379 results about "Classification rule" patented technology

The invention relates to a safe backup and recovery method and system for medical data, and the method comprises the steps: obtaining the multi-dimensional attribute information of the medical data, and carrying out the dynamic classification of the multi-dimensional attribute information based on a preset classification rule, and obtaining an attribute information label; performing differential encryption on the medical data according to the attribute information tag to obtain encrypted medical data, writing the encrypted medical data into a multi-level storage node, and performing multi-level backup construction to obtain a cross-platform backup topology; performing traceability node extraction on the multi-level storage nodes according to the abnormal node data access behavior to obtain traceability node information, and performing data traceability verification on the traceability node information according to the backup topology configuration to obtain a trusted recovery chain; and performing recovery node selection on the multi-level storage nodes according to the trusted recovery chain, and performing multi-node collaborative decryption reconstruction to obtain a recovery verification report. According to the method, the redundancy storage cost can be reduced while the data recovery reliability is guaranteed.

The invention discloses a robot real-time potential safety hazard recognition system based on multi-modal sensor fusion, and particularly relates to the technical field of intelligent inspection and safety monitoring, the system comprises five parts of data acquisition, information fusion, behavior response, trajectory analysis and risk output, and the potential safety hazard recognition system is used for recognizing potential safety hazards through image acquisition, thermal imaging, gas concentration and temperature and humidity information. Carrying out numerical value normalization and feature extraction, identifying potential abnormity and generating early warning; triggering data enhanced acquisition and track recording in the target area, and constructing a space-time path model to analyze an abnormal evolution trend; and finally, outputting a potential safety hazard assessment result according to a risk level classification rule by combining the enhanced information and the trajectory features. According to the method, high-precision early warning is realized through multi-source data acquisition and normalization fusion, the local recognition capability is improved based on dynamic enhanced acquisition of a behavior response mechanism, an abnormal development trend is tracked by combining track evolution modeling, risk level assessment is output according to the abnormal development trend, and accurate recognition and dynamic management and control of hidden dangers are realized.

The invention discloses an intelligent data query method and device based on a large model and a medium, and relates to the technical field of large model application. The method comprises the steps of receiving a natural language query text input by a user, and determining a target interface group from pre-associated interface groups through a preset business line classification rule and a keyword matching algorithm; inputting the natural language input text into a pre-trained large model to analyze a parameter set required by interface calling; wherein the parameter set comprises at least two of the following items: a time range field, a business main body field and a statistical dimension field; calling an API interface in the target interface group according to the parameter set, and obtaining original business data; performing context supplementation on the original business data based on the business logic rule to generate a standardized data result containing business background description; and reconstructing the standardized data result according to a display format preset by a user, and outputting a combined visual result containing charts, tables and text description.

The invention relates to the technical field of text classification, and provides a text classification method based on a large model and a rule engine, and the method comprises the steps: S1, storing multi-level rule classification labels, and constructing a classification rule template library; s2, receiving text data from various data sources, and preprocessing the text data; s3, performing rule matching on the text data based on the classification rule through a rule engine, and outputting a rule classification result; and S4, when any one of the following conditions is met, large language model classification is triggered: a, a classification rule is not matched; b, matching a classification rule, wherein the rule confidence is smaller than a rule confidence threshold; c, the text data length exceeds the preset text data length; d, matching a specific business scene label; outputting a model classification result; and S5, when the rule engine classification in the S3 and the large language model classification in the S4 are parallel, executing the strategy. The output reliability and the service continuity are guaranteed, and the method is suitable for scenes with high accuracy requirements such as financial compliance examination and the like.

The invention discloses an electric power marketing management information platform daily electric power fitting method and related equipment, and relates to the technical field of electric power data management, and the method comprises the steps: obtaining historical load data, external environment data and equipment operation state data of a target region; constructing an initial load fitting model based on the historical load data; extracting a date characteristic factor according to a preset time classification rule, and dynamically correcting the initial load fitting model based on the date characteristic factor to generate a corrected load model; generating a multi-source fusion feature based on the external environment data and the equipment operation state data; and inputting the multi-source fusion features into the corrected load model, and outputting a target load prediction result.

The invention relates to the technical field of databases, in particular to an intelligent mapping and classifying method based on heterogeneous data sources, which comprises the following steps: collecting heterogeneous data streams through an API (Application Program Interface) gateway and converting the heterogeneous data streams into structured data packets; using a semantic topology engine to fuse BERT semantic extraction, a graph convolutional network and a dynamic time warping technology to generate a cross-source association graph; constructing a field type clustering center by adopting a meta-learning framework based on the atlas, generating an initial classification rule through mode compatibility measurement, and dynamically updating the rule by means of adversarial training; outputting a DSL configuration script in combination with a template engine and an AST compiling technology; and dynamically adjusting a graph convolution weight and classifier parameters by using a strategy gradient algorithm through a reinforcement learning agent, and establishing a mapping-classification-verification collaborative optimization mechanism. According to the method, cross-source data semantic association accuracy is improved, small sample adaptive classification is realized, and system robustness and efficiency are improved.

The invention relates to the technical field of geological disaster early warning, in particular to a geological disaster early warning method and system based on dynamic data monitoring, and the method comprises the following steps: obtaining geological displacement monitoring data, setting a time window range, respectively calculating a local extreme point and a local zero point of the monitoring data in a time window, and calculating the local extreme point and the local zero point of the monitoring data in the time window; obtaining a multi-scale reference value; and based on the multi-scale reference value, calculating a comprehensive statistical magnitude, and generating a data quality evaluation parameter. According to the method, a local extreme point and a zero point are dynamically calculated by setting a time window range, a multi-scale reference value is constructed to carry out comprehensive statistic generation and data evaluation, and collaborative discrimination of piecewise linear regression and threshold comparison are executed to form a critical index; and performing characteristic quantity recursive residual spectrum analysis and matrix decomposition operation by combining the power spectral density distribution and the spectral density matrix to establish an early warning characteristic sequence, and realizing multi-stage early warning indication output based on a grading rule.

The invention discloses a multi-mode teenager idiopathic scoliosis screening method based on a back RGB-D image, and belongs to the technical field of image processing and scoliosis screening, and the method specifically comprises the steps: collecting an RGB-D image pair of a subject in an Adams anteflexion posture, carrying out the manual marking of an RGB color image, constructing and training a back key point detection model, and carrying out the detection of the back key point. And establishing a calculation rule of a maximum trunk rotation angle, a minimum trunk rotation angle and a back contour asymmetry index, and establishing an AIS classification rule based on the RGB-D data set, thereby completing AIS classification of the to-be-detected subject. According to the non-contact radiation-free AIS screening method, comprehensive utilization of back two-dimensional and three-dimensional information is achieved by means of data complementation of the RGB mode and the D mode, the problem of feature deficiency in a single mode is solved, and the method has the advantages of being fast, efficient, robust and objective.

The invention relates to a data transmission method and device, a DTU and a cloud, and the method comprises the steps: obtaining original perception data collected in a preset time period; dividing the original sensing data according to a preset importance degree division rule to obtain high-priority data, medium-priority data and low-priority data; the high-priority data, the middle-priority data and the low-priority data are compressed according to different compression strategies respectively, a high-priority data packet, a middle-priority data packet and a low-priority data packet are obtained, and the compression ratios of the compression strategies corresponding to the high-priority data, the middle-priority data and the low-priority data are sequentially increased; and uploading the low-priority data packet, the medium-priority data packet and the high-priority data packet to the cloud. According to the embodiment of the invention, the method can prevent the perception data with low importance from wasting transmission resources, reserves more transmission resources for the perception data with high importance, and guarantees the timely, accurate, smooth and efficient transmission of the perception data with high importance.

The invention provides a data classification and grading method and device based on an artificial intelligence large model technology, and relates to the field of data security. The method comprises the following steps: constructing a data classification and grading catalog according to a preset classification rule, and storing each row of the classification and grading catalog as a document record in a target vector database; respectively carrying out first retrieval operation, second retrieval operation and third retrieval operation on the data classification and grading catalog by taking the business system name, the data table name and the field name as first query conditions so as to obtain a first matching document, a second matching document and a third matching document in the target vector database; sorting the first matching document, the second matching document and the third matching document according to a preset sorting mode, and obtaining a sorting result; and inputting a sorting result into the large language model, and performing data security early warning through the large language model. According to the method and the device, the security problems that corresponding security measures cannot be taken, sensitive data is leaked and the like when a data governance engineer identifies the metadata according to the data classification and classification catalog and judges the classification and security level of the metadata are solved.

The invention provides a medical record information extraction and analysis method and system based on a deep learning large model, and the method comprises the steps: carrying out the semantic understanding of a medical record text through a deep learning model, and recognizing key clinical features; extracting the key clinical features, and carrying out vectorization processing on the extracted key clinical features; performing sentence segmentation and word segmentation on the case text based on a rule engine and the key clinical features; adjusting the segmented words based on a symptom classification rule; and carrying out word frequency statistics on the adjusted segmented words. According to the method, accurate recognition and structured conversion of key clinical features can be ensured through a semantic understanding technology, knowledge graph construction and machine learning model optimization are supported, and a new view angle and a new tool are provided for exploration and research of diseases.

The invention relates to the technical field of urban space information processing, in particular to an urban functional area fine classification method based on multi-modal data collaboration. According to the method, POI data, AOI data, land utilization data, remote sensing images and other multi-source heterogeneous data are fused, a multi-dimensional feature system is constructed, and an urban functional area classification model covering a single functional area, a composite functional area, a mixed functional area and a non-functional area is provided. According to the method, firstly, block-level basic units are constructed based on an open street map (OSM), then coordinate transformation and function classification are carried out on POI and AOI data, features such as POI density, AOI area proportion and land utilization structure are extracted respectively, and finally, function type identification is carried out on urban block units according to set classification rules. Compared with the prior art, the method has the advantages that through multi-source information fusion and discrimination logic optimization, the accuracy and applicability of urban functional area recognition are remarkably improved, and the method is suitable for scenes such as urban planning, land utilization evaluation and urban management.

A method and system for controlling a heating, ventilation, and air conditioning system (HVAC) in a building based on adaptive model predictive control (MPC) is disclosed, and belongs to the technical field of intelligent building adaptive control. The method may include: acquiring operation data of the HVAC under different working conditions, and performing preprocessing on the acquired data; calculating a marginal fitness measure (MAD) of each data point for each class based on the preprocessed data, and calculating a fitness measure of each data point for each class by combining MADs of all descriptors to obtain a global fitness measure (GAD); setting a GAD threshold, and classifying current operation states of the system according to classification rules; and adopting adaptive control strategies for different operation states since model parameters of each class are fixed after classification.

The invention relates to the technical field of tea quality evaluation, and particularly discloses a tea quality digital comprehensive evaluation system based on multi-source data fusion, and the system comprises a multi-source data collection module which is used for obtaining multi-dimensional data of tea; the multi-dimensional feature extraction module is used for extracting chemical quality parameters, physical characteristic parameters and process correlation parameters and generating a comprehensive quality embedding vector; the digital evaluation model construction module is used for generating a quality grade division rule and a quality defect traceability path; and the tea quality comprehensive evaluation module is used for generating a multi-dimensional quality evaluation report according to the quality grade division rule and the quality defect traceability path. Through three technical breakthroughs of multi-source heterogeneous data fusion, intelligent algorithm collaborative analysis and dynamic feedback control, the method has remarkable advantages in the aspects of accuracy, efficiency, traceability, industrial application value and the like, and an innovative solution is provided for standardized production and quality control of tea leaves.

The invention discloses an AI-based archive digitization full-process management and control method, which comprises the following steps: acquiring archive image data and text data, constructing an original archive data set, and generating an initial directory set; constructing an archive directory template set, executing template matching and structure adaptation, and generating an initial structure recognition result; setting a classification rule of directory deviation and structural contradiction, constructing a deviation rule base, executing rule scanning and node comparison, and generating a deviation contradiction label sequence; constructing an evolution state vector, inputting the evolution state vector into an improved GMN model, and generating a structure repair state vector through a structure guiding unit, a deviation sensing unit, a path fusion unit and a node evolution unit in sequence; and executing result updating and conflict rearrangement, completing standard template alignment and semantic coverage verification, and generating a final structure recognition result. According to the method, the structural integrity, the semantic consistency and the automatic repair capability of directory recognition are improved, and the method has relatively high adaptability and intelligent level.

The invention relates to the technical field of researching or analyzing materials by testing chemical or physical properties of materials, in particular to an intelligent analysis method for corrosivity of sulfide molecules in transformer oil. The influence of the oil environment on the adsorption energy is simulated and corrected in combination with molecular dynamics, a microscopic parameter database is constructed, and a corrosion risk grade classification rule and a generation rate prediction model are established. Oil data are collected in real time and input into the model to output a dynamic risk level, and a circulation rate adjusting or corrosion inhibitor adding instruction is triggered. Model parameters are regularly corrected by combining a closed-loop feedback mechanism, and a microscopic database is expanded, so that transfer learning prediction of unknown sulfide corrosion behaviors is realized. According to the method, the limitation of an existing single experiment condition is broken through, the relevance between a microcosmic molecular mechanism and a macroscopic corrosion behavior is dynamically analyzed, and the prediction precision under the multi-factor coupling effect is improved.

The invention provides a refractory material quality detection method and system based on machine learning, and the method comprises the steps: collecting a multi-angle surface image set of a refractory material sample, covering images under different illumination conditions, so as to comprehensively capture material surface features; a pre-trained image feature analysis model is called to extract surface texture features, pore distribution features and microscopic crack features, and an image feature set is formed; performing joint classification processing on the image feature set based on a preset defect classification rule base, and identifying and quantifying defect types and distribution parameters thereof; generating a quality evaluation parameter according to the defect type set and the distribution parameter, comparing the quality evaluation parameter with a preset quality standard threshold value, and outputting a quality detection result; and a detection result is fed back to a production line control system, and sorting operation and process parameter adjustment are automatically triggered to realize intelligent monitoring and optimization of the production process, so that the accuracy and efficiency of quality detection of the refractory material are improved, and the improvement of the production quality is promoted.

The invention provides a screen mixed content classification method and system oriented to efficient compression, and relates to the technical field of screen mixed content classification, and the method comprises the steps: obtaining a screen mixed content image; performing block processing on the screen mixed content image to obtain a plurality of target blocks; determining a first category classification result of each target block in a feature-based classification mode; constructing an improved convolutional neural network model; determining a second category classification result of each target block through an improved convolutional neural network model; and based on the first category classification result and the second category classification result, determining a final category of each target block through a preset classification rule. According to the invention, by fusing the classification results of the first category and the second category, collaborative optimization of the traditional features and the depth features is realized, so that the image content recognition precision in a complex scene is remarkably improved.

The invention is suitable for the technical field of privacy data processing, and provides a privacy grading and sharing control method for tourist travel itineraries, and the method comprises the steps: obtaining travel itinerary data of tourists; dividing the travel data into at least three basic privacy levels based on a preset region sensitivity classification rule, a user identity classification rule and an equipment type classification rule; in the basic privacy level, performing fine-grained sub-level division on the data according to interval division of travel duration, classification of activity risk levels and a historical sharing frequency threshold; based on the basic privacy level and the fine-grained sub-level, geographic position fuzzification and user identity anonymization processing are carried out on the travel data of the high-sensitivity level, and the access range of a shared object is limited through permission grading; and the sensitivity of the fine-grained sub-level is adjusted according to the real-time sharing frequency, and the sharing permission is updated, so that the accurate balance of the travel itinerary data between privacy protection and sharing effectiveness can be realized.

The invention discloses an idle land recognition method, device and equipment and a storage medium, and belongs to the technical field of machine learning, and the method comprises the steps: obtaining a first land feature set based on the land text data of a plurality of pieces of land; screening the first land feature set to obtain a second land feature set; semi-supervised learning cooperative training is carried out on a first classifier and a second classifier based on the second land feature set, the first classifier and the second classifier are used for idle land identification, the first classifier is realized based on a rotating forest algorithm, and the second classifier is realized based on an extreme gradient lifting algorithm; wherein in the semi-supervised learning cooperative training process, the prediction labels are screened based on a preset idle land classification rule. The method can realize accurate and efficient idle land identification.

The invention relates to an automatic driving risk identification model training method and system, and a risk assessment method and system, and belongs to the technical field of vehicle automatic driving. The automatic driving risk identification model training method comprises the steps of performing time synchronization processing on multi-source sensing data to form a driving data set, and performing category labeling on the driving data set according to a set classification rule; the categories comprise traffic infrastructures, traffic auxiliary facilities, traffic participants and traffic environments; performing physical state information labeling on the driving data set subjected to category labeling to obtain a preprocessed data set; generating character risk cues describing different driving scenes based on the preprocessed data set; risk assessment is carried out on different driving scenes, risk states of the driving scenes are marked based on risk assessment results, and a set number of risk targets are marked for the driving scenes which are confirmed to be risk scenes; and forming a sample set by using the character risk prompt words, the risk states and the risk targets corresponding to different driving scenes.

Described embodiments provide systems and methods for mapping high-priority WiFi packets to low latency data pipe. An access point can receive the one or more low latency classification rules from a device monitoring one or more low latency classification rules at a cable modem termination system (CMTS). The access point can be in communication with the CMTS and provide wireless access to one or more devices in a local network. The access point can receive packets corresponding to the one or more low latency classification rules from the CMTS. The access point can communicate, based at least on the one or more low latency classification rules, the packets to a queue used for transmission of low latency traffic to the one or more devices in the local network.

The invention relates to the technical field of non-road mechanical vehicle monitoring, in particular to a non-road mechanical vehicle fuel consumption real-time estimation method and system. According to the method, multi-source data are collected and preprocessed, and the current working state of a vehicle is recognized in real time based on classification rules; dynamically setting Kalman filtering parameters according to the identified working state of the vehicle; predicting a predicted fuel state vector and an error covariance matrix at the next sampling moment based on Kalman filtering; meanwhile, the refueling event is detected in parallel, and if the refueling event is detected, the refueling event is recorded, and the fuel oil state and the error covariance matrix are reset; and carrying out fusion processing on the actual observation information and the predicted fuel state vector based on Kalman filtering to obtain optimal fuel state estimation. According to the method, the multi-dimensional observation model is constructed, and the fuel consumption estimation precision is greatly improved through multi-source information complementation; and a dynamic mapping relation between working condition identification and filter parameters is established, so that the filtering algorithm keeps excellent performance under different working conditions.

The invention relates to a food water content calculation method and device based on a visual image, equipment and a medium. First, food image data is collected, and first image data is obtained by using a standardized preprocessing technology. Secondly, extracting surface glossiness and texture feature values from the first image data, segmenting the image by using a feature detection algorithm, marking a key region related to the water content, and forming a visual feature set; thirdly, the relevance of all elements in the visual feature set is analyzed, food types are distinguished through a classification rule, feature weight distribution of different types of food is determined, and optimization of feature description is achieved; and finally, based on the optimized feature description, establishing a mapping relation model between the water content and the visual features, substituting a newly input food image into the model, predicting the water content, and outputting a real-time detection result. An efficient and convenient detection means is provided for the fields of food processing, storage and the like, the detection cost is reduced, and the production efficiency and the product quality control capability are improved.

The invention provides an adaptive performance optimization method, device and equipment for a distributed database, and relates to the technical field of database system optimization, and the method comprises the steps: for any database node, obtaining first performance data of the database node and second performance data of each fragment in each database contained in the database node; for any fragment in any database contained in the database node, calculating a popularity score of the fragment according to the first performance data of the database node and the second performance data of the fragment; according to the popularity score of the fragment and a configured popularity classification rule, determining a popularity classification result of the fragment; according to a configured performance optimization rule, determining a target optimization strategy corresponding to the popularity score and the popularity classification result; and based on the target optimization strategy, carrying out adaptive performance optimization on the fragments. According to the method, the overall resource utilization efficiency and service performance of the database system are remarkably improved.

The embodiment of the invention provides a training method of a substation anomaly detection model and a substation anomaly detection method. The training method comprises the following steps: acquiring an initial image data set; classifying the initial image data set according to a classification rule to obtain two image data sets corresponding to the classification rule; based on the two image data sets corresponding to the classification rule, a pre-trained generator is adopted to carry out image expansion, an expanded image data set is obtained, and the generator is obtained by training a CycleGAN network model based on the two image data sets corresponding to the classification rule; obtaining a sample image data set according to the initial image data set and the expanded image data set corresponding to each classification rule; and training the initial anomaly detection model according to the sample image data set to obtain an anomaly detection model for detecting the anomaly of the transformer substation based on the image. The method is used for achieving the effect of improving the Luban performance of the anomaly detection model.

The invention belongs to the technical field of education data management, and particularly discloses an intelligent recording and broadcasting management method for education courses, which comprises the following steps of: acquiring multi-modal data in real time through a multi-sensor group, and comprehensively capturing teaching actions of teachers, limb actions of students, speech audios and running states of teaching equipment; further comparing the feature combination with a pre-stored template feature combination to identify different teaching stages; recording modes are automatically switched according to the identified teaching stage, a main and auxiliary picture polling switching rule is established, and videos recorded at different camera positions are integrated into a composite video with synchronous main and auxiliary pictures; key teaching links of teaching videos are automatically identified, index tags are established, the videos are classified and stored in combination with a preset classification rule, and accordingly video contents meeting user requirements are pushed to different users. Structured management and rapid retrieval of recorded and broadcast video content are realized, and personalized service and utilization efficiency of teaching resources are enhanced.

The present disclosure provides a method for selectively encrypting and decrypting portions of a document. The method includes receiving a document containing plaintext, analyzing the document using an artificial intelligence system to identify portions containing sensitive information, assigning predefined security levels or attributes to the identified portions, sending an encrypt file content request to a cryptographic engine, receiving an encrypted file content response, building an encrypted document by replacing portions with masking symbols and storing encrypted portions as metadata, and sending the encrypted document to a client device. The artificial intelligence system uses machine learning models trained on domain-specific data and employs pattern recognition and contextual analysis to identify sensitive content based on semantic understanding and classification rules.

The invention provides a soil lead element distribution monitoring method and system, and the method comprises the steps: obtaining topographic feature variables, remote sensing index variables, human activity intensity variables and geological feature variables of a plurality of sampling points of a target monitoring region in a specified current monitoring period in a preset variable database, constructing a lead migration potential index, an enhanced vegetation coverage dynamic index and a soil erosion-lead migration coupling index, and obtaining an actual soil lead spatial distribution probability based on a preset soil lead element distribution monitoring model; according to the actual soil lead space distribution probability, based on a preset lead element early warning grading rule, carrying out early warning grading on the sampling points; and according to the early warning level of the sampling point, generating a corresponding lead element early warning instruction and transmitting the lead element early warning instruction to early warning equipment, so that the soil lead element distribution monitoring efficiency can be improved.

The invention discloses a data modeling method and device, computer equipment and a storage medium, relates to the technical field of artificial intelligence, and is applied to data modeling of a financial big data platform. The original data elements are processed in a standardized mode, and the type labels are automatically marked according to the preset classification rules; in combination with a feature coding technology, converting multi-dimensional data elements into uniform characterization representation; calculating by using a data field standard map and semantic similarity, automatically identifying an abnormal mode, and performing quantitative scoring through a pre-training review model; and finally, performing iterative optimization according to the review index score, and generating a detailed feedback report. According to the method, the modeling efficiency is greatly improved, automation and intelligentization of the field creation, type judgment and standard binding processes are realized, the data standard coverage rate and the modeling quality are effectively improved, the risk of manual review and script deployment errors is reduced, and the modeling efficiency is improved. And the modeling reliability and the operation and maintenance efficiency of the financial big data platform in a complex business scene are enhanced.