strength, the thickness deviation and the pressure resistance of the paperboard are determined; predicting the factor group by adopting a neural network prediction model to obtain a prediction score, comparing the prediction score with a target standard value, and calculating a deviation factor; fusing the deviation factor and the process parameter range through a weighted summation method to obtain a preliminary quality control index set; performing weight adjustment on the index set according to the paperboard category and the production batch to obtain an optimized quality control index; and a change trend is extracted from the optimization indexes, and a real-time quality early warning signal is generated by adopting threshold comparison. The quality control and real-time early warning capability in the corrugated board production process is improved.

The invention provides a method for predicting microorganism-drug interaction through a double hypergraph contrast learning framework based on hierarchical attention, and solves the problems that the graph structure is simple, the interpretable analysis is increased, and the prediction precision is improved. The method consists of three modules: S1, data preprocessing: carrying out nonlinear fusion on similarity data of microorganisms and drugs to obtain comprehensive similarity of the microorganisms and the drugs, and forming a double hypergraph by combining an original incidence matrix and the comprehensive similarity with KNN and K-means algorithms; s2, feature extraction: sending the double-hypergraph obtained in the step S1 into a hierarchical attention and hypergraph convolutional network to obtain feature embedding, updating features based on comparison learning of the double-hypergraph, and fusing the double-hypergraph features through an integration network. And S3, result prediction: mapping the microorganism and drug characteristics obtained in the step S2 by using a full connection layer to obtain final characteristics, and performing dot product to obtain a prediction score. And outputting the obtained prediction score in combination with specific literature verification and a network pharmacology verification model.

The invention discloses a self-updating method and device of a financial risk control model, a storage medium and a terminal, relates to the technical field of data processing, can be applied to the field of financial risk control, and mainly aims at solving the problem that the model recognition accuracy is low due to the fact that an existing financial risk control model is not timely updated. The method mainly comprises the following steps: acquiring anomaly prediction scores and confidence coefficients of different newly-added samples obtained in a process of performing anomaly identification on newly-added sample data in a production environment by a financial risk control model; extracting a low-confidence sample from the newly added samples according to the abnormal prediction score and the confidence; performing clustering processing on the low-confidence samples, and constructing an updated training sample set according to target samples extracted from each cluster; and performing incremental learning update training on the financial risk control model based on the training sample set, so as to continue to execute anomaly recognition of subsequent sample data based on the financial risk control model completing update training. The method is mainly used for updating the financial risk control model so as to improve the timeliness of model updating.

The invention discloses a cross-domain recommendation system based on multi-view double distillation and comparative learning, and belongs to the technical field of data processing. The system comprises a local view module, a global view module and a prediction layer. The local view module constructs a single-domain user-article interaction graph, aligns domain shared representations by using instance-level contrast learning, and distinguishes domain specific representations by using feature-level contrast learning. The global view module improves the mobility of user and article representation by constructing a heterogeneous user-article interaction graph, an isomorphic user-user similarity graph and an article-article relation graph and adopting a bidirectional distillation mechanism to optimize knowledge migration. And the prediction layer fuses the user and article representations of the local view and the global view, calculates a preference prediction score and generates a recommendation result. According to the method, the user preference representation can be effectively decoupled among different fields, the negative migration risk is reduced, and efficient transmission of cross-domain knowledge is realized. According to the method, the recommendation precision is remarkably improved.

The invention discloses a synthetic lethal prediction method based on non-generality forgetting and low-rank interaction multi-mode adaptive fusion, and relates to the technical field of bioinformatics, computational biology and drug target screening. The method comprises the following steps: firstly, constructing a multi-view gene embedding expression based on a gene graph structure, a hypergraph structure and a knowledge graph; then constructing a difference vector and an interaction vector for any gene pair so as to characterize the difference and potential complementary relationship between the genes; multi-source information fusion is realized by using a deep learning model containing interactive attention, a forgetting gate and a multi-modal fusion mechanism; on this basis, multi-task optimization is carried out through joint classification loss, modal interaction loss and redundancy suppression loss, and optimal discrimination is realized through a dynamic threshold search strategy; and finally, outputting a sorting result of the candidate synthetic lethal gene pairs according to the prediction score. According to the method, the accuracy and generalization ability of synthetic lethal relationship prediction can be effectively improved, and a high-reliability calculation auxiliary tool is provided for anti-cancer drug target screening and gene therapy strategy design.

The invention relates to the technical field of biological gene mutation prediction, in particular to a multi-omics feature-based missense mutation function classification method and system. The classification method comprises the following steps: S1, collecting a multi-omics comprehensive feature set related to missense mutation; s2, constructing a heterogeneous graph; s3, learning a meta-path-based weight of each node in the heterogeneous graph; s4, calculating weights of different meta-paths by using attention of a semantic level; s5, performing weighted summation on the node weight and the meta-path weight to obtain final embedding representation of each node; s6, training to obtain a multi-omics feature-based missense mutation function classification model; and S7, outputting a prediction score based on the multi-omics feature missense mutation function classification model, and dividing missense mutation according to the prediction score. According to the method, the heterogeneous graph is constructed by utilizing multiple omics characteristics, and the relationship between biological entities is explicitly modeled, so that GOF and LOF mutations are distinguished more accurately.

The invention discloses a recommendation system preference forgetting method based on a user angle, and the method comprises the steps: firstly building an interaction matrix of a user and an article based on a recommendation system; secondly, a recommendation system module is constructed to learn preference information from the interaction matrix, an attack module is constructed to modify the interaction matrix through an interaction disturbance matrix, so that the preference of a target user set forgotten by preference cannot be learned by the recommendation system module, and finally, the interaction disturbance matrix of the modified interaction matrix is trained and optimized through double-layer optimization. According to the method, by adding the forged interaction data to the interaction matrix of the target user in a limited manner, the prediction score of the recommendation system on the sensitive article of the target user can be remarkably reduced, so that the forgetting effect is achieved, the model does not need to be invaded or relied on model parameter information, and the overall performance of the recommendation system can be maintained while the forgetting effect is guaranteed.

The invention provides a content popularity prediction method and device and a medium, and the method comprises the steps: determining a target evaluation user set, and obtaining the behavior characteristics of each target evaluation user in the target evaluation user set for to-be-predicted content and the touch scene characteristics; generating a user evaluation feature vector corresponding to each target evaluation user based on the behavior feature of each target evaluation user for the to-be-predicted content and the touch scene feature; inputting the user evaluation feature vector of each target evaluation user into a pre-trained content popularity prediction model, so that the content popularity prediction model calculates the score of the to-be-predicted content in combination with the touch scene features to obtain a comprehensive score of each target evaluation user for the to-be-predicted content, and determining a popularity prediction score of the to-be-predicted content. In this way, richness of the popularity prediction dimension of the to-be-predicted content can be improved, and then the efficiency and accuracy of popularity prediction of the to-be-predicted content are improved.

The present disclosure describes a system, method, and computer program for predictive autoscaling for faster searches of event logs in a cybersecurity system. In one embodiment, the system receives search-related signals from a plurality of signal sources. The signals are indicative of: (1) a user's intent to perform a search for event logs in a cybersecurity database, (2) how computationally intensive the potential search is likely to be, and (3) the currently available computational resources. The signals are evaluated, and an autoscale prediction score is calculated. The autoscale prediction score reflects the likelihood of a user to submit search, the computational resources required for the potential search, and the currently available computational resources. The system scales computational resources in accordance with the autoscale prediction score. These steps are performed before any search is submitted by the user in a search user interface.

The invention belongs to the field of educational informatization, and discloses a student-teacher-student double-selection recommendation method and system oriented to individual requirements and used for assisting in lesson of graduate students. Calculating the demand similarity of both teachers and students, and screening part of candidate lists; establishing a scoring matrix according to the matched historical data; the singular value decomposition scoring matrix is a product of three matrixes, namely a student feature matrix, a teacher feature matrix and a diagonal matrix; optimizing a student feature matrix and a teacher feature matrix; calculating a prediction score matrix according to the optimized student feature matrix and the optimized teacher feature matrix; and for each user, obtaining a teacher or a student with the highest predicted score from the objects in part of the candidate list through the predicted score matrix, and selecting the teachers with the top M predicted scores or the students with the top M predicted scores as a recommendation list. According to the method, the teacher-student matching accuracy can be improved, similarity calculation can be carried out by utilizing attribute information of newly added students or teachers, and content-based recommendation is carried out.

The invention relates to a postoperative complication prediction method and device based on a two-way map neural network, and the method comprises the steps: collecting ICG video images before and after cerebrovascular bypass surgery, and obtaining the clinical feature data of a patient; preprocessing the collected ICG video image, extracting a blood vessel network by adopting an image segmentation algorithm, and generating corresponding blood vessel graph structures before and after bridging; extracting blood vessel features, blood flow dynamic features and clinical features from the collected ICG video images and clinical feature data, and constructing a multi-modal high-dimensional feature set; and taking the multi-modal high-dimensional feature set and the corresponding vascular graph structures before and after bridging as input, and generating a complication prediction score through a pre-trained prediction model based on a two-way graph neural network. Compared with the prior art, the method has the advantages that the multi-modal feature information is fused, the graph neural network structure is optimized, the model performance is stable, the accuracy is high, and the method has important clinical application value.

The invention relates to the technical field of computers, in particular to a component risk prediction method and system fusing a time decay factor and a storage medium, and the method comprises the steps: analyzing a component interaction relationship, building a node set and an edge set, and introducing the time decay factor to construct a time decay weighted component dependency relationship graph; performing dynamic graph structure feature analysis and time sequence modeling on the nodes and the associated edges thereof, and by extracting multilayer neighborhood features of the nodes, constructing a time sequence behavior model and calculating a structure lag coefficient to generate a graph feature set; introducing a service weight to carry out multi-layer dependency fusion and weight dynamic adjustment to obtain a graph feature set after weight fusion; and constructing a risk prediction input matrix, inputting the constructed risk prediction input matrix into a preset risk prediction model, and outputting a risk prediction score and a risk level of each component by the model. According to the method, the accuracy and timeliness of component risk prediction in a complex system can be remarkably improved.