7994 results about "Feature (machine learning)" patented technology

In embodiments, systems and methods for improving machine-learning systems are disclosed. In embodiments, a system includes a data pool system that is configured to receive data from a plurality of different data sources and maintain a training data set that is used to train a specific machine-learning model based on the data from the plurality of different data sources. In embodiments, the system further includes a data scoring system that determines a data reliability score corresponding to the new data based on a set of intrinsic features of the new data and a data scoring model, wherein the data pool system selectively adds the new data to the training data set based on the reliability score of the new data. The system also includes a machine learning system that trains the specific machine-learning model based on the training data set.

According to some embodiments, a method is performed by a wireless device for fallback operation of a machine learning (ML) model. The method comprises: transmitting a message indicating a capability of the wireless device for supporting a combination of at least one ML-based feature for a functionality and at least one fallback feature for the functionality to a network node; operating the at least one ML-based feature for the functionality; and operating the at least one fallback feature for the functionality.

The invention provides a flow analysis and threat detection method and device based on machine learning, and the method comprises the steps: collecting a real-time flow data package of a target network environment, carrying out the protocol analysis and session recombination, and generating a real-time flow feature data set containing multi-dimensional flow features; loading a pre-trained multi-level threat classification model, inputting the real-time traffic feature data set into a feature extraction layer of the model, carrying out normalized coding on traffic features of corresponding dimensions through feature coding channels, generating a real-time feature vector sequence, inputting the real-time feature vector sequence into a primary classifier of the model, and classifying the real-time traffic features according to the real-time feature vector sequence; and performing abnormal probability calculation and cluster division on the real-time feature vector sequence through a mixed detection unit, outputting a primary threat tag and an abnormal confidence coefficient corresponding to each real-time feature vector, inputting the primary threat tag and the abnormal confidence coefficient into an aggregation classifier, performing dynamic weighted aggregation, and generating a comprehensive threat score so as to judge whether a threat response strategy is triggered or not. According to the invention, the accuracy and timeliness of threat detection in a complex network environment can be improved.

The invention provides a livestock breeding risk intelligent assessment method and system based on multi-source data fusion, and the method comprises the steps: collecting livestock individual vital sign data, breeding environment parameters, management behavior data and risk-related historical data through Internet of Things equipment, and carrying out the data preprocessing to form a standardized multi-source data set; extracting risk features of individual, group and environment levels based on the data set, and fusing the risk features to form a multi-dimensional risk feature library; utilizing machine learning to construct a differentiated risk assessment model; analyzing the incidence relation between the risk factor and the actual event through the Bayesian network to calibrate the model; realizing livestock risk grade dynamic division and early warning based on the calibrated risk scoring system; and finally, generating intervention suggestions for risk quantitative evaluation, risk prevention and control decision and loss evaluation. According to the method, accurate evaluation of livestock breeding risks is realized, decision support is provided for breeding safety management, and the method has relatively high application value.

The invention discloses a grassroots power supply enterprise compliance risk intelligent system and method based on big data analysis. The data acquisition unit is used for acquiring business operation data, historical violation records and policy and regulation update data of basic power supply enterprises to form a unified compliance data set. And the natural language processing unit performs text word segmentation and correlation analysis on the policy and regulation and violation record data, extracts key risk factors and labels compliance risk labels. And the risk feature construction unit performs multi-dimensional feature fusion on the business operation data and the compliance risk label data to generate a feature matrix for risk identification. And the intelligent risk assessment unit performs real-time analysis on the feature matrix by using a pre-trained machine learning model, identifies compliance risk categories and levels, and generates early warning information. According to the method, the accuracy of compliance risk identification is improved by using big data analysis and an intelligent algorithm, the compliance management cost of basic-level power supply enterprises is reduced, and the operation safety and compliance of the enterprises are improved.

The invention discloses a sewage plant total nitrogen concentration real-time prediction and process regulation and control method based on an optimization integration algorithm, and belongs to the technical field of environment monitoring and treatment. According to the method, sewage plant data are monitored and collected, a sliding window and a time sequence are combined to analyze and clean the data and reconstruct features, total nitrogen concentration strong correlation variables are screened, data quality is standardized and optimized, a plurality of machine learning algorithms are adopted to construct a prediction model, and an optimal model is optimized through cross validation and performance evaluation. The robustness is improved by global parameter adjustment in combination with optimization algorithms such as a particle swarm, process schemes such as aeration intensity and carbon source adding are generated through multi-objective optimization after containerization deployment, and a whole-process intelligent management and control system is constructed by integrating virtual verification, graded early warning and a self-adaptive feedback mechanism. According to the method, the problems of detection lag, insufficient model generalization ability, regulation response delay and the like of a traditional method are solved, and the operation energy consumption and the medicament cost are remarkably reduced while it is guaranteed that the effluent quality stably reaches the standard.

The invention relates to the technical field of hydraulic engineering safety monitoring, and particularly discloses a hydraulic engineering potential safety hazard assessment and prediction system and method based on image recognition. A multi-scale convolutional neural network is combined with a three-dimensional point cloud registration technology to extract surface visual feature parameters, and adaptive time-frequency analysis and a wavelet packet reconstruction algorithm are used to extract physical feature parameters of internal concealment defects; constructing a dual machine learning framework, eliminating environmental interference through a deep residual network, analyzing a causal relationship between features based on a gating cycle unit, and screening a key feature parameter set; a Gaussian process regression model of an adaptive kernel function is used for dynamic risk prediction, risk abrupt change points are identified in combination with multi-scale wavelet transform, and finally a safety state score and a grading early warning signal are generated through a fuzzy comprehensive evaluation algorithm.

The invention provides a particulate matter and ozone source monitoring method and system based on a distributed sensor, and relates to the technical field of pollution treatment. According to the invention, sensor nodes with geographic perception capability are deployed in a monitoring area in a high-density manner, pollutant concentration and meteorological parameters are collected in real time, and data are uploaded to a cloud platform for preprocessing and dynamic calibration; a machine learning model is constructed based on the combined features of the pollutants and the meteorological factors, a driving relation is mined, and pollutant influence factors are extracted; further performing joint modeling on the influence factors and regional emission source data, identifying the coupling strength between pollutants and emission sources by adopting a classification or clustering method, and judging the categories of main sources; backward trajectory simulation, source fingerprint analysis and multi-source regression decomposition are combined to realize pollution path inversion and source contribution rate quantification; and finally, constructing a geographic information visualization platform, displaying a pollution thermodynamic diagram, a contribution change diagram and an evolution path diagram, and providing support for multi-source pollution traceability and scientific management and control.

The invention relates to the technical field of hard disk testing and verification, in particular to a multi-thread-based solid state disk reliability automatic testing method and system.The method comprises the steps that firstly, SMART information is deeply analyzed through microsecond-level high-granularity continuous performance monitoring, and multi-thread parallel processing is assisted; according to the method, fine performance fluctuation of the solid state disk under the concurrent load can be quickly captured, a fault mode can be identified, then early warning is realized by utilizing the extracted multi-dimensional features and a machine learning model, and a detailed fault diagnosis report is generated; and through dynamic error correction code strength verification and data integrity verification under pressure, an internal error correction mechanism of the solid state disk is actively detected and optimized. And finally, in combination with prediction reliability modeling, the system can estimate the remaining service life and predict faults, and provides product optimization suggestions for design, manufacturing and firmware optimization of the solid state disk, so that automation, intelligence and full life cycle management of the fault detection reliability of the solid state disk are realized.

The invention relates to the field of distributed machine learning, provides a big data-oriented distributed model training optimization method, and solves the problems of load imbalance, low resource utilization rate, large communication overhead, insufficient fault-tolerant efficiency and the like caused by data fragmentation staticization in the prior art. Load balancing is realized through intelligent clustering and overlapping control; the multi-dimensional heterogeneous resource evaluation model monitors calculation / storage / network indexes in real time, and realizes adaptive scheduling in combination with a task prediction and optimization algorithm; the hierarchical gradient synchronization mechanism adopts a tree-shaped parameter server and a dynamic compression technology, so that the communication traffic is reduced by 50%, and the precision loss is less than 0.8%; the incremental checkpoint system uses erasure code coding and parallel recovery to shorten the fault recovery time from 15 minutes to within 2 minutes, the resource utilization rate reaches 85% or above, the convergence speed is improved by 30%-40% in ResNet, BERT and other model training, and the large-scale training efficiency and the system stability are remarkably optimized.

The invention relates to a network flow anomaly detection strategy generation method based on machine learning, and belongs to the technical field of machine learning. The method comprises the following steps: firstly, collecting network traffic data in a preset time window, and extracting feature vectors containing traffic, a time sequence and a protocol type; and inputting the feature vector into a long short-term memory auto-encoder model, and calculating a reconstruction error to judge whether the network flow is abnormal or not. Aiming at the abnormal feature vector, adopting a multi-agent depth deterministic strategy gradient algorithm to construct a plurality of cooperative agents, and independently generating a candidate abnormal detection strategy by each agent; through a cross-agent strategy evaluation mechanism, the difference between a joint strategy and a single-agent strategy in the aspect of anomaly detection accuracy is compared, cooperation gain is calculated, strategy exploration parameters of all agents are adjusted according to the cooperation gain, and a global optimal anomaly detection strategy is optimized and determined in real time. According to the method, high-precision and low-missing-report network traffic anomaly detection can be realized, and the method has good self-adaptability and real-time performance.

The invention relates to the technical field of biomedical engineering, in particular to a multi-mode brain anomaly detection method and system based on machine learning. The method comprises the following steps: acquiring brain medical image data of different modalities, and realizing spatial registration and alignment through a multi-modal registration algorithm based on mutual information; a multi-branch feature extraction model including a convolutional neural network, a converter and a state space model is utilized to perform feature embedding on the original image of each modal; performing frequency decoupling on the features of each mode through adaptive approximate wavelet transform, and decomposing the features into high-frequency detail information and low-frequency global information; a frequency band fusion strategy based on an attention mechanism is implemented on high and low frequency features of different modal images, and fused frequency sub-band features are input into a space-frequency Mama module. Through the adaptive frequency domain decomposition and cross-modal fusion mechanism, the multi-modal brain image information is effectively integrated, and the accuracy and robustness of brain anomaly detection are remarkably improved.

The invention relates to the technical field of robot control, and discloses a robot control method, system, equipment and medium based on a multi-modal large model, and the method comprises the steps: collecting the multi-source modal data of a scene where an operation task is located, and carrying out the processing through a machine learning model, obtaining a multi-modal feature, and carrying out the position coding and Transform fusion processing, multi-modal fusion features are obtained, the multi-modal fusion features and the constructed job task knowledge base are input into a large language model to decompose a target job task, a human-in-the-loop mechanism is introduced to optimize a decomposition result, and a sub-task sequence is obtained; according to a subtask type in the subtask sequence, processing the subtask sequence through a visual language action model or a reinforcement learning model, and generating a motion instruction to enable the robot to start an execution process of the target operation task; live-line work tasks are processed through the multi-modal large models LLM, VLA and the like, and the work efficiency of the autonomous distribution network live-line work robot is improved.

The invention discloses a semiconductor packaging device electromagnetic compatibility comprehensive test method and system, and belongs to the technical field of electromagnetic compatibility testing. The method comprises the following steps: collecting structure parameters, packaging topology and predefined function states of a to-be-tested packaging device, and constructing a polymorphic working model; establishing a disturbance injection control model according to each state and configuring disturbance source parameters; implementing dynamic disturbance injection and acquiring response data in a real working state of the device; performing time domain and frequency domain conjoint analysis on the response data, constructing an electromagnetic response dynamic feature sequence, inputting the electromagnetic response dynamic feature sequence into a machine learning model, extracting multi-dimensional coupling features and predicting tolerance; calculating performance indexes such as an interference tolerance score and a coupling strength index based on model output indexes, comparing the performance indexes with a standard, and evaluating a compatible risk level in a full state; the method realizes quantitative evaluation of the EMC performance of the packaging device with high reduction degree and multi-state coverage, and has the advantages of comprehensive test, accurate prediction, explainable attribution and the like.

Systems and methods are provided for machine learning models configured as zero-shot personalized text-to-speech models which comprise a feature extractor, a speaker encoder, and a text-to-speech module. The feature extractor is configured to extract acoustic features and prosodic features from new target reference speech associated with the new target speaker. The speaker encoder is configured to generate a speaker embedding corresponding to the new target speaker based on the acoustic features extracted from the new target reference speech. The text-to-speech module is configured to generate the personalized voice corresponding for the new target speaker based on the speaker embedding and the prosodic features extracted from the new target reference speech without applying the text-to-speech module on new labeled training data associated with the new target speaker.

The invention discloses an educational resource image intelligent recommendation and multi-scale matching system and method based on machine learning, and relates to the technical field of intelligent education, a permeability evaluation model is constructed by collecting interactive behavior data of students and educational resource images to quantify cognitive stability and knowledge mastery parameters; synchronously performing multi-scale analysis on the image to generate a visual feature, a knowledge association feature and a cognitive guide feature; and determining an optimal recommendation level based on dynamic granularity selection processing, generating an enhanced image with a permeability feedback mark, and interactively generating learning evaluation data through the enhanced image. According to the method, cognitive dynamic evaluation and multi-scale feature matching are fused, so that the technical bottleneck of a traditional recommendation system on image granularity selection and cognitive adaptation is solved, and the educational resource recommendation accuracy and learning efficiency are remarkably improved.

The invention relates to the technical field of material verification management, and discloses a government affair material intelligent verification method and system based on machine learning, and the method comprises the steps: obtaining the text information data and verification rules of historical materials; multi-modal feature extraction is carried out; establishing a dynamic verification rule knowledge graph, and verifying the multi-modal features and the target feature points of the text segments to generate a primary verification strategy; performing deep verification processing on the material text fragment to generate a secondary verification strategy; performing third-time verification on the material text fragment subjected to the second-time verification by triggering cross-page feature matching to realize cross-page text fragment association and multi-modal feature fitting so as to generate a third-time verification strategy; performing iterative comparison on the results of the three rounds of verification to generate a credibility score; and the comparison result is adjusted according to the credibility score in combination with the verification risk abnormity assessment, so that a final verification decision strategy can be obtained, and efficient, accurate, dynamic and comprehensive intelligent verification can be realized.

The invention relates to the technical field of fault prediction, in particular to a centrifuge fault prediction system based on machine learning, which comprises a data channel synchronization module, a multi-dimensional feature extraction module, a state evolution index construction module, a trend aggregation trajectory recognition module and a fault section recognition module. According to the method, different types of data are synchronously aligned by a multi-channel signal segmentation processing mechanism based on a periodic state, a state evolution sequence is constructed in combination with a unified sampling structure based on a time scale, a state characteristic track is established through a multi-dimensional parameter set, a trend change index is constructed by means of a difference root-mean-square between adjacent states, and the state evolution sequence is analyzed. According to the method, the aggregation section is recognized and the trajectory deviation frequency is counted by utilizing continuous trend mutation, so that dynamic migration of the trajectory boundary and intelligent recognition of the fault section are realized, the boundary failure problem caused by static preset conditions is avoided, and the continuous prediction stability of long-period equipment and the application range under a non-standard working condition are effectively enhanced.

The invention relates to the technical field of computers, and discloses a computing power resource multi-dimensional scheduling method and system based on dynamic weight, and the method comprises a data perception step, a weight generation step, an intelligent decision-making step and a scheduling optimization step. The system corresponds to the method. The method comprises the following steps: a data sensing step: collecting multi-dimensional state parameters of computing power nodes and carrying out feature modeling to construct a global feature space; a weight generation step: dynamically adjusting the weight of each dimension based on a machine learning model and a rule engine; an intelligent decision-making step of screening candidate nodes in the global feature space and evaluating priorities, generating an optimal node cluster and performing resource dynamic slice distribution; and a scheduling optimization step: monitoring an execution effect and performing closed-loop feedback so as to iteratively optimize a weight strategy and decision logic. The problems that in the prior art, the sensing dimension is single, and decision-making weight is rigid are solved, and multi-dimensional accurate sensing, dynamic weight decision making and elastic resource allocation of computing power resources are achieved.

The invention provides an intelligent real-time detection method and system for PCCP welding quality, and relates to the technical field of online detection and intelligent evaluation of pipeline welding quality through machine learning. Light energy data and multi-light-source images of a spiral weld pool are collected, exposure parameters are dynamically adjusted through the energy difference of visible light near-infrared bands, and the real-time detection of the PCCP welding quality is achieved. Inhibiting strong light interference and generating a weld surface image; a stress concentration area is positioned by scanning a welding seam thermal deformation area and combining speckle pattern change, sound frequency change and the elastic characteristic of the thin-wall steel cylinder; inputting the surface image and the deformation data into a space-time convolutional neural network, fusing light energy change, image details and spatial features to construct a weld joint space structure diagram, and adaptively correcting the position of a sensor; and comparing the sinking depth of the three-dimensional point cloud reconstruction, analyzing the correlation between the sinking degree and the stress, and generating a probability thermodynamic diagram to output the pressure-bearing failure risk level, so that the probabilistic early warning of the pressure-bearing failure risk can be realized.

The invention discloses a remote sensing image generation method based on a federal visual language model, which belongs to the technical field of machine learning and specifically comprises the following steps: receiving text instruction description by each client; extracting a multi-scale feature map from private remote sensing image data through a visual encoder, and generating a semantic embedding vector by text instruction description through a language encoder; inputting the semantic embedding vector and the multi-scale feature map into a dynamic attention mask generator to generate pixel-level space weight distribution; carrying out weighted fusion operation on the multi-scale feature map, and generating visual feature representation of text conditionalization; generating a remote sensing image according with the description of the text instruction through an image decoder; the client uploads model parameter increments of the visual encoder, the language encoder and the dynamic attention mask generator to the central server; the central server aggregates the model parameter increments, and distributes the updated global model parameters to each client; according to the method, the flexibility and semantic consistency of remote sensing image generation are effectively improved.

The invention relates to the technical field of robot control, and discloses a heavy-load robot motion trail method and system based on machine learning. The method comprises the steps that historical movement track data of the heavy-load robot in a working scene are collected, and the data comprise a joint position sequence, an end effector pose sequence and environment obstacle distribution information; the data is preprocessed, track features are extracted, a space-time correlation matrix is constructed, and the matrix is used for representing the dynamic coupling relation between joint movement and the tail end pose; training a trajectory prediction model containing a long and short-term memory network and an attention mechanism based on the matrix, and generating a collaborative mapping relation between a joint position and a tail end pose; obtaining a current task target pose sequence and an environment constraint condition in real time, and outputting a candidate track set meeting dynamic constraint through a model; and adopting a multi-objective optimization algorithm to screen candidate tracks, generating an optimal track instruction and issuing the optimal track instruction to an execution mechanism. The method adapts to the complex characteristics and variable working conditions of the heavy-load robot, and the track adaptability is improved.

Deep learning tools such as convolutional neural networks (CNNs) and transformers have spurred great advancements in computational biology. However, existing methods are constrained architecturally in context length, computational complexity, and model size. This application introduces a sub-quadratic architecture for modeling, which combines projected gated convolutions and structured state spaces to achieve local and global context with, for example, single-nucleotide resolution. These models outperform CNN-, GPT-, BERT-, and long convolution-based models in many tested genomics tasks without pre-training and with 4×-781× fewer parameters. In the proteomics domain, these models similarly outperform pretrained attention-based models, including ESM-1B and TAPE-BERT, on remote homology prediction without pre-training and while using 3,308×-23,636× fewer parameters.

The present invention relates to the technical field of the control of wastewater treatment devices. Disclosed is an intelligent control method for wastewater treatment devices at a dry bulk cargo terminal, which is used for solving the problem of poor control of wastewater treatment devices at a terminal. The method comprises the following steps: installing a plurality of types of sensors at key locations of a dry bulk cargo terminal, and using edge computing nodes to perform real-time data collection and preprocessing; on the basis of historical features and temporal features, using a machine learning model to perform wastewater type classification, thereby realizing efficient dynamic adjustment of operating parameters of wastewater treatment devices; then, by means of weighted voting and confidence evaluation, integrating a plurality of classification results to ensure an optimal treatment effect; and analyzing actual wastewater treatment conditions to continuously optimize device control, thereby preventing faults, extending the service life of devices, and improving the wastewater treatment effect.

Systems, devices, and methods for wireless intrusion detection based on deep learning are provided. A network device collects legitimate network traffic over a time period and learns a first set of features that represents the legitimate network traffic. The network device generates synthetic network traffic based on the learned first set of features and trains a machine learning model based on the learned first set of features and the synthetic network traffic. Based on the training, the machine learning model learns a second set of features that differentiates the synthetic network traffic from the legitimate network traffic. The devices and methods precisely detect potential security threats, while reducing false positives, thereby ensuring a sensitive and accurate response to genuine anomalies. Further, the devices and methods improve accuracy of detection of potential security threats including known and new attacks in wireless networks, while adapting to evolving attack techniques and network dynamics.

The invention provides a telecommunication service fraud-related risk security assessment system based on a multi-modal fusion model, and the system comprises a multi-source data collection assembly which is responsible for obtaining original telecommunication service data in multiple ways, carrying out the preprocessing of the original telecommunication service data, and obtaining first telecommunication service data; performing feature extraction and behavior pattern analysis on the first telecommunication service data to obtain key features related to the telecommunication service; the assessment model construction component is responsible for constructing a fraud-related risk assessment model based on machine learning, inputting the key features into the fraud-related risk assessment model, outputting an intelligent assessment control matrix, classifying and rating fraud-related risks, and generating a risk assessment result; and the service collaborative linkage assembly is responsible for early warning the risk level of the telecommunication service system according to the risk assessment result, and taking prevention measures according to the risk level. According to the method, an objective evaluation standard is established, and the conversion of risk identification from experience judgment to data driving is realized.

Techniques for document recommendation based on conversational log for real time assistance are described. A first machine learning module identifies key phrases of a conversational log in real time. The first machine learning module executes multiple machine learning models trained to determine a probability that a portion of a conversation includes a key phrase. A second machine learning module identifies assistance pertaining to the identified key phrases of the conversational log. The second machine learning module executes a machine learning model trained to identify semantic similarity and word matching features of embedding representations of the key phrases and a knowledge base of assistance. The assistance is provided to a user during a conversation in real time.

The decision process of a first machine learning (ML) model may be explained based on a second ML model implemented on an apparatus. The apparatus may obtain a prediction about an image made based on the first ML model. The apparatus may further determine visual concepts associated with the image that may have been used by the first ML model to make the prediction, and determine respective contributions of the visual concepts to the prediction made by the first ML model. The apparatus may then generate, based on the second ML model, a textual description that explains the respective contributions of the visual concepts to the prediction made by the first ML model. The second ML model may determine respective image features associated with the visual concepts, map the determined image features to corresponding text features, and generate the textual description based at least on the text features.

The invention discloses a fatigue life simulation evaluation method for a lightweight aluminum alloy material of a new energy automobile, and relates to the technical field of material life evaluation. A microstructure image is collected, coupling features are extracted through machine learning, and heterogeneous data fusion and enhancement are completed; generating a topological optimization structure based on a GAN, introducing a VPSC model to describe anisotropy according to a stress gradient dynamic grid, and constructing a dynamic finite element model; fusing vehicle driving data, predicting a load by using LSTM, performing VMD decomposition and environment correction, and realizing space-time correlation load spectrum reconstruction; a phase field model is used in a microcosmic mode, cracks are tracked in a macroscopic mode through XFEM, damage parameters are transmitted in a bidirectional coupling mode, and multi-physics field coupling simulation is carried out; fusing simulation and test data by adopting Bayesian reasoning, calculating life probability distribution, and correcting parameters when errors exceed the limit; according to the method, the fatigue life prediction error is finally reduced, the time consumption of single simulation is reduced, full-life-cycle evaluation and visual early warning are realized, an efficient scheme is provided for lightweight design, and industrial technology upgrading is promoted.

The invention relates to the field of underground powerhouse construction risk identification, and provides an underground powerhouse construction risk identification and disposal method, system, device and medium, and the method comprises the steps: collecting multi-source heterogeneous data in real time, obtaining historical risk case data, and carrying out the preprocessing to obtain structured time-space correlation data; constructing a multi-dimensional analysis model based on a parallel computing algorithm, and performing multi-scale risk analysis on the structured time-space associated data to obtain multi-level risk feature data; performing risk feature recognition through a multi-modal machine learning model to obtain risk quantitative indexes, and performing recognition based on a fuzzy comprehensive evaluation algorithm to obtain construction risk levels; and matching emergency strategies of construction risk levels, carrying out parameter expansion through a combinatorial optimization algorithm, generating a plurality of candidate disposal schemes, carrying out weight calculation and sorting on the candidate disposal schemes by adopting a multi-criterion evaluation model, and outputting an optimal disposal scheme. According to the invention, efficient identification and accurate emergency decision-making of the construction risk of the underground powerhouse are realized.