33 results about "Classification scheme" patented technology

The invention discloses a shale oil geological engineering dessert evaluation method and device based on geological modeling, and the method comprises the steps: determining a shale oil facies classification scheme and shale oil dessert evaluation key parameter values according to the shale oil geological engineering logging interpretation data, determining the corresponding scores of different shale rocks, and obtaining the shale oil dessert evaluation key parameter values of the shale oil facies classification scheme and the shale oil dessert evaluation key parameter values of the shale oil facies classification scheme; and determining a target sweet spot evaluation parameter weight based on sweet spot evaluation attributes of the lithofacies three-dimensional model and actual well drilling productivity data, establishing a shale oil geological engineering integrated comprehensive sweet spot factor, and determining a shale oil geological engineering comprehensive sweet spot evaluation result based on the sweet spot factor and the target sweet spot evaluation parameter value. Through the technical scheme of the embodiment of the invention, comprehensive evaluation of shale oil geological engineering desserts is realized, complexity of shale oil geological features can be adapted, correlation between geological and engineering dessert evaluation parameters is improved, and fine evaluation of shale oil dessert space distribution is ensured.

The invention provides a pore geometric morphology fine characterization method based on confining pore forming of mineral particles, which comprises the following steps: S1, performing fine characterization of the mineral particles based on geometric morphology characteristics, and accurately identifying the mineral particles in combination with characterization results; s2, performing geometric shape quantitative characterization on different types of pores from the angles of pore diameters, roundness, flatness, shape indexes and corner angle indexes to obtain geometric shape parameters of the pores; s3, on the basis of the recognized mineral particles, the names of confinement pores are determined, clustering analysis is conducted on the confinement pores in combination with the geometric morphology quantitative characterization result obtained in the step S2, and a pore classification scheme is established; and S4, based on a classification scheme and a quantitative characterization result, determining development characteristics of shale reservoir pores, establishing a porosity contribution index model formula, and matching and verifying a reservoir physical property result. According to the method, through comprehensive analysis of the pore morphological characteristics, solid data support and theoretical basis can be provided for scientific evaluation of a high-quality pore development mode.

Taxonomy driven multi-system networking and content delivery systems and methods are provided herein. An example systems includes a website authoring platform generating a taxonomy structure for a webpage, the taxonomy structure indicating assets to be included in the webpage a layout for the assets, a technical documentation authoring system for creating technical documents related to the assets, and a metadata binder providing the technical documentation authoring system with access to the taxonomy structure of the website authoring platform so that technical document authors are provided with the taxonomy structure to guide creation and structuring of the technical documents so that the technical documents comprise data that corresponds to the assets.

The invention relates to computer-implemented systems and methods for data risk management that provides traceability across governing artifacts which is key to driving effective data risk management and achieving compliance to policy, standards and controls. An embodiment of the present invention is directed to a standardized data risk taxonomy, harmonized classification schema, policy and standard hierarchy, control catalog and standard of care. These components facilitate an alignment to a full information lifecycle with a common definition of data risk, traceability across governing artifacts (e.g., obligations, policies, standards, risks, controls, etc.) and standardized catalogs with defined treatment across harmonized classification of assets.

The invention discloses an import and export commodity intelligent classification method based on knowledge graph metadata topology, and relates to the technical field of reinforcement learning, and the method comprises the steps: inputting an initial data packet into a dynamic interaction model, carrying out explicit association mining through a semantic enhancement layer, optimizing a rule matching path through a rule evolution layer, and constructing a dynamic commodity knowledge graph; performing topological structure derivation on the dynamic commodity knowledge graph to generate a graph topological analysis report and a metadata list, and performing knowledge reasoning integration on the graph topological analysis report and the metadata list to generate an intelligent navigation engine; calling an intelligent navigation engine to execute multi-path semantic query and rule verification on the dynamic knowledge graph to generate a candidate classification scheme set; and performing multi-target collaborative optimization on the candidate classification scheme set to generate a sorting scheme sequence, performing traceability packaging on the sorting scheme sequence, and outputting an intelligent classification scheme. According to the invention, through the dynamic interaction model and multi-target collaborative optimization, the rule adaptation efficiency in a complex scene is improved.

The invention discloses a multi-dimensional text data classification processing method and system, and relates to the field of text data classification processing. Collecting structured, unstructured and semi-structured texts, and carrying out denoising, word segmentation, standardization, missing value processing and abnormal text processing; extracting features from semantics, grammar, emotion and domain exclusive dimensions; optimizing features through standardization, correlation analysis, PCA dimension reduction and feature importance ranking; a'pre-training language model + traditional machine learning model 'fusion framework is constructed and trained; real-time and batch reasoning is supported; evaluating a classification result and feeding back a to-be-optimized sample; when the to-be-optimized sample reaches the standard or the performance is reduced, model iteration is started, and the model is updated after test verification. According to the method, the multi-source text preprocessing suitability and the data quality are improved, the classification accuracy and the field suitability are enhanced through multi-dimensional feature fusion, the performance and efficiency are optimized through the fusion model and incremental training, data changes are continuously adapted, and an efficient and reliable classification scheme is provided.

The invention relates to the technical field of ocean sample management, and discloses an ocean sample full-life-cycle digital management method, which comprises the following steps of: firstly, collecting characteristic data such as sample types, source positions and chemical component information of ocean samples, and calculating a classification scheme and storage parameters according to the characteristic data; evaluating a management demand and generating a target management strategy by combining the real-time management data of the sample library; making and optimizing a preliminary management plan by using algorithms such as k-means clustering and a decision tree in combination with historical management data and sample use frequency data, and generating a personalized management scheme; and finally, implementing the scheme, monitoring a sample preservation rate index through a sensor, dynamically adjusting a temperature threshold value and a storage position of classified storage equipment, introducing a real-time feedback mechanism to iteratively correct and adjust parameters, and generating a full-life-cycle management scheme. According to the method, full-process digital management of the ocean samples is realized, and the management efficiency and the sample storage quality are improved.

The application discloses an import and export commodity intelligent classification method based on knowledge graph metadata topology, relates to the technical field of reinforcement learning, and comprises the following steps: inputting initial data packets into a dynamic interaction model, performing explicit association mining in a semantic enhancement layer, optimizing rule matching paths in a rule evolution layer, and constructing a dynamic commodity knowledge graph; performing topology structure derivation on the dynamic commodity knowledge graph, generating a graph topology analysis report and a metadata list, integrating knowledge reasoning on the graph topology analysis report and the metadata list, and generating an intelligent navigation engine; calling the intelligent navigation engine to perform multi-path semantic query and rule verification on the dynamic knowledge graph, generating a candidate classification scheme set; performing multi-objective collaborative optimization on the candidate classification scheme set, generating a sorting scheme sequence, performing traceability packaging on the sorting scheme sequence, and outputting an intelligent classification scheme. The application improves the rule adaptation efficiency in complex scenarios through the dynamic interaction model and multi-objective collaborative optimization.

A classification apparatus includes: a feature quantity output unit that outputs a feature quantity of input data; and a classification unit that retains, as a classification weight, a feature quantity obtained by averaging, per each class, the feature quantity output by the feature quantity output unit in response to a base class dataset and a novel class dataset with a smaller number of data items than the base class dataset and that outputs a result of classification of the input data by using the feature quantity of the input data and the classification weight.

The invention relates to a content recommendation method and device, computer equipment and a storage medium. The method comprises the following steps: clustering user browsing data of different user accounts in an acquisition period, thereby dynamically dividing interest grading thresholds corresponding to different content types, and further pushing interested contents of different content types to each user account according to the dynamically determined interest grading thresholds. Compared with the prior art, the method does not depend on fixed statistical quantiles or normal distribution hypotheses, interference of hot content or extreme users on grading results can be reduced, objectivity of interest grading threshold values is improved, and the problem that certain interest grading schemes are insufficient in interest grading generalization is solved.

The application discloses a kind of based on deep learning's round billet end face billet number direction classification scheme, belong to intelligent metallurgical technology field, the method includes: establishing round billet end face billet number data set;Round billet end face billet number data set is divided into training data set and test data set;Each image in training data set is preprocessed;Based on the image after pre-processing, improved AlexNet network is trained, and classification model is obtained;Classification model is tested based on test data set, in the case where test result is classification model meets preset condition, it is determined that the classification model training is completed;Based on the classification model of training completion and text recognition unit, to be tested image is handled, to determine the class that round billet end face billet number direction in to-be-tested image belongs to.The based on deep learning's round billet end face billet number direction classification scheme provided in application can efficiently and accurately identify round billet end face billet number direction.

The application provides a webpage information extraction and classification method and device, and belongs to the technical field of artificial intelligence. The webpage information extraction and classification method comprises the following steps: converting the source code of a target webpage into a dom tree; processing each node of the dom tree to obtain four feature matrices, i.e., a text feature matrix, an Xpath feature matrix, a layout feature matrix and a visual feature matrix; inputting the four feature matrices into an encoding network respectively to obtain four representation vectors; performing feature fusion on the four representation vectors to obtain fused features; inputting the fused features into a classification network to obtain and store the classification result of an information unit of the target webpage, wherein the classification result comprises at least one of the following: a table, a form to be filled, a text unit that needs to be linked to a next webpage for display, a navigation bar or a display bar, pure text, an advertisement and useless information. The technical scheme of the application can improve the universality of the webpage information extraction and classification scheme.

This invention proposes a structurally enhanced discriminative width learning system for image classification. First, it introduces graph regularization based on class relationships to ensure that similar samples remain closely distributed in the projection space. Second, it adds intra-class divergence regularization to further enhance the aggregation of samples within the same class. Furthermore, it introduces inter-class divergence regularization to ensure that the centers of different classes are far apart in the feature space. Finally, in the output layer regression stage, it replaces the Frobenius norm regularization in ridge regression with sparse norm regularization to effectively suppress noise and redundant information. This image classification scheme, while maintaining the simplicity and efficiency of the optimization process, organically integrates three different types of structured information—intra-class structure, inter-class structure, and local manifold information—into a unified model, enabling the learned feature representations to possess stronger structure preservation and discriminative capabilities.

The application provides a data mining-based storage period prediction method and system, relates to the technical field of archive management, and comprises the following steps: analyzing a classification scheme and a storage period table to construct a labeled sample set, and training a period prediction mapping relationship by combining the text and metadata features of historical archiving data. For the data to be predicted, the modified period result is calculated by determining the rule matching degree, and the source system identifier is bound and returned. The application can improve the accuracy and automation level of the storage period prediction.

The invention discloses a forest multi-type disturbance remote sensing classification scheme under an unbalanced small sample condition, and the scheme comprises the steps: carrying out the classifier training of a limit gradient lifting classifier and a lightweight gradient lifting classifier through employing the manually labeled forest disturbance type sample data; generating a first classifier and a second classifier, and performing classification prediction on the unlabeled forest disturbance detection result based on the two classifiers to obtain amplified forest disturbance classification prediction result data, and performing pseudo sample generation on the amplified forest disturbance classification prediction result data by using a potential diffusion model to obtain first forest disturbance type data; and based on the first forest disturbance type data, iteratively performing forest disturbance type sample data amplification and classifier training by adopting a mode of mutual guidance of a first classifier and a second classifier to obtain a target limit gradient lifting classifier and a target lightweight gradient lifting classifier, and performing classification prediction on the unlabeled forest disturbance data. The forest multi-type disturbance classification result is obtained, and the forest multi-type disturbance classification precision can be improved.

The process is for use with: a plurality of data sets including a data set for each of a plurality of variants of a product, each set including data about the variant for which the set is provided, the data about the variant for which the set is provided including details of at least portions of a spectra created by NMR analysis of the variant; and a predetermined arrangement of a predetermined number of cells. The process comprises defining a classification scheme. The scheme divides that part of an NMR spectrum which is present in the plurality of sets into a plurality of parts equal in number to the predetermined number of cells and assigns each of said parts to a unique cell of the predetermined arrangement; and defines a relationship between the range of NMR intensities found in the sets and a predetermined portion of the visible spectrum.

The invention discloses a physical constraint-fused time sequence intelligent correction method, which comprises the following steps of: carrying out double clustering based on a meteorological station geographic position and a meteorological element time sequence curve shape by adopting a self-adaptive evolutionary optimization geographic space clustering model; an optimal classification scheme is selected through a weighted evaluation index of the contour coefficient and the intra-cluster distance quadratic sum; on the basis of the classification result, noise identification and marking are carried out on each type of meteorological element data by jointly using a local abnormal factor algorithm and dynamic physical constraint conditions based on cut-in wind speed, cut-out wind speed and a theoretical power curve; constructing a mixed feature refining model, and forming a feature set with high expression ability; carrying out 15-minute single-step prediction by using a self-adaptive dynamic correction prediction model driven by mixed features; comprehensively evaluating the prediction result; the prediction precision is obviously improved.

The invention discloses a model classification method and device, equipment, a storage medium and a computer program product, which are used for solving the problems that an existing model classification scheme cannot deeply analyze an internal structure of a model only according to a name, creation time or a few surface parameters of the model, so that a classification result is rough, the precision is insufficient, and the efficiency is high. And refined model management and scenarized calling requirements are difficult to support. The method comprises the steps of obtaining model structure parameters of a to-be-classified model; determining a to-be-classified parameter coefficient corresponding to the to-be-classified model according to the model structure parameter; a predetermined parameter domain reference coefficient corresponding to each classification parameter domain is obtained, and each parameter domain reference coefficient is determined based on a structure parameter coefficient of a preset experiment model of the corresponding classification parameter domain; determining a deviation value between the to-be-classified parameter coefficient and each parameter domain reference coefficient; and determining a target classification parameter domain corresponding to the to-be-classified model according to the deviation value.

The invention relates to a training system and a training method for training an artificial neural network, in particular for medical applications. The training system comprises an image-processing device, which is configured to receive a set of acquired images, an artificial intelligence module, configured to implement an artificial neural network, and a controller device, configured to train the artificial neural network implemented by the artificial intelligence module, wherein the image-processing device is further configured to generate modified images based on a modification of the acquired images, wherein each of the modified images generated by the image-processing device contains less texture information than the respective acquired image from which they stem, and wherein the controller device is configured to receive a set of training images comprising at least part of the modified images and at least part of the acquired images and train an artificial neural network which, based on a classification scheme, is configured to output a probability distribution for each training image according to the classification scheme.

The invention relates to the technical field of oil and gas field development geology and geophysics, in particular to an ultra-deep clastic rock fractured phase characterization method based on multi-scale data fusion, which comprises the following steps: fusing field outcrop, logging and seismic data, and adopting a'trend constraint-variation function quantization-collaborative simulation 'three-step geostatistical model to characterize fractured phases of ultra-deep clastic rocks. The method comprises the following steps: firstly, establishing a clastic rock fracture facies classification scheme including a fault control sandstone type, a fault control sand-mud interbed type, a pleat control sandstone type and a pleat control sand-mud interbed type on the basis of a lithology combination, and finally, determining a control mechanism of different fracture facies spatial combinations (such as a seepage-shielding body) on a water invasion path and residual gas distribution. According to the ultra-deep clastic rock fracture phase characterization method based on multi-scale data fusion, the span of the fracture from static description to dynamic prediction is realized, and a precise geological basis is provided for efficient development and residual gas potential tapping of an ultra-deep fractured gas reservoir.

The invention relates to the technical field of rescue evaluation, and discloses a rescue evaluation system for intelligent triage classification. The system comprises a data acquisition unit, a feature reconstruction unit, a core feature extraction unit and a dynamic evaluation unit. The data acquisition unit obtains rescue site multi-source heterogeneous data, a wounded person multi-dimensional data set is formed through space-time alignment, and the data contains wounded person vital signs, injury condition images, environmental parameters and rescue resource states; the feature reconstruction unit fills the dynamic tensor of the data set to generate a standardized sign feature tensor and an abnormal feature identification set; the core feature extraction unit is combined with the injury condition classification knowledge base to construct an injury condition feature matrix, determine a dimension association graph and generate an injury condition evolution graph; the dynamic evaluation unit extracts feature vectors, calculates injury condition dimension entropy, and screens a core feature sequence to generate a wounded person priority classification scheme. The system can integrate multi-source information, realizes dynamic evaluation, and is suitable for on-site rescue evaluation of various disaster accidents.

The application discloses a communication engineering informatization system and method based on big data, relates to the technical field of engineering informatization, and realizes real-time access to an engineering management system, a sensor network and a user operation log, constructs a dynamic knowledge graph containing a sub-project topological relationship, data generation space-time distribution and a cross-project calling chain, generates a classification scheme through a sliding window clustering algorithm based on historical sub-project quantity distribution characteristics, dynamically calibrates a critical threshold of data storage priority integration in combination with an engineering complexity factor, analyzes sub-project dependency relationship for a new engineering, activates an optimization process when a joint index of the sub-project quantity and topological complexity exceeds the threshold, analyzes sub-project correlation scores based on space-time correlation and business semantic characteristics of historical calling data, and generates a storage integration strategy, thereby improving clustering efficiency, reducing manual configuration requirements and improving engineering deployment efficiency.

The invention discloses a region category estimation method, device and equipment, a medium and a program product, which can be applied to the map field and the traffic field. The region category estimation method comprises the steps of obtaining a region area of a target region, and detecting whether the region area is located in a preset area range; when the area of the region is in a preset area range, acquiring historical positioning information of the target region in a historical time period; dividing the historical time period into a first time period and a second time period according to the historical positioning information; based on the historical positioning information, the first time period and the second time period, the category of the target area in the target time period is estimated, and the universality of the area classification scheme can be improved.

A computer-implemented method comprises obtaining export data representing a first classification in a first database that is to be imported into a second database, the first classification including a first set of classification markings that correspond to a first classification scheme identifier of the first database and that determine access to the first database and including a first plurality of origin classifications, each original classification of the first plurality of origin classifications including a classification scheme identifier; determining that a specific original classification of the first plurality of origin classifications has a specific classification scheme identifier matching a second classification scheme identifier of the second database; importing the specific original classification into the second database as an imported classification.

The present disclosure relates to a classification method for classifying records by relevance to nodes of a hierarchical structure representing classification schemes for different classification options. The classification method includes receiving an input record having a plurality of record content features, and a context attribute indicating the received context of the input record. The method includes retrieving relationship data indicating one or more nodes of the hierarchical structure associated with the context attribute of the received input record. The method also includes determining a relevance score of one or more nodes in the nodes of the hierarchical structure to classify the input record. The relevance score of each of the one or more nodes is determined based on a comparison of the plurality of record content features of the input record relative to the classification option represented by the node, and further, wherein the relevance score of each of the one or more nodes is determined based on the retrieved relationship data.

The application provides a data preparation method and device for low-code GIS system use scheme recommendation, and relates to the technical field of geographic information systems. The structured data of the scheme generated by the method includes scheme classification, scheme description, coded data marked with user operation sequence, scheme coding structure, data corresponding to preset scheme execution efficiency indicators, data corresponding to preset scheme usage indicators, and the like. These rich data dimensions enable the preset large model to more accurately understand user needs after learning and training, find the most suitable scheme under the corresponding classification according to user needs, and optimize and improve the scheme according to the dialogue content.

The present disclosure provides systems, methods, and computer readable media for implementing security policies at the level of a software call stack. In one example, a method includes generating a call stack classification scheme; detecting a call stack during deployment of an application; classifying, using the call stack classification scheme, the detected call stack as one of an authorized call stack or an unauthorized call stack during a runtime of the application to produce a classification; and applying a security policy based on the classification.

The application discloses a lightweight clothing classification method based on improved RVFL for intelligent settlement terminals. In view of the problems of morphological interference caused by clothes folding and stacking in unmanned retail self-service settlement and low computing power of edge computing terminals, a fast and lightweight classification scheme is proposed. The method first performs PCA dimensionality reduction and denoising on the clothing images collected by the settlement table camera; a regularized random vector function link network is constructed as a classification model to enhance the fault tolerance and anti-interference ability in the cash register environment; and the PCA principal component direction is used as a guide to perform local joint optimization on the hyperparameters of RRVFL by using the Schrodinger optimization algorithm. The application breaks the dependence of traditional large models on cloud high computing power, greatly reduces the hardware cost of the cash register terminal, and can realize fast learning and model updating of new and seasonal clothing in tens of seconds in the local store, effectively guaranteeing the efficiency and pricing accuracy of unmanned cash register.

A system and related methods for generating class-specific data are disclosed. The data belong to an input space with an unknown initial probability distribution and a known classification scheme. From a relatively small, unbalanced dataset of samples with respect to the classification scheme in the input space, the system is programmed to learn a series of invertible transformations from the input space to a target space, a target probability distribution for the samples in the target space, and a trainable parameter probability distribution for each parameter of the target probability distribution to represent uncertainty information related to the target probability distribution. The system is programmed to further identify how to sample from each parameter probability distribution, which determine how to sample from the target probability distribution, to generate samples in the input space that are more likely to belong to specific classes.

The invention relates to a customer financial risk grading method and device, equipment, a medium and a product. The method comprises the steps of performing rule scoring on target customer data based on a preset rule engine and at least one risk index to obtain a first risk score value for evaluating dominant risks; performing model scoring on the feature vector of the target customer data based on a trained machine learning model to obtain a second risk score value for evaluating the hidden risk; the machine learning model is obtained by training historical customer data as a sample, and the historical customer data comprises judicial check and control data, blacklist data, suspicious list data and untrusted licensed pedestrian data; performing weighted fusion on the first risk score value and the second risk score value to obtain a final risk score; and mapping the final risk score into a corresponding customer risk level. According to the method, the defects that a pure rule system is high in subjectivity and poor in generalization ability are effectively overcome, and an objective and comprehensive financial risk grading scheme with prediction ability is provided.