45 results about "Pattern learning" patented technology

An adaptive random access system and method with learned query optimization for compacted data files that enhances random access performance through machine learning and pattern recognition. The system incorporates a query pattern learning module that analyzes historical access patterns and user behavior to build statistical models of data usage. An adaptive estimator module improves location estimation accuracy by incorporating learned patterns rather than relying solely on mathematical calculations. A predictive boundary detector uses learned codeword patterns to more accurately identify boundaries in compacted data, reducing misalignment errors. An intelligent search engine coordinates optimization strategies including context-aware search string parsing and encoding strategy selection based on learned performance data. A dynamic codebook optimizer reorganizes sourceblock layout based on access frequencies and co-occurrence patterns to improve retrieval speed. An enhanced search cache implements predictive caching algorithms that anticipate user queries and proactively load relevant data.

The invention relates to the technical field of medical information, and discloses a knowledge graph driven intelligent analysis system based on the medical field, and the system comprises a knowledge graph construction module which is used for constructing a medical knowledge graph containing a subject, an event, an object entity and an information reachable relation; the behavior pattern learning module is used for learning a parameterized behavior decision model for the subject entity; the dynamic deduction module is used for deducing the market change after the event is injected according to the behavior decision model and the information reachable relation; and the strategy generation module is used for reversely generating a market strategy by taking the dynamic deduction process as a fitness function. The method comprises the following steps: constructing the medicine field knowledge graph; learning a behavior decision model of the subject entity based on a graph; dynamically deducing the market change in a virtual environment; and reversely solving and generating the market strategy. According to the method, prospective dynamic deduction can be carried out on the market, the optimization strategy is actively generated, and the scientificity and timeliness of decision making are improved.

The invention discloses a robot application secondary development method supporting user behavior pattern self-learning, and relates to the field of robot application development, and the method specifically comprises the following steps: S1, data collection and treatment; s2, performing behavior abstraction; s3, mode learning; s4, performing double-domain scheduling during operation; s5, performing mode capitalization; and S6, low code reuse. According to the robot application secondary development method supporting user behavior mode self-learning, during operation, a double-domain isolation framework physically isolates a safety domain from a learning domain, and the situation that key operation of a robot is disturbed in the learning process is avoided; the scheduling algorithm combining the SMP and the EDF and a real-time monitoring mechanism can accurately control core indexes such as thread scheduling delay and interrupt response time, ensure that key tasks such as robot motion control and emergency fault processing are executed preferentially, meanwhile, parallel operation of behavior learning and execution is achieved, the strict requirement for real-time performance of an industrial scene is met, and the real-time performance of the industrial scene is improved. And the operation safety of the robot is ensured.

The invention provides an interactive real-time loop feedback programming teaching system based on a large language model, and the system comprises a task driving module which is used for generating a multi-language exercise library based on the large language model and dynamically recommending tasks in combination with the programming capability of a user and an interest label; the code writing and debugging module is used for writing multi-language codes by a user, receiving the multi-language codes submitted by the user, executing multi-language compilation and generating compilation data, and the compilation data comprises execution logs, performance data and error information; the programming capability portrait module is used for analyzing the compiling data by utilizing a large language model to obtain programming behavior data, and processing the programming behavior data to obtain a user programming capability portrait; the intelligent interaction module is used for analyzing the programming ability portrait by adopting a large language model to generate targeted teaching feedback, and calling AI assistant teaching to execute five-step guide teaching based on the targeted teaching feedback; the reflection internalization module is used for generating a reflection log based on the user ability portrait generated by the programming ability portrait module, the reflection log comprises an error mode, a learning habit and a cognitive blind spot, and an internalization strategy is recommended according to the content of the reflection log. The internalization strategy refers to providing personalized learning methods, learning plans, learning guide, learning guidance and recommendation tasks of updating the task driving module for the user according to the current error type.

The invention provides an industrial data dynamic sensing and acquisition strategy adjusting method, and belongs to the technical field of industrial data analysis. An unsupervised label generation module based on pattern learning generates high-quality phase labels for unlabeled data in batches through multi-dimensional features; the multi-feature fusion hybrid detection network decouples an input signal into a periodic component and a residual component based on a signal decomposition algorithm, extracts depth features of the periodic component and the residual component by adopting a parallel double-branch structure, and comprehensively senses and predicts a system state through a feature fusion network; the reinforcement learning-driven adaptive sampling system is based on a deep reinforcement learning technology, an intelligent decision-making system for dynamically adjusting sampling frequency and acquisition strategies according to the real-time state of the system is constructed, sampling decisions are continuously optimized through interactive learning with the environment, and on the premise of ensuring that key information is not lost, the sampling efficiency is greatly improved. And the collection efficiency and the resource utilization rate are maximized.

The invention relates to a home violation risk identification method, system and device based on artificial intelligence and a medium. The method comprises the following steps: preprocessing audio data to obtain a voice segment; performing automatic voice recognition on the voice segments to obtain a dialogue text sequence, and performing acoustic feature extraction to obtain a time sequence acoustic feature sequence; performing key feature extraction based on context semantics and risk knowledge on the dialogue text sequence to generate a semantic feature vector; performing deep emotion mode learning on the time sequence acoustic feature sequence to generate an acoustic feature vector; performing multi-modal fusion on the semantic feature vector and the acoustic feature vector to obtain a fusion feature vector; and carrying out collaborative risk judgment on the fused feature vector to generate a result containing high, medium and low risk levels and corresponding judgment confidence coefficients. By adopting the method, the limitation of single modal analysis can be overcome, and the home violence risk can be identified more comprehensively and accurately.

The invention relates to the technical field of medical artificial intelligence diagnosis, and discloses a diabetic nephropathy auxiliary identification method based on multi-source data. The method comprises the following steps: starting a medical data traceability process, capturing original indexes from a multi-heterogeneous system, and arranging the original indexes into an initial data set; and performing hierarchical data fusion according to the pathological evolution knowledge graph to generate a fusion health record with a timestamp. And introducing a dynamic feature weaving process, and extracting a multi-level abnormal feature network from the physiological parameter trajectory according to the priority of the biomarker. And importing the abnormal feature network into a progressive mode learning process, training an identification engine for mapping feature combinations to different disease risk levels, and outputting an identification conclusion by the engine. According to the method, the structured abnormal mode network is constructed through priority-driven dynamic feature weaving, and the recognition engine is trained by using a progressive learning mechanism, so that the accuracy and clinical interpretability of risk stratification of diabetic nephropathy are improved.

A Large Codeword Model (LCM) is a deep learning architecture that operates on discrete, compressed representations of data called codewords. Unlike traditional models that use raw tokens and dense embeddings, LCMs can efficiently process and generate data in various modalities, including text, images, audio, and time series. By capturing the inherent structure and patterns in the data, LCMs learn more generalizable and interpretable features, enabling transfer learning across different domains. The LCM architecture offers a scalable, flexible, and computationally efficient approach to building AI systems, with potential applications in natural language processing, speech recognition, and beyond.

This invention belongs to the field of soft measurement modeling technology and discloses a soft measurement modeling method based on mode-related spatiotemporal diffusion generation, which includes the following steps: (1) acquiring data of multi-mode dynamic processes; (2) data partitioning and preprocessing operations; (3) establishing a mode-related spatiotemporal diffusion model and generating virtual samples; (4) predicting pressure variables of three-phase flow processes and evaluating model performance. This invention proposes a soft measurement modeling method based on mode-related spatiotemporal diffusion generation. By using the DSTN-Net noise prediction network to capture the dependence of dynamic data in the time and space dimensions, the mode learner learns the multi-mode distribution characteristics and distinguishes the data distribution of different modes, which can generate virtual samples with high similarity to the original samples, thereby improving the prediction performance of the soft measurement model in the case of small samples.

An object of a system according to an embodiment is to efficiently and accurately perform a checking operation when an alarm is generated.SOLUTION: A system includes an alarm data collection part, a pattern learning part, a confirmation work proposal part, a confirmation work execution part, a report part, and a history storage part. The alarm data collection unit collects past alarm data. The pattern learning unit learns a pattern based on the data collected by the alarm data collection unit. The confirmation work proposal unit proposes confirmation work when an alarm is generated. The checking work execution unit executes the checking work proposed by the checking work proposal unit. The reporting unit reports a result of the checking work executed by the checking work execution unit. The history storage unit stores a history of the confirmation work executed by the confirmation work execution unit.SELECTED DRAWING: Figure 1

The application relates to a kind of multi-data set X-ray security image target detection methods based on dual-mode network, comprising: lattice appearance enhancer network connection basic feature extraction network forms target detection network;Through target detection network and K X-ray security image data set, the teacher model of K target detection network is trained;Common class mode learning network and unique class mode learning network of K X-ray security image data set are constructed;Using the teacher model of K target detection network after training, student model of target detection network is alternately carried out knowledge distillation under common class mode learning network and unique class mode learning network;Student model of target detection network after knowledge distillation is trained using back propagation algorithm;The student model of target detection network after training is used to carry out target detection to X-ray security image, and the position and corresponding category of contraband in X-ray security image are obtained. The position and corresponding category of contraband can be accurately detected, and contraband is marked.

The application relates to a pedestrian trajectory prediction method based on a multi-disconnected mode learning, a prediction trajectory is generated based on a social disconnected mode generation adversarial network model, an encoder is used to extract visual features and observed pedestrian trajectory features, physical scene and social attention features are obtained based on an attention module; a structured graph sequence based on attention labeling is established by using the social attention, and the structured graph sequence is encoded by using a space-time encoder to extract transient changes of a physical background and pedestrian movement; the physical scene attention, the social attention and space-time encoder output features are spliced and input into a multi-generator architecture, and the future trajectory of a predicted pedestrian is output; a generator selector is used for prior learning of the multi-generator, and a spectral trajectory clustering module is used for updating an upper limit of the number of generators in the prior learning process. Compared with the prior art, the application can capture transient changes of space-time information, reduce model redundancy, and be flexibly adapted to multiple prediction scenes.

The application discloses a kind of information cascade prediction system and method based on the enhanced hawkes process of Transformer, including user embedding module, global dependence module, local dependence module, intensity function acquisition module and popularity prediction module, first, obtain user position-wise embedding and time coding as user embedding, then in the perspective of topological structure, introduce path perception hypothesis, and design two layers of attention layer from two angles of global embedding and local mode, parameterize the intensity function of hawkes process, and combine hawkes process, global embedding and local mode, learn the time and topological random characteristics coupled in information cascade diffusion process, to carry out popularity prediction;The application expands the traditional hawkes process, and effectively obtains knowledge from continuous time domain, improves the accuracy of popularity prediction.

The application relates to the technical field of medical information, and discloses a knowledge graph driving intelligent analysis system based on the medical field, which comprises a knowledge graph construction module, which is used for constructing a medical knowledge graph containing subjects, events, object entities and information accessibility relationships; a behavior pattern learning module, which is used for learning a parameterized behavior decision model for the subject entity; a dynamic deduction module, which is used for deducing market changes after injecting events according to the behavior decision model and the information accessibility relationship; and a strategy generation module, which is used for taking the dynamic deduction process as a fitness function to reversely generate a market strategy. The method comprises the following steps: constructing the medical field knowledge graph; learning the behavior decision model of the subject entity based on the graph; dynamically deducing the market changes in a virtual environment; and reversely solving and generating the market strategy. The application can perform forward-looking dynamic deduction on the market and automatically generate an optimized strategy, thereby improving the scientificity and timeliness of decision-making.

The invention relates to a multi-modal artificial intelligence generation content unified identification method and system. For multi-modal to-be-identified data, directly identifying the multi-modal to-be-identified data by adopting a fast thinking mode or performing interpretability identification by adopting a slow thinking mode based on the scene demand identification large model; the large identification model is trained by adopting a fast and slow thinking dual-mode two-stage training framework, an interpretable authentic identification data set is constructed, and standardized labeling from evidence extraction, space-time positioning to thinking chain construction is carried out. In a dual-mode supervision fine tuning stage, for a fast thinking mode, supervision fine tuning is utilized to realize fast discrimination and multi-classification traceability in a high-concurrency and low-delay scene; for a slow thinking mode, learning a complete interpretable reasoning chain from evidence extraction to criterion induction to conclusion generation; then, the model output is optimized through a preference alignment stage. Compared with the prior art, the method can provide a unified and reliable technical scheme for authenticity verification and interpretable review in a complex content scene.

Introduced here is a supervised spatial pooler for spatial pattern recognition using distance-based overlap measurement and distributed threshold-based winner selection. The supervised spatial pooler can incorporate batched learning and directed initialization. Moreover, a temporal memory system architecture is introduced, using synchronized spatial pooler components for spatial and temporal pattern learning. A language model extends the temporal memory system for predictive text generation, featuring an autoregressive encoder-decoder architecture with context-dependent token representations and end-of-sequence prediction. Additionally, an integrated pipeline architecture is described. The pipeline architecture uses the language model as its core learning algorithm and includes embedding and tokenization stages for neuromorphic computing applications.

The system according to this embodiment aims to detect the possibility of dementia early based on the lifestyle patterns and conversations of seniors, and to take appropriate action. [Solution] The system according to the embodiment comprises a conversation analysis unit, a pattern learning unit, a notification / advice unit, and an anomaly detection unit. The conversation analysis unit analyzes conversations and detects the possibility of dementia. The pattern learning unit learns the lifestyle patterns of seniors. The notification / advice unit detects behavioral anomalies early based on the lifestyle patterns learned by the pattern learning unit and provides notifications and advice. The anomaly detection unit detects anomalies from GPS information.

The invention discloses an intelligent door and window automatic ventilation and purification system based on air quality monitoring, and relates to the technical field of indoor air governance, the system comprises sparse sensor networks, a reverse solution engine, a targeted execution terminal, a mode learning and prediction module and a system control center, the sparse sensor networks are distributed at indoor key positions, and the sparse sensor networks are distributed at indoor key positions; synchronously collecting and uploading concentration time sequence data of various pollutants; a reverse solving engine is combined with the indoor space three-dimensional structure digital twinborn model, the position and intensity parameters of the pollution source are solved through an optimization algorithm, and a pollution source probability hotspot map is generated; the system control center dispatches a targeted execution terminal according to a resolving result, and local directional purification operation is carried out; the mode learning and prediction module realizes rapid response of a pollution source by learning historical data, realizes accurate monitoring, traceability and targeted treatment of pollutants, reduces energy consumption, improves treatment efficiency and pertinence, and adapts to various indoor scenes.

The application discloses a cross-sequence multi-behavior sequence recommendation method based on a Transformer. The method is as follows: firstly, a user historical interaction sequence is segmented, and then cross-sequence heterogeneous relationship propagation is performed; then, short-term interaction mode learning is performed; next, global representation aggregation is performed; finally, a recommendation content list is generated by combining user recent preferences and item embedding, and meanwhile, content priority is considered; and model parameters are adjusted by using user historical interaction for label data enhancement. The application proposes a user sequence segmentation method, guarantees that a subsequence maintains relatively concentrated and stable preferences of a user, and enables the model to more accurately capture dynamic changing preferences of the user, proposes a similar subsequence determination mechanism and a method for performing cross-sequence information propagation between similar subsequences, and preserves time coding information in the information propagation process, thereby improving the diversity and accuracy of recommended contents.

The system according to this embodiment aims to efficiently display and support the use of applications based on the user's usage patterns. [Solution] The system according to the embodiment comprises a collection unit, a learning unit, a display unit, a timing unit, and a notification unit. The collection unit collects the user's usage patterns. The learning unit learns the usage patterns collected by the collection unit. The display unit displays the application based on the patterns learned by the learning unit. The timing unit displays the application at specific times or on specific days of the week. The notification unit notifies the user when it is not possible to display the application fluidly.

An object of a system according to an embodiment is to intuitively assist an elderly person with a video or a sound when the elderly person operates a smartphone.SOLUTION: A system according to an embodiment includes an action pattern learning part, a video assist part, and a voice assist part. The behavior pattern learning unit learns a behavior pattern of the elderly person. The video assist unit assists an operation with a video on the basis of the action pattern learned by the action pattern learning unit. The voice assist unit assists the operation by voice based on the action pattern learned by the action pattern learning unit.SELECTED DRAWING: Figure 1

The invention discloses a document structure self-adaptive recognition method and device based on machine learning and a medium, and the method comprises the steps: obtaining input document data, preprocessing the input document data to convert a document file into uniform text format data, and extracting format feature data in the uniform text format data; performing pattern learning on the format feature data by using a machine learning algorithm to generate analysis rule data; the analysis rule data comprises a regular expression rule and a semantic analysis rule; performing document structure identification on the unified text format data by applying the analysis rule data, and constructing a document structure tree based on an identification result; converting the document structure tree into a standardized structured data format, and outputting structured data through an application program interface; the structured data format comprises a JSON format or an XML format.

This invention relates to the field of intelligent medical data analysis, specifically to an intelligent data analysis method for hospital infection control. The method includes the following steps: multi-source infection control behavior data collection, infection control behavior feature modeling, infection control behavior correlation modeling, graph-based behavior pattern learning, multi-modal fusion infection risk assessment calculation, and dynamic evolution analysis of infection risk. This invention, by constructing an infection control behavior relationship graph and introducing a joint embedding representation mechanism of behavior nodes and related features, achieves modeling of complex correlations between multi-source infection control behaviors, overcoming the limitations of existing technologies that rely solely on single rules or independent data indicators for analysis. Furthermore, this invention, by constructing an anomaly identification mechanism based on the deviation between predicted and actual behaviors, and combining cross-modal attention fusion and feature contribution analysis methods, achieves dynamic assessment and interpretable analysis of infection risk.

A system is provided.SOLUTION: A system comprising: means for performing pattern learning of a comment using a generative AI model; means for analyzing a newly input comment; means for automatically diagnosing whether or not the newly input comment corresponds to a harassment action based on an analysis result; and means for transmitting a diagnostic result to a user device.SELECTED DRAWING: Figure 1

The invention relates to the technical field of navigation, and provides an orientation method and device based on an atmospheric polarization mode learning network. By constructing a polarization mode learning network comprising two parallel feature extraction structures which are consistent in structure, feature extraction is performed on a polarization angle image and a polarization degree image by using the same feature extraction structure, and each branch can concentrate on processing polarization features of corresponding types while feature extraction logic is unified. A polarization mode learning network is trained through a data set comprising polarization image samples under multi-period and multi-weather conditions and corresponding target orientation parameters, data enhancement is carried out through polarization angle periodic processing, atmospheric attenuation simulation and regional disturbance in the training process, and a polarization mode is obtained through a data driving mode. A nonlinear mapping relation between a complex polarization mode and direction information is mined, so that the trained polarization mode learning network can output an accurate, stable and high-robustness orientation result under various meteorological conditions.

The application relates to a staff mobile mode learning method applied to coal mine underground work type identification, and belongs to the technical field of coal mine underground data analysis. A work type identification model is designed by using coal mine underground staff track data, the model realizes identification of which work type the input track data belongs to, and comprises a multi-semantics embedding module, a global feature extraction module and a work type identification module. Advantages: graph embedding and word embedding are combined to learn the embedding representation of staff, external factors are considered to affect the staff mobile track, richer semantic information can be obtained, the transformer technology is used to process variable long track sequences, global semantic features of the track are extracted, a contrast learning network for self-supervised track classification is involved, potential values of unlabeled data are fully tapped, the accuracy of work type identification is improved, technical support is provided for a coal mine underground scheduling platform, the problem of one person with multiple cards is avoided, intelligent scheduling of coal mine underground staff is realized, and underground production safety is maintained.

Introduced here is a supervised spatial pooler for spatial pattern recognition using distance-based overlap measurement and distributed threshold-based winner selection. The supervised spatial pooler can incorporate batched learning and directed initialization. Moreover, a temporal memory system architecture is introduced, using synchronized spatial pooler components for spatial and temporal pattern learning. A language model extends the temporal memory system for predictive text generation, featuring an autoregressive encoder-decoder architecture with context-dependent token representations and end-of-sequence prediction. Additionally, an integrated pipeline architecture is described. The pipeline architecture uses the language model as its core learning algorithm and includes embedding and tokenization stages for neuromorphic computing applications.

For providing recommendations for bill of materials revision, a database is storing bills of materials for a set of products. A pattern learning module processes the stored bills of materials to learn patterns. Embodiments for learning structural and temporal patterns are provided. A pattern application module applies the patterns to a current bill of materials for a product of interest and forecasts recommendations, with each recommendation indicating how the current bill of materials should be updated. A user interface outputs the recommendations along with the applied patterns and their confidence values. The method and system provide an automized framework that forecasts revision for products. That framework helps to boost product quality by avoiding late recognition of change needs that would most likely negatively impact product and cost performance. Automatically assessing the change needs reduces hours spent by domain experts on these tasks, which saves internal costs.

The invention relates to the technical field of network security, in particular to a network security monitoring system based on artificial intelligence, which comprises a data preprocessing module, an anomaly detection module, a fault mode learning module, a threshold dynamic adjustment module, an attack source positioning module, a behavior analysis and identification module, a decision support module and a recovery strategy optimization module. According to the method, the abnormal state can be efficiently identified through the isolation forest algorithm and the self-encoder model, the learning of the fault mode is deeper through the application of the long and short-term memory network, and the precise attack source positioning is realized through the combination of the self-regression differential moving average model and the graph database query technology; the accuracy of attack pattern recognition is improved by introducing the convolutional neural network and the graph neural network, the security decision is more scientific and reasonable by applying the analytic hierarchy process and the support vector machine algorithm, and the resource allocation and recovery strategy are optimized by using the genetic algorithm and the particle swarm optimization algorithm.