68 results about "Concept drift" patented technology

The invention discloses a multi-head time sequence intelligent risk control method and system based on weighted trend and fluctuation, and relates to the technical field of financial risk management. Comprising the following steps: S1, collecting multi-channel time sequence data in real time, and carrying out data preprocessing; s2, calculating the global time weight, quantifying the multi-scale fluctuation stability of the channel pair, and judging the asynchronous alignment degree of the channel pair; s3, extracting an effective frequency band interval, calculating frequency domain characteristic parameters of the signal, evaluating frequency domain energy phase characteristics of a channel signal, and quantifying fluctuation states of a channel under different scales; s4, constructing a sparse coupling relation graph, quantifying an edge weight in the sparse coupling relation graph, and obtaining a network average coupling weight; and S5, evaluating the dynamic evolution characteristics of the channel risk state, and generating risk trend prediction and control suggestions. The problem that risk control accuracy is affected due to the fact that trend and fluctuation feature extraction of multi-source heterogeneous high-noise multi-head time series data is unstable under the condition of concept drift and multi-scale coexistence is solved.

The invention discloses an encrypted traffic adaptive update classification method and system oriented to an open network environment. The method comprises the following steps: firstly, extracting endogenous semantic features and exogenous environment features based on a causal decoupling mechanism, and stripping an environment confusion factor through anti-fact disturbance and invariance constraint to obtain invariant semantic representation; constructing a macroscopic drift state vector representing a network situation, and inputting a trained meta-learning super-network intelligent decision adaptive control hyper-parameter; performing cross-modal element calibration by utilizing large language model thinking chain reasoning, and calculating the subspace direction consistency of an instantaneous gradient vector of a candidate sample and a category optimization trajectory prototype so as to screen credible samples; and in combination with the capacity-limited playback queue, gradient orthogonal projection constraints are introduced to update low-rank adaptation layer parameters. According to the method, the concept drift problem is solved through causal decoupling and meta-learning decision, forgetting prevention is achieved through orthogonal projection updating, and the online adaptability and robustness of the model in the open environment can be improved without additional manual annotation.

The invention discloses an interference prediction method and device fusing an incremental KD tree and a graph convolutional neural network. The method comprises the following steps: S1, extracting time sequence features of sparse historical interference samples; s2, constructing a dynamic graph structure according to an incremental KD tree model to capture a spatial relationship among sparse historical interference samples; s3, extracting a local sub-graph based on the spatial relationship between sparse historical interference samples, and inputting the local sub-graph into the graph convolutional neural network to obtain a prediction result; s4, obtaining a new sparse historical interference sample, and carrying out incremental parameter updating on the local subgraph based on a prediction result to obtain an incremental parameter updating result; and S5, judging whether a concept drift problem exists in the confrontation process, if the concept drift problem exists, adopting an online updating mechanism to adapt to environmental changes, then returning to S1, and if confrontation is finished, exiting. According to the method, the problem that interference behavior time change and space correlation are difficult to describe at the same time in a traditional method is solved, and the problem of calculation efficiency of a traditional graph model in a dynamic scene is solved.

The invention discloses a flight landing time prediction method based on machine learning, and relates to the technical field of flight pre-judgment, and the method comprises the steps: carrying out the preprocessing of data based on historical flight data and historical meteorological data, and outputting feature sample data; performing data cleaning and standardization processing on the feature sample data, and outputting a training data set and a test data set in combination with actual landing and planned landing moment information of historical flights; establishing a landing prediction model, and performing optimization by combining reinforcement learning and a generative adversarial network; using the average absolute percentage error to test the landing prediction model and output a trained landing prediction model; and outputting a dynamically updated landing prediction model by adopting a concept drift detection algorithm, online gradient lifting and adaptive weighted fusion. Through combination of the deep Q network and multi-agent reinforcement learning, the prediction result is optimized in real time, and dynamic adjustment can be performed according to changes of actual flights.

The invention relates to the technical field of building automation and intelligent operation and maintenance, in particular to a building equipment fault positioning method based on a sensor network and a layer model, and the method comprises the steps: data collection and preprocessing, dual-time scale concept drift detection, dynamic model management and adaptive correction, and fault positioning and output. According to the method, system changes with sudden change and slow change can be captured at the same time through the double-time-scale design, the dynamic model pool and the historical concept reproduction recognition mechanism enable the system to memorize and reuse the historical operation mode, false alarms caused by periodic changes such as season switching are avoided, and the reliability of the system is improved. The problems that a traditional static or simple online updating model is poor in adaptability in a dynamic environment and the accuracy rate is reduced are fundamentally solved.

The invention relates to a dynamic semantic evolution tracking and concept drift adaptive updating method and system, and the method comprises the steps: obtaining scene demand information, and determining an information collection frequency and a historical data time window; obtaining target multi-source data and historical multi-source data; constructing an initial map; semantic feature vectors are extracted, and similarity matching is carried out; when the semantic similarity is lower than a similarity threshold value, updating the initial graph to generate an incremental graph; when it is monitored that the context distribution difference value of the nodes in the incremental atlas is larger than a drift threshold value, it is judged that a concept drift event exists, and a drift response strategy is triggered; according to the method, continuous tracking of semantic evolution and self-adaptive updating of concept drift are realized by constructing the initial map, matching semantic features, generating the incremental map and triggering the response strategy; the method has the advantages of tracking a semantic evolution process, adaptively updating the knowledge graph, responding to a concept drift event, and improving semantic analysis accuracy and real-time performance in a dynamic scene.

Approaches herein relate to model decay of an anomaly detector due to concept drift. Herein are machine learning techniques for dynamically self-tuning an anomaly score threshold. In an embodiment in a production environment, a computer receives an item in a stream of items. A machine learning (ML) model hosted by the computer infers by calculation an anomaly score for the item. Whether the item is anomalous or not is decided based on the anomaly score and an adaptive anomaly threshold that dynamically fluctuates. A moving standard deviation of anomaly scores is adjusted based on a moving average of anomaly scores. The moving average of anomaly scores is then adjusted based on the anomaly score. The adaptive anomaly threshold is then adjusted based on the moving average of anomaly scores and the moving standard deviation of anomaly scores.

The invention provides a self-adaptive network security situation awareness method and device combined with online learning, and the method comprises the steps: collecting real-time network state data and system load index data of a preset data source, carrying out the preprocessing of the data to form a multi-modal time sequence segment, inputting a pre-training time sequence data prediction model activated by employing Monte Carlo Dropout through a sliding window, and carrying out the prediction of the real-time network state data and system load index data. The method comprises the following steps of: calculating a prediction value of a next time period, outputting a prediction value and an uncertainty quantity of the next time period, calculating a threat probability through historical residual probability distribution fitting, realizing double-index risk assessment based on a preset threshold interval system, dividing into three types of states, and finally, respectively triggering online learning, configuration maintenance or intervention disposal flow for different states. According to the method, by introducing uncertainty quantized double-index evaluation and state-driven online learning closed loop, crossing of network security situation awareness from static detection to dynamic self-adaption is achieved, and the two core problems of insufficient real-time performance and concept drift in an edge computing scene are effectively solved.

This invention discloses a charger life prediction system based on big data analysis, belonging to the field of computer-aided design technology. It includes a data fusion unit, a benchmark modeling unit, a drift monitoring unit, a risk assessment unit, and a strategy generation unit. The data fusion unit generates structured feature vectors and sends them to the benchmark modeling unit. The benchmark modeling unit establishes a benchmark life prediction model based on the structured feature vectors and sends the benchmark life prediction model to the drift monitoring unit. The drift monitoring unit generates a conceptual drift index and sends it to the risk assessment unit. This invention profoundly reflects the inherent attributes and operating conditions of the equipment, and its accuracy and reliability far exceed those of traditional models relying on a single data source.

The invention discloses a load prediction method based on STL decomposition and concept drift detection, and relates to the technical field of power system prediction, and the method comprises the following steps: collecting historical power load data of a power system, checking whether there is a missing value in the collected data, and carrying out the interpolation processing of the missing value, decomposing the processed load data into three components, namely a trend component, a period component and a residual error component through STL (Standard Template Library) decomposition; secondly, carrying out concept drift detection on the trend component through an ADWIN algorithm, capturing load characteristics changing along with time, and dynamically adjusting the size of a time window trained by a prediction model; and under the dynamically adjusted time window, training a Transform-based prediction model for the three components, and adding prediction results of the three components to realize load prediction. According to the method, STL decomposition is carried out on the load data, the load data is decomposed into trend, period and residual components, refined model training is carried out for different components, and finally high-precision load prediction is realized.

The invention relates to the technical field of network security and information security, and particularly discloses an internal threat detection method based on entity behavior analysis, which comprises the following steps of: firstly, converting a multi-source heterogeneous log into a unified event semantic vector, and constructing a double-flow neural network architecture: extracting a long-term portrait vector from a long-period history by utilizing LSTM (Long Short Term Memory), and in the other path, a short-term intention vector is extracted from the neighbor session by utilizing Transform. And the short-term intention is projected to the semantic space of the long-term portrait to carry out consistency residual calculation, so that the hidden slow poison attack is effectively identified. In order to solve the problem of misinformation caused by feature dimension stability difference, a variance estimation and anisotropy measurement mechanism is introduced, portraits are modeled as probability distribution, variance vectors are utilized to carry out reliability weighting on residual errors, and global entropy regularization is combined to dynamically adjust penalty sensitivity, so that the probability distribution of the portraits is determined. Therefore, legal concept drift and malicious behaviors can be accurately distinguished in a complex dynamic scene.

This invention discloses a method and system for identifying encrypted traffic based on spatiotemporal features and semantic alignment. The method first extracts the spatial and temporal feature sequences of the network flow, and uses a byte-pair encoding algorithm to convert the spatial packet length into discrete symbols. Then, the discrete symbols and temporal features are mapped to a high-dimensional space and fused together. A global spatiotemporal feature vector is extracted through a network using a concatenated one-dimensional convolution and multi-head self-attention mechanism. Next, the text semantic bullseye matrix of fine-grained behaviors of various known applications is obtained offline from a large language model. Finally, the similarity between the spatiotemporal features and the text bullseye is calculated, and a multi-instance learning max-pooling mechanism is introduced for dynamic routing. Based on this, a contrastive learning loss function optimization model is constructed or cross-modal inference is performed. This invention completely overcomes the conceptual drift problem caused by changes in encrypted features, achieving extremely high generalization accuracy and feature interpretability across generations.

The invention relates to the technical field of computer security, in particular to a dynamically updated Android malicious software continuous learning detection method, which comprises the following steps of: acquiring and sequencing historical Android application samples according to time to construct a training set, and training a hierarchical comparison classifier; new applications are collected regularly to form a to-be-tested batch, and the current classifier is used for prediction; for each sample in the to-be-detected batch, executing pseudo loss uncertainty calculation to obtain an uncertainty score of the sample; according to the score, selecting a predetermined number of most uncertain samples for labeling, and obtaining a real label; adding the new labeled sample into the training set, and performing incremental training by adopting a hot start mode based on the weight of the current classifier to obtain an updated classifier; and circularly executing, and detecting the new application in the next period by using the updated classifier. According to the method, the labeling cost can be remarkably reduced, the concept drift can be effectively coped, and efficient, stable and continuous malicious software detection is realized.

The invention discloses a self-adaptive clean coal ash content prediction method during testing of hidden space concept drift evaluation, belongs to the technical field of computers, and aims to improve the generalization ability and robustness of an artificial intelligence model under the non-stationary working condition of dense medium coal separation. According to the method, a time shift driven sample-offset sample pairwise modeling mechanism is constructed, a double-input hidden space representation network with LSTNet as a backbone is designed, and direct estimation of the dynamic offset of the ash content of the clean coal is realized based on the difference of representation at two moments. According to the method, the target offset loss and the hidden space reconstruction constraint are fused, the composite loss function is constructed, and the excessive dependence of the model on single data distribution is reduced by enhancing the stability of the representation space structure; according to the adaptive strategy during testing, concept drift under the new working condition can be evaluated and compensated under the condition of not depending on large-scale retraining, and the accuracy, stability and reliability of ash content prediction under the unknown or extreme working condition are improved.

The invention discloses a wind turbine generator reliability rating method and system, a medium and computer equipment. The method comprises the steps of data preparation and alignment; aRIMA / seasonal ARIMA modeling is carried out, and characteristic indexes are obtained; aRIMA output is structured into enhanced features available for a tree model, and the enhanced features are fused with time coding, lagging and rolling statistics and unit static parameters; a CART or gradient boosting tree is adopted to output'reliability grade / risk score 'and provide an interpretable rule path, and online updating and auditing are carried out to cope with concept drift and seasonal variation. The reliability rating method which is more stable, explainable and audible is formed, and the accuracy rate, the recall rate and the cross-seasonal stability are remarkably improved; and low-cost operation can be realized through lightweight feature calculation and periodic updating in an edge deployment environment.

This invention discloses an adaptive update classification method and system for encrypted traffic in open network environments. The method first extracts endogenous semantic features and exogenous environmental features based on a causal decoupling mechanism. It then removes environmental confusion factors through counterfactual perturbations and invariance constraints to obtain invariant semantic representations. Next, it constructs a macroscopic drift state vector representing the network situation and inputs it into the intelligent decision-making adaptive control hyperparameters of a trained meta-learning hypernetwork. Cross-modal meta-calibration is performed using a large language model's thought chain reasoning, and the consistency between the instantaneous gradient vector of candidate samples and the subspace direction of the category optimization trajectory prototype is calculated to screen credible samples. Finally, combined with a capacity-constrained replay queue, gradient orthogonal projection constraints are introduced to update the parameters of the low-rank adaptation layer. This invention solves the concept drift problem through causal decoupling and meta-learning decision-making, and uses orthogonal projection updates to prevent forgetting, improving the model's online adaptability and robustness in open environments without additional manual annotation.

The application discloses a stream data self-adaptive classification method and system based on artificial intelligence, relates to the technical field of artificial intelligence and machine learning, and comprises the following steps: acquiring stream input data and a classification model of a current time step; using the model to obtain a relative position feature representing a decision boundary and to calculate an instantaneous gradient vector; updating the feature to a feature sequence, determining a target update resistance threshold based on the distribution discrete degree of the feature in the time sequence dimension; attenuating a historical accumulated gradient vector and superimposing the instantaneous gradient vector to obtain a target accumulated gradient vector; then judging whether the length of the vector is greater than the resistance threshold; if not, keeping the model parameters unchanged and retaining the accumulated gradient to the next time step; if yes, updating the classification model parameters based on the difference and resetting the target accumulated gradient vector; and the application effectively removes transient random interference and real concept drift, and breaks the noise resistance and sensitivity bottleneck.

The invention discloses an open intelligent algorithm adaptive combination method and system based on meta learning and a dynamic pipeline. According to the method, by introducing an algorithm directed acyclic graph construction mechanism of a type system and an adapter, dynamic integration of heterogeneous components in an open algorithm library is realized, and the limitation that a traditional AutoML system algorithm combination space is closed and depends on a preset process is overcome. Environment perception and a performance monitoring closed loop during operation are integrated into a decision process, and a dynamic adjustment mechanism based on deviation triggering is designed, so that the system has the capability of coping with concept drift and resource fluctuation after deployment. By constructing and continuously updating the knowledge base recording the task-environment-performance mapping relation, the system has the ability of making decisions based on historical experience. According to the meta-learning mechanism, when the system processes similar tasks, repeated search overhead can be reduced, continuous evolution of performance is achieved, and a feasible technical path is provided for solving the continuous learning problem of a universal intelligent system.

The application discloses a cloud-edge collaborative and federated learning-based electroencephalogram prediction model self-evolution system and method. The system comprises an edge device and a cloud server cluster: the edge device collects physiological signals of a user and runs a first machine learning model, and uploads desensitized data of a marked event asynchronously; the cloud server performs continuous wavelet transform on received one-dimensional electroencephalogram signals to generate two-dimensional time-frequency images, trains a second deep learning model by using the images, generates updated weights, and completes silent updating by issuing the updated weights to the edge device. The application solves the model concept drift problem through a cloud-edge collaborative architecture, realizes continuous evolution of a prediction model, improves feature extraction depth and generalization ability through time-frequency conversion and a multi-specialist sub-network architecture, and effectively protects user privacy through a data desensitization and asynchronous uploading mechanism.

The application provides an integrated learning classification method and device based on an online sequential pre-interference layer, and relates to the technical field of data stream classification.The application obtains historical vibration sequential data and real-time online vibration sequential data; designs an integrated classifier composed of multiple heterogeneous base classifiers; the base classifier is a four-layer feedforward neural network model, an interference layer is added between the input layer and the hidden layer of the OS-ELM network to realize nonlinear kernel mapping of online sequential samples; each base classifier of the integrated classifier is initialized and learned sequentially based on the obtained historical vibration sequential data and online vibration sequential data; and the newly obtained online vibration sequential data is classified and predicted based on the integrated classifier.The application can effectively solve the concept drift and class imbalance problems in dynamic data streams, and improve the recognition ability of the model for minority class samples and the adaptability of the model to data distribution changes.

The present application provides an electroencephalogram sound stimulation evaluation prediction and adaptive output method and system, which is based on an individual-group bidirectional migration collaborative mechanism of shared feature space and distribution alignment constraint to fuse individual specificity and group commonality knowledge, forming a mechanism to overcome the individual differences of electroencephalogram; the method also uses an adaptive decision fusion mechanism based on prediction uncertainty evaluation to ensure output stability when the prediction model reliability fluctuates, and a model dynamic online updating mechanism based on concept drift detection to adapt the prediction model to the long-term changes of the user state, forming a mechanism to improve accuracy and robustness; the present application can effectively overcome the individual differences of electroencephalogram, realize high-precision system prediction and output, and has robustness.

Embodiments of the invention provide an electrical load prediction method and apparatus. The method comprises the steps of obtaining multi-source data; preprocessing the multi-source data to obtain preprocessed multi-source data; constructing load characteristics based on the preprocessed multi-source data; inputting the load characteristics into a load prediction model selected based on a prediction task, and outputting a predicted load prediction result by the load prediction model; and when concept drift occurs in the multi-source data or the prediction result deviation is greater than a set deviation threshold value, updating the load prediction model. The prediction accuracy and reliability can be improved.

The application provides a time-aware instant software defect prediction method, device and readable storage medium, wherein the time-aware instant software defect prediction method comprises the following steps: acquiring real change samples; inputting the real change samples into a generation model based on a conditional generative adversarial network to acquire simulated change samples; adjusting the optimization strategy of an instant software defect prediction intelligent agent based on a flexible action-evaluation algorithm according to the time sequence of changes by using the real change samples and the simulated change samples; and predicting defects of software by using the trained instant software defect prediction intelligent agent. The generation model based on the conditional generative adversarial network is used to acquire simulated software changes, and the optimization strategy of the instant software defect prediction intelligent agent is adjusted in a timely manner according to the time sequence of changes by using the real change samples and the simulated change samples after each time step ends, by using the reward result and the current state, so that the concept drift problem is solved.

The application discloses an unknown threat identification method based on adaptive snapshot integration, which is applied to instant fine classification of network encrypted traffic and can identify unknown threats caused by concept drift. The method comprises an encrypted traffic preprocessing and feature extraction module, an encrypted traffic time sequence feature neural network classification module, a new category traffic discrimination module, a snapshot integration module and a transfer learning module. The method can cope with increasing unknown abnormal traffic, give unknown abnormal traffic clustering results for experts to mark, effectively solve the concept drift and category imbalance problem, and thus guarantee the reliability of abnormal traffic detection.

The application discloses a kind of data under the whole life cycle meta-learning anomaly detection method and system, comprising: according to the original data stream of pre-set sampling frequency acquisition, obtain original data set with stage label;Null-zero cleaning is carried out to it, and the output rough screening data set;According to life cycle stage, it is divided into several meta-learning tasks, and the model-independent meta-learning framework based on gradient is used to carry out outer loop and inner loop iteration with task as unit, obtain stage-independent initialization parameter, generate stage adaptive anomaly detection model;Calculate anomaly score to unlabeled query set and to-be-measured stream data, and recheck in combination with median deviation threshold, output final anomaly data set.The application is divided into meta-learning task by whole life cycle stage, and the model-independent meta-learning framework based on gradient is used to solve the problem that data characteristics are significantly different in different stages of data whole life cycle, anomaly detection model generalization ability is insufficient and difficult to adapt to interstage concept drift.

The invention discloses a satellite telemetry data online anomaly detection method and device based on an optimal segmentation isolated forest. The method comprises the following steps: in an offline stage, training a BSIForest model by using satellite historical telemetry data, adopting a two-dimensional optimal hyperplane division strategy for the model during isolation tree node division, and taking a midpoint of a subinterval with the lowest histogram frequency as a segmentation point, thereby effectively inhibiting a'ghosting 'phenomenon occurring between clusters and a parallel axis region in a traditional algorithm; in the online stage, abnormal scores of telemetry data are calculated sample by sample, and continuous abnormal windows are accurately recognized and output in combination with a threshold value and a sliding window local abnormal rate mechanism. Through incremental updating of a training strategy, the detection model can cope with concept drift caused by orbit environment change. It has been verified that the single-sample anomaly detection duration is less than 0.2 millisecond on the basis of telemetry data of a real satellite power supply system, no missing report or false report exists in an abnormal sequence, and the real-time and high-reliability anomaly detection requirement of a satellite is met.

The application discloses a kind of prediction and abnormality detection algorithm and application for time series data, belong to time series data abnormality detection technical field, comprising: the original time series data of input is preprocessed, and the standardized sample segment is generated;Multi-scale feature fusion time series prediction model is constructed and trained;Using the trained time series prediction model, the standardized sample segment is multi-step forward prediction, and the predicted sequence is obtained;Based on the predicted sequence and the corresponding actual observation sequence, dynamic residual threshold is calculated, and real-time abnormality detection and marking are carried out according to the threshold.The application solves the problem that the existing complex space-time correlation feature is not sufficient, cannot dynamically adapt to data change and model cross-scene migration and online learning ability is weak.The application fully captures complex nonlinear space-time correlation features, improves the adaptability of abnormality detection to data nonstationarity and concept drift, realizes online self-adaptation and threshold drift tracking, and improves the practicability and long-term robustness of the algorithm.

The invention relates to a charger life prediction system based on big data analysis, which belongs to the technical field of computer aided design and comprises a data fusion unit, a reference modeling unit, a drift monitoring unit, a risk assessment unit and a strategy generation unit. The data fusion unit is used for generating a structured feature vector and sending the structured feature vector to the reference modeling unit; the reference modeling unit is used for establishing a reference life prediction model based on the structured feature vector and sending the reference life prediction model to the drift monitoring unit; the drift monitoring unit is used for generating a concept drift index and sending the concept drift index to the risk assessment unit, the intrinsic attribute and working condition influence of equipment are deeply reflected, and the accuracy and reliability of the system are far higher than those of a traditional model depending on a single data source.

This invention relates to the field of fault detection technology, and in particular to an industrial fault detection method and system based on dynamic drift sensing and diffusion enhancement. The method includes constructing an unsupervised fault detection DDA-DE model, wherein dynamic drift sensing DDA is used to process industrial data streams to establish a statistical distribution baseline and drift threshold; diffusion-enhanced anomaly detection DE is used to determine initial model parameters and anomaly thresholds; the robustness of the unsupervised fault detection model to concept drift is enhanced based on a diffusion strategy; the proposed industrial-enhanced Mahalanobis distance real-time drift sensing algorithm can more efficiently detect concept drift phenomena, and the parallel design of the drift sensing algorithm and the fault detection classifier achieves collaborative detection of data drift and anomaly events.