2351 results about "Sequencing data" patented technology

The invention relates to the technical field of artificial intelligence, can be applied to business scenes such as agent autonomous decision making, financial science and technology and medical health, and discloses a multi-modal sequence data processing method, device and equipment and a medium. Extracting multi-scale feature hierarchies, combining the multi-scale feature hierarchies into a multi-scale feature pyramid set, performing cross-modal feature alignment to generate a multi-scale alignment feature sequence, executing local and global attention processing to generate long-distance dependency features, performing cross-layer information interaction to generate comprehensive multi-scale features, and performing multi-scale feature extraction; and dynamically fusing the multi-modal information and inputting the multi-modal information into a task decision network to obtain a target task result. According to the method, through the multi-scale feature pyramid, cross-modal alignment, attention processing and cross-layer information interaction, the problem of insufficient relevance between different modals and different scales in multi-modal long sequence data is solved, and fine modeling and dynamic fusion of multi-modal and multi-scale features are realized.

The invention discloses a three-dimensional human body posture estimation method based on hybrid architecture space-time modeling, and the method comprises the steps: obtaining historical human body posture data, and generating a three-dimensional human body posture estimation network training set; constructing a three-dimensional human body posture estimation network model, wherein the network model comprises a space-time position embedding module, a space-time Mama block, a space-time self-attention block, a full connection layer and a regression layer; based on the estimation network training set and a loss function training network model, learning a mapping relation from a two-dimensional attitude joint point sequence to a three-dimensional attitude joint point sequence; and inputting two-dimensional attitude joint point sequence data to be estimated into the trained network model, and outputting a three-dimensional attitude joint point sequence. According to the method, the global spatio-temporal features are rapidly extracted by using Mama, then the global spatio-temporal feature information is further supplemented by using Transform, and the spatio-temporal features are fused and complemented more effectively by using a hybrid architecture.

The invention discloses a train dispatcher behavior anomaly detection system based on LSTM-FCN and a dynamic intervention method thereof, and relates to the technical field of intelligent monitoring. Scheduling operation time sequence data, environment state data and physiological monitoring data are collected in real time, a space-time joint feature vector is constructed, and a train dispatcher behavior anomaly detection result is obtained. The LSTM-FCN hybrid network is used to extract time sequence dependence features and spatial pattern features, and an attention weight matrix generated by physiological data and a compensation factor matrix generated by environmental parameters are combined to dynamically calculate a comprehensive abnormal probability value; and when the probability value exceeds a self-adaptive adjustment threshold value, triggering a hierarchical intervention mechanism based on the driving scheduling knowledge graph to realize progressive intervention from interface prompting to manual takeover. By fusing multi-modal data and self-adaptive dynamic compensation, the problem of poor adaptability to complex scenes due to adoption of a static model in traditional train monitoring is solved, and the accuracy and intervention timeliness of abnormal behavior detection of the train dispatcher are remarkably improved.

The invention discloses an SPR response region identification method based on image semantic segmentation and time sequence alignment, and the method comprises the following steps: collecting SPR image frame sequence data, and constructing an original image sequence; performing image preprocessing operation on the original image sequence, and outputting a standardized image sequence; constructing a time sequence window image set composed of multiple continuous frames; inputting the time sequence window image set into an improved SegFormer model, and generating a response region segmentation mask image corresponding to each frame; executing cross-frame time sequence alignment operation of the response area, and outputting time sequence consistency identification mapping of the response area; performing area statistics, intensity analysis and time positioning operation; and generating a structured response region recognition result. The spatial-temporal evolution process of the response area in the SPR image sequence can be effectively recognized, the accuracy and stability of response area recognition are improved, and the method is suitable for high-precision biological detection and real-time molecular analysis scenes.

The invention discloses a civil aircraft PHM modeling method based on cross-modal coupling, a medium and equipment, and belongs to the technical field of aircrafts. The method comprises the steps of S1, multi-modal data preprocessing: performing feature extraction and time-space / event alignment for three modalities of flight time sequence data, text data and image data, and laying a foundation for subsequent fusion modeling; s2, cross-modal feature fusion and holographic data sample construction: constructing a holographic data sample library covering the whole life cycle through cross-domain fusion and dynamic optimization; and S3, PHM physical-knowledge fusion comprehensive modeling: constructing a double-layer collaborative framework combining physical simulation and knowledge reasoning, and realizing controllable fusion of physical simulation and multi-modal knowledge. The problems that in the prior art, multi-source heterogeneous data alignment is difficult, physical mechanism and data driving fusion is insufficient, and model interpretability is poor can be solved.

The invention discloses a posture feature recognition method based on a video stream, and the method comprises the steps: carrying out the preprocessing of a continuous video stream, obtaining video frame training data, extracting a key frame and an adjacent frame in each frame of image, constructing a feature extraction module for a human body region, and obtaining a global frame, performing local extraction on the human body area by using adjacent frames on the left side and the right side to obtain local frames, and constructing semantic association information for the global frame through time sequence continuity between the local adjacent frames and the current key frame; acquiring enhanced feature representation by adopting a conditional feature aggregation algorithm; obtaining attitude sequence data through the attitude detail features; the method comprises the following steps: establishing three-dimensional coordinates, adaptively extracting posture change data by adopting a human body motion decoupling model, predicting human body posture characteristics through a smooth optimization strategy, and introducing a cross attention mechanism to realize deep fusion of spatio-temporal characteristics, so that the understanding ability of the model to a complex action mode is enhanced; and the attitude expression capability of the model in a sheltered or fuzzy region is obviously improved.

The invention relates to the technical field of typhoon prediction, and discloses a typhoon rapid enhancement prediction method based on time-space sequence and multi-modal feature fusion, and the method comprises the steps: constructing a multi-modal time-space sequence data set and an auxiliary data set based on typhoon optimal path data and multi-source satellite observation data; a unified manifold approximation and projection method is adopted to carry out dimension reduction preprocessing on the high-dimensional multi-modal space-time sequence data, and one-dimensional time sequence embedding representation of the typhoon observation sequence is generated; taking the one-dimensional time sequence embedded representation and the auxiliary data as independent input channels, and inputting a trained typhoon observation network model to predict a typhoon rapid enhancement probability; wherein the typhoon observation network model is a multi-mode time-space fusion deep learning architecture, the core of the typhoon observation network model is composed of a variational attention recurrent neural network, and hyper-parameter optimization is carried out through an improved Harris eagle optimization algorithm. According to the invention, accurate and robust identification of the typhoon rapid enhancement process is realized.

The invention discloses a space-time decoupling human behavior recognition method, device and equipment based on a dynamic semantic guide mask. The method comprises the following steps: extracting human skeleton sequence data from an original video and performing data enhancement; based on a dynamic semantic guide mask mechanism, performing mask operation on the skeleton sequence after data enhancement in two dimensions of space and time; respectively sending the skeleton sequence data after mask operation into a query encoder and a momentum encoder, respectively obtaining spatial representation and time representation, constructing cross-domain contrast loss, and carrying out contrast learning training to obtain a human behavior recognition model; and inputting the to-be-recognized video into the human body behavior recognition model to obtain a prediction result of the human body behavior in the to-be-recognized video. According to the space-time decoupling human body behavior recognition method, device and equipment based on the dynamic semantic guide mask, the human body behavior recognition precision and the environmental adaptability are remarkably improved under the condition that manual labeling is not needed.

The embodiment of the invention provides a radar echo extrapolation prediction method and system based on a test time training strategy. According to the embodiment of the invention, the method comprises the steps: carrying out the time axis calibration and space grid alignment processing of a radar echo historical collection sequence collected by an atmosphere detection system, and generating radar echo sequence data with time-space consistency; constructing an echo evolution relevance representation model based on the data, and generating an echo evolution dependency graph and a multi-step extrapolation initial condition set; then introducing a test time training strategy to carry out adaptive prediction processing, and adjusting edge weight parameters in real time to obtain a radar echo multi-step extrapolation prediction sequence with time dynamic characteristics; and finally, according to the sequence, extracting an echo intensity change characteristic and a space movement track rule, and combining historical climate cycle evolution data to generate a climate evolution trend description set containing an echo intensity trend curve and a path offset vector so as to realize more accurate climate evolution prediction and weather forecast.

The invention belongs to the technical field of wind power equipment fault diagnosis, and relates to a wind turbine generator maintenance method based on multi-modal data fusion and a knowledge graph. Comprising the steps of obtaining text description data, component image data and equipment operation time sequence data of a wind turbine generator; predicting the residual life of the component based on the equipment operation time sequence data, and generating a residual life prediction value; performing feature extraction on the data to obtain corresponding features; performing fusion processing on the features to obtain joint feature representation; analyzing a fault evolution time sequence mode of the wind turbine generator based on the joint feature representation; performing matching retrieval in a historical case library according to a fault evolution time sequence mode; and inputting a matching retrieval result into the dynamic knowledge graph for reasoning, generating a fault traceability result, and generating a maintenance decision scheme including a maintenance priority list in combination with the residual life prediction value. According to the method, accurate fault diagnosis, automatic source tracing of root causes and dynamic optimization of maintenance strategies are realized, the operation and maintenance efficiency is remarkably improved, and the cost is reduced.

The invention discloses an intestinal cancer metastasis prediction method and device based on spatial omics, a medium and equipment, and the method comprises the steps: collecting original multi-omics data, and carrying out modal alignment and quality control processing to obtain pre-processed multi-omics data comprising second spatial transcriptome data, second single-cell RNA sequencing data and second pathological image data; performing cross-modal semantic embedding on the second spatial transcriptome data based on the second single-cell RNA sequencing data to generate a spatial enhanced expression profile; performing multi-scale graph construction on the second spatial transcriptome data and the second pathological image data, and extracting spatial heterogeneity features; inputting the spatial enhancement expression spectrum and the spatial heterogeneity features into a pre-trained metastasis risk prediction model, and outputting a liver metastasis probability spatial heat map and a key driving feature list; and finally generating a clinical prediction report containing high-risk area positioning. According to the method, through dynamic optimization of spatial resolution and multi-scale feature collaborative modeling, the sensitivity of early transfer detection is remarkably improved.

The invention relates to a building structure health state assessment method and device, and belongs to the technical field of building renovation, and the method comprises the steps: obtaining original stress time sequence data generated by dynamic strain response of a stress part of a target building under the action of a natural load or artificial excitation, building health state category labeling is carried out on the original stress time sequence data, and a multi-dimensional labeling vector with a timestamp is obtained; obtaining a health state evaluation model based on a hierarchical coding architecture of spatial-temporal feature decoupling, and training the health state evaluation model based on the clean stress time sequence data as labeled sample data in combination with unlabeled sample data to obtain a target health state evaluation model; and based on the target health state assessment model, data analysis is performed on the collected target building stress data, and the building structure health state of the target building is determined, so that the accuracy, continuity and early warning performance of building structure health state assessment can be improved.

The invention relates to the technical field of bioinformatics, and provides a visual analysis method and system for a rice multi-tissue single cell expression profile. The method comprises the following steps: comparing sequencing data of an original single cell transcriptome of a rice tissue to obtain a standardized transcriptome data set; performing batch effect correction and integration on the standardized transcriptome data set to obtain a whole plant expression matrix; performing cell type annotation on the whole plant expression matrix to obtain a cell type annotation system; carrying out visual dimension reduction processing on the whole plant expression matrix fused with the cell type annotation system, and carrying out co-expression network construction to obtain a modular tissue correlation analysis model; and establishing an interaction end based on the module organization correlation analysis model, and realizing data visualization analysis through the interaction end. The invention provides a one-stop analysis platform for rice cell heterogeneity research, functional gene mining and molecular breeding.

The invention discloses a water heavy metal concentration simulation and prediction method, which comprises the following steps: integrating original monitoring data, hydrodynamic data, total suspended solids, image remote sensing data and human activity data, and generating a multi-source cleaning sequence data packet; executing cross-modal adsorption capacity estimation by using image remote sensing data in the data packet, inferring particle chemical composition and adsorption isotherm parameters from image textures, and generating a capacity feature packet containing an adsorption capacity upper bound; time-varying travel time is calculated based on the hydrodynamic data and the human activity data, causal alignment is performed on the capacity feature packet and the upstream signal, and a travel time alignment feature packet is generated; and in combination with metal fingerprint parameters, applying an adsorption capacity upper bound as a physical constraint on a form distribution constraint head, explicitly decoupling and predicting the form, and generating a prediction result packet. According to the method, the hydrodynamic physical mechanism and the particle adsorption chemical mechanism are deeply coupled, and the prediction precision and the physical consistency of the model under the unsteady state condition are improved.

The invention discloses a distribution transformer state evaluation method, system and device and a storage medium, and the method comprises the steps: obtaining current and voltage long sequence data through high-frequency sampling, and forming a high-quality data set through expert labeling; signal preprocessing is realized by adopting dual-channel adaptive wavelet packet denoising and cross-correlation sinchinger interpolation phase correction; constructing an asymmetric convolution pyramid to extract current high-frequency and voltage low-frequency multi-scale features, and performing cross-modal fusion by using compressed multi-head attention and a channel-space double-gating mechanism; dynamic random depth regularization is introduced, and a time domain loss function in a fault sensitive period is focused, so that high-accuracy and low-delay real-time state evaluation is realized; the capacity of capturing and recognizing early weak fault features of the distribution transformer in a complex operation environment is remarkably enhanced, and therefore more sensitive and more accurate state early warning is achieved.

The invention relates to the technical field of bioinformatics, and provides a crop whole genome phenotype prediction method and system fused with an environmental indicator gene, and the method comprises the following steps: collecting re-sequencing data, and carrying out genome comparison to obtain variation site data; performing whole genome association analysis by using the variation site data to obtain phenotype association site information; carrying out gene expression quantity measurement on samples of the crop population material in different environments to obtain gene expression quantity data; performing differential expression analysis on the gene expression quantity data to screen environmental indicator genes to obtain an environmental indicator gene set; constructing a phenotype prediction model of double-branch fusion; and predicting a to-be-predicted material through the phenotype prediction model to obtain phenotype prediction results for different environments. According to the method, environmental factors are incorporated into the whole genome selection model, so that the phenotype prediction precision in different environments is improved.

The invention discloses a resource allocation method of a vehicle-mounted information system and a program product. The method comprises the following steps: acquiring historical load time sequence data of a vehicle-mounted information system in a historical time period, and determining load prediction data of the vehicle-mounted information system in a prediction time period according to the historical load time sequence data and a load prediction model; under the condition that the load prediction information meets a preset load condition, determining user operation prediction data of the vehicle-mounted information system in the prediction time period, and determining a user interaction task according to the user operation prediction data; the method comprises the steps of obtaining a system task needing to be executed by the vehicle-mounted information system within a prediction time period, determining the system task and a user interaction task as a to-be-executed task, determining a priority score of the to-be-executed task, and allocating resources to the to-be-executed task according to the priority score so as to realize dynamic allocation of task resources.

The invention provides a spatial domain identification method based on data interpolation and cell type deconvolution, and belongs to the technical field of bioinformatics. In order to solve the problems that gap information between adjacent points cannot be utilized in low-resolution spatial transcriptome data and prior information of cell types in a tissue space structure level cannot be fully integrated in a traditional method, the method comprises the following steps: acquiring a spatial transcriptome data set and a single-cell RNA sequencing data set, and performing data preprocessing on the acquired data sets; and carrying out data interpolation on the preprocessed spatial transcriptome data, and carrying out cell type deconvolution in combination with single-cell RNA sequencing data. And constructing a deep learning model based on the graph convolutional network. And training a deep learning model according to gene expression information, spatial position information and cell type information of the spatial transcriptome data after cell type deconvolution by using a self-supervised contrast learning strategy. And performing spatial domain identification on the to-be-detected data based on the trained model.

The invention relates to the technical field of ecological system service function evaluation. The invention relates to a frequency offset correction algorithm in HPLC (High Performance Liquid Chromatography) + HRF (High Radio Frequency) dual-mode communication. The method comprises the following steps: S1, collecting phase sequence data from an HPLC communication link, collecting frequency offset estimation data and corresponding confidence data from an HRF communication link, and adaptively adjusting a preset detection threshold value of the phase sequence data according to the confidence data; s2, performing pulse abnormal sample point detection and elimination on the phase sequence data according to the adjusted preset detection threshold value; according to the method, the detection threshold is dynamically adjusted through the confidence data, so that the anomaly detection process can be matched with the current interference environment in real time, the integrity of the phase sequence data is maintained, and the problem of insufficient adaptability of a fixed detection threshold in an electromagnetic interference dynamic change scene is solved; and a more reliable phase data basis is provided for subsequent interpolation compensation and frequency offset estimation, so that the frequency offset correction precision of the dual-mode communication system is improved.

The invention discloses a relay protection method and device based on time sequence data prediction, and the method comprises the steps: collecting the operation sequence data of a main network power transformation relay protection device in real time through a wireless sensing node, and transmitting the operation sequence data to a data processing layer through a Zigbee wireless network protocol; historical data features are extracted based on a sliding window method, abnormal data are recognized and removed through a quartile method, and missing data are filled up through a linear interpolation algorithm; constructing a CNN-GRU prediction model fused with the attention mechanism, dynamically weighting key features by the attention mechanism, inputting the preprocessed data into the model for training, and generating an operation state evaluation value based on a prediction result, and when the evaluation value exceeds a preset threshold, triggering a multi-level early warning mechanism. And the prediction precision is verified by using a decision coefficient R2, the R2 is required to be greater than or equal to 0.95, and model parameters are dynamically updated to adapt to the working condition change of the power grid.

The invention discloses an interpretability analysis and decision sharing verification method and system for a rectal cancer prognosis model, and relates to the field of medical artificial intelligence interpretability. The method comprises the following steps: carrying out gradient weighting class activation mapping analysis on a prognosis model to generate an image thermodynamic diagram; calculating the contribution degree of the multi-modal features by using an SHAP interpreter; an integrated visual interface is constructed, and patient data, model prediction and the explanation result are presented to a doctor together; the doctor performs independent risk assessment based on the interface information; finally, decisions of doctors and the model are compared, and model auxiliary efficiency is evaluated. Through a doctor-model decision sharing verification mechanism which is explained and innovated in a multi-level mode, the transparency and clinical credibility of the complex AI prognosis model are remarkably improved, the value of time sequence data in dynamic risk assessment can be verified, and clinical landing application of the AI model is powerfully promoted.

The embodiment of the invention provides a model training method, a carbon emission prediction method, a device and equipment. The method comprises the following steps: firstly, obtaining carbon emission sample sequence data; then, preprocessing the carbon emission sample data sequence to obtain preprocessed carbon emission sample sequence data; further, according to the preprocessed carbon emission analysis sample sequence data, determining a time sample feature and an interaction feature sample sequence; and then. Processing the sample time feature and the preprocessed power consumption sample sequence data to obtain a sample time feature component and a sample power consumption feature component; and finally, inputting the sample time characteristic component, the sample power consumption characteristic component, the interaction characteristic sample sequence and the preprocessed carbon emission sample sequence data into an initial Transform model for optimization training, and obtaining an improved target Transform model. In this way, the prediction precision of the prediction model and the generalization ability of the model are improved, and therefore accurate prediction of carbon emission data is achieved.

The invention relates to the technical field of data processing, and discloses a research report text extraction optimization method and device, equipment and a medium, which can be applied to the financial field, and the method comprises the following steps: obtaining research report text data, and processing to obtain a segmented text fragment set; establishing a fragment vector index database by utilizing the segmented text fragment set; receiving a query request based on the fragment vector index database and processing the query request to obtain candidate fragment sequence data; and processing the candidate fragment sequence data to obtain context data, injecting a cue word template to generate a target cue word, inputting the target cue word into a language model for reasoning, and outputting a text optimization result. In the invention, aiming at the problem that the output of the existing research and report text is lack of standardization, the target cue word can be generated by utilizing the context data obtained by calculation, and is input into the language model for reasoning, and finally the text optimization result is output, so that a multi-level abstract and key information summary can be quickly generated, the manual reading cost is reduced, and the efficiency is improved. The working efficiency of personnel is improved.

The invention discloses a method for evaluating enhancement and release seedling resources based on environmental DNA polymerization analysis. The method comprises the following steps: carrying out gridding partition on a target water area, collecting a water sample through a designed sampling scheme, and carrying out DNA extraction and high-throughput sequencing to obtain species sequence information of each sampling point. Sequencing data is subjected to species identification by using a bioinformatics method, released species are identified, a spatial abundance model is established, and a preliminary distribution map is generated. And establishing a DNA degradation kinetic model in combination with water area environmental parameters, and carrying out reverse correction on abundance distribution. And through a resource inversion model coupled with hydrodynamics, analyzing biomass distribution characteristics and migration laws of the release group, and obtaining a resource evaluation result. And finally, a species environment preference model is constructed based on migration path analysis, an optimal release area is matched in a target water area, a scientific scheme including release point locations, opportunities and quantity is generated, and a whole-process technical support and a decision basis are provided for enhancement and release.

The invention discloses an injection molding equipment anomaly detection method based on a graph neural network, and the method comprises the following steps: collecting multi-dimensional monitoring data in the operation process of injection molding equipment, and carrying out the normalization, denoising and missing value filling, and obtaining structured monitoring sequence data; dividing the injection molding production process into a plurality of process stages based on timestamps and process stage labels in the structured monitoring sequence data, and constructing a corresponding stage sub-graph set; based on the stage sub-graph set, the energy transfer path, the material flow path and the historical abnormal propagation path, constructing a nested graph structure; inputting the nested graph structure into an improved dynamic graph convolutional network for feature extraction to obtain a time-aware graph embedding representation; and performing anomaly detection on the time perception type graph embedded representation to generate an anomaly detection result of the injection molding equipment. According to the method, nested graph structure modeling and the dynamic graph convolutional network are adopted, and high-precision detection and rapid positioning of the abnormal state of the injection molding equipment are achieved.

The invention discloses an internal medicine drug curative effect evaluation method based on big data analysis of an electronic health record, and the method comprises the steps: extracting basic health data, diagnosis and treatment time sequence data and drug intervention data from the electronic health record, and carrying out the time-space alignment to generate a dynamic feature set; subgroups are obtained based on disease typing standard hierarchical clustering, and historical data and real world data are fused through transfer learning to construct a subgroup curative effect reference matrix; collecting data after medication in real time, and generating an evaluation vector containing short-term physiological response, middle-term symptom improvement and long-term prognosis risk through deep learning; dynamically matching the evaluation vector with the reference matrix, and introducing an individual weight coefficient to correct deviation; taking the deviation correction value as input, constructing a self-adaptive evaluation model through reinforcement learning, and performing iterative optimization; and generating an individualized report containing the curative effect level, the medication suggestion and the risk early warning, and quantifying the curative effect level through a fuzzy comprehensive evaluation method. According to the method, individual differences are accurately captured, full-cycle dynamic evaluation is realized, and the curative effect evaluation accuracy and the clinical decision-making efficiency are improved.

The invention relates to the technical field of biological information, in particular to a translation efficiency prediction method, a screening method, a model training method and electronic equipment. The translation efficiency prediction method is realized based on a translation efficiency prediction model, and the translation efficiency prediction model comprises an embedding module, a coding module and a prediction module. The method comprises the following steps: acquiring sequence data of messenger ribonucleic acid; extracting features of the sequence data through the embedding module to obtain feature representation of messenger ribonucleic acid; the feature representation is input into the coding module, the input feature representation is subjected to convolution processing through the coding module, a convolution result is coded based on an attention mechanism, and the updated feature representation is output according to a coding result; and performing translation efficiency prediction on the updated feature representation through the prediction module to obtain a translation efficiency prediction result. According to the embodiment of the invention, the translation efficiency of the messenger ribonucleic acid can be accurately predicted with high throughput, the screening efficiency is improved, and the screening cost is reduced.

Provided herein are technologies for framing and evaluating biological sequence-based analysis tasks in a unified, natural-language-based, text in and text out format. Among other things, methods and systems of the present disclosure provide machine-learning technologies for combining biological sequence data, representing, for example, DNA, RNA, and protein sequences, with natural language, conversational style prompts that set out particular analysis tasks to be performed on the biological sequence data. This approach, for example, allows complex analysis tasks, including, but not limited to, identification of various sequence modifications, genes, and regulatory elements in DNA sequences, and quantification of properties such as degradation propensity of RNA and protein stability, to be input to a machine learning model in a uniform text-based format and for output to be generated in a same, unified, text-based format.

The present invention relates to the field of tissue-derived cell identification, in particular to a method for identifying tissue-derived cells in the body fluid based on single-cell sequencing technology. In the present invention, on the basis of high-throughput single-cell RNA sequencing data of cell samples obtained from body fluids, by means of reference component analysis (RCA), expression of known tissue-related marker genes, and a tissue-derived cell prediction model constructed based on logistic regression, the specific identification of tissue-derived cells in body fluids is achieved.

The invention relates to a product recommendation optimization method based on user portrait dynamic updating, which comprises the following steps: acquiring user behavior sequence data, and analyzing based on behavior consistency indexes including time interval, intention change rate and interest inversion degree; when it is detected that the behavior path has a fault, frequent switching or reverse direction, revoking the update of the path to the user portrait and marking the path as a low credible behavior segment; before user portrait updating is triggered, the portrait mutation risk is judged according to the feature distribution difference between the current behavior and the historical portrait, the behavior uniqueness and the conversion closed loop condition, and when mutation is found, portrait updating is frozen and recorded to a to-be-verified sequence; based on self-consistent state recognition of a historical behavior track, local increment updating is carried out on a staged portrait dimension consistent with a current behavior; and before the recommendation result is generated, performing consistency evaluation on the recommendation target and the user portrait intention, if the alignment degree is lower than a preset threshold value, adjusting or reconstructing the recommendation content, and correcting the deviation label according to continuous negative feedback.