Product
Patsnap Eureka
For R&D, Patsnap Eureka makes reading and utilizing patents & technical documents easy.
Patsnap Eureka AIR
Designed for self-driven R&D workflows. Generate viable solutions, solve complex R&D challenges, empower your innovation with AI.
Patsnap Eureka Materials
Designed for material experts only. Revolutionize your material R&D, from search, analyze, to developing new materials.
Feature
TechResearch
Generate reliable direction feasibility study reports for your R&D in just a few steps.
TechSeek
Discover and master advanced knowledge NOW. Basics, ideas, possibilities, all at once.
TechMind
As an expert in R&D Theories, TechMind can generates customized viable solutions instantly.
TechRisk
Analyze your overall solution with one click, know your potential R&D risks in advance.
TechMonitor
Get weekly tech updates, stay abreast of the latest tech innovations and key insights.
Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
33 results about "Multiview learning" patented technology
Efficacy Topic
Property
Owner
Technical Advancement
Application Domain
Technology Topic
Technology Field Word
Patent Country/Region
Patent Type
Patent Status
Application Year
Inventor
System and method for a multi view learning approach to anomaly detection and root cause analysis
A system and method for detecting anomalies in a communication network includes detecting first outliers in a first set of quality indicators for a cellular group, detecting second outliers in a second set of performance indicators for the cellular group, correlating the first outliers and the second outliers to produce an anomaly candidate, determining a confidence threshold for the anomaly candidate, and indicating a network anomaly in response to the confidence threshold exceeding a predetermined threshold.
Owner:FUTUREWEI TECH INC
System and Method for a Multi View Learning Approach to Anomaly Detection and Root Cause Analysis
A system and method for detecting anomalies in a communication network includes detecting first outliers in a first set of quality indicators for a cellular group, detecting second outliers in a second set of performance indicators for the cellular group, correlating the first outliers and the second outliers to produce an anomaly candidate, determining a confidence threshold for the anomaly candidate, and indicating a network anomaly in response to the confidence threshold exceeding a predetermined threshold.
Owner:FUTUREWEI TECH INC
Shop optimization addressing method based on transfer learning
The present invention discloses a shop optimization addressing method based on transfer learning. The method comprises: dividing areas according to an urban road network; employing social media in each area and sensor data to construct features, employing data such as urban planning to construct rule features, multiplying factors such as a distance by the corresponding weight, and employing multi-view learning to perform feature fusion. The transfer learning is employed to transfer large-city addressing knowledge with rich data bulk to a relatively small city to train the optimization shop addressing model of the small city. The rule features and the trained model are fused to obtain a final optimal addressing model. The multi-view learning is configured to fuse different views of data to effectively extend the data of other sources, and based on the transfer learning technology, small cities having small samples also can obtain good effects.
Owner:ZHEJIANG UNIV
Webpage classification method for semi-supervised multi-view learning
ActiveCN110413924ASolve redundancyEfficient divisionCharacter and pattern recognitionWebsite content managementWeb page categorizationAlgorithm
The invention relates to the technical field of Internet, in particular to a webpage classification method for semi-supervised multi-view learning which comprises the following steps of: obtaining data from a webpage, and establishing a training set; training a classifier through the marked training set; encoding the marked training set and the unmarked training set through a trained classifier toobtain sample features; performing density clustering on the sample features to obtain a clustering result; and classifying the samples of the unmarked training set according to a clustering result.According to the scheme, the marked training set is used for training the classifier; orthogonal constraints and adversarial similarity constraints are added on the basis of an existing multi-view classification method, density clustering marking is carried out on all data in a training set through a trained classifier, finally, accuracy verification is carried out on the classifier, and the classification performance of the classifier can be improved through multiple iterations of the process.
Owner:GUANGDONG UNIV OF PETROCHEMICAL TECH
Semi-supervised multi-view dictionary learning based color face recognition method
ActiveCN107392190AEasy to identifyCharacter and pattern recognitionLearning basedDictionary learning
The invention discloses a semi-supervised multi-view dictionary learning based color face recognition method. According to the method, a multi-view leaning technology is applied to semi-supervised color facial image dictionary learning. In the training stage, by learning structured dictionaries of all chrominance components separately and making the dictionaries be orthogonal to one another, the correlation among all the chrominance components is removed, and complementary difference information among all the chrominance components is fully utilized; in the dictionary learning process, class-label-free color facial image samples are used for participating in training, and all training sample information is fully utilized. In the class test stage, all the chrominance components are added up, the dictionary corresponding to each class of training samples is used for reconstructing reconstruction errors of test samples, and finally the test samples are classified as the class with the smallest accumulative reconstruction error. Accordingly, the recognition effect is higher, and by means of semi-supervised multi-view dictionary learning, the color face recognition capacity is obviously enhanced.
Owner:NANJING UNIV OF INFORMATION SCI & TECH
Science subject extraction method based on multi-view learning
ActiveCN103530316ASimple Data PreprocessingOvercome the lack of information on a single aspectSpecial data processing applicationsDocument management systemsDocumentation procedureData information
The invention provides a science subject extraction method based on multi-view learning. The extraction method includes the steps that thesis data are obtained from a thesis database to serve as target files where science subjects are to be extracted; data information of multiple views in the target files are extracted to serve as bases of science subject extraction; simple data pre-processing is carried out on the data information of each view, the data information of all the target files is expressed to form a data matrix, and data information of each target file is one row vector of the data matrix; by means of the multi-view learning method, the target files are clustered, the target files of the same kind correspond to the same science subject; the science subject of the target files of each kind is extracted and expressed in a mode of multiple key words. The method has the advantages of making up for the defect that in a traditional method, data information of only one aspect is considered, well making use of data information of various aspects and obtaining better science subject extraction effects by means of complementary relationships between the data information and consistent auxiliary clustering of potential subjects.
Owner:ZHEJIANG UNIV
Search engine user information demand satisfaction evaluation method capable of integrating multiple views and semi-supervised learning
ActiveCN105488522AEfficiently Assess Requirements SatisfactionImprove the evaluation effectWeb data indexingCharacter and pattern recognitionLabeled dataStudy methods
The invention relates to a search engine user information demand satisfaction evaluation method capable of integrating multiple views and semi-supervised learning. The method is divided into the following six stages: preprocessing data, training a subview satisfaction model, distributing a dummy tag for unlabeled data, training a user satisfaction model based on the multiple views and the semi-supervised learning, and carrying out evaluation. Through a semi-supervised learning method, a small quantity of labeled data and a great quantity of unlabeled data are used for improving the performance of an evaluation model, and a multi-view learning method is imported to overcome the problem that a traditional single-view based semi-supervised learning method is always caught in local optimum. The search engine user information demand satisfaction evaluation method has the beneficial effects: (1) under the condition of the small quantity of labeled data, the search engine user information demand satisfaction can be effectively evaluated; (2) the small quantity of labeled data and the great quantity of unlabeled data can be used for improving the evaluation performance of the user satisfaction model; and (3) a search process of the user can be independently described from angles of behaviors and time, and the model can be prevented from being caught into the local optimum through mutual learning.
Owner:ZHEJIANG HONGCHENG COMP SYST
Short-term traffic forecasting method based on multi-task multi-view learning model
ActiveCN109598939AReasonable methodBalance of spatio-temporal heterogeneityDetection of traffic movementTraffic predictionTemporal heterogeneity
The invention discloses a short-term traffic forecasting method based on a multi-task multi-view learning model. The method comprises the following steps: 1, constructing a space-time data model individually for each road segment; 2, constructing a multi-core learning model; 3, constructing an objective function by adopting a multi-task multi-view feature learning model; 4, introducing a particleswarm optimization algorithm to optimize the objective function obtained in the step 3; and 5, repeating the steps 1 and 2 for any road segment to obtain input features, bringing the input features into an optimized objective function, and realizing short-term traffic prediction for any road segment. The method realizes the efficient prediction of short-term traffic, solves the problem that the spatial-temporal heterogeneity and the global prediction ability of the model can not reach equilibrium, solves the problem of parameter optimization of the model, and can be widely used in urban planning, human mobility survey, automobile navigation, emergency response, space-time accessibility analysis and traffic pollution modeling.
Owner:INST OF GEOGRAPHICAL SCI & NATURAL RESOURCE RES CAS
High-speed train operation condition identification method
ActiveCN107563403AHigh feasibilityEasy to operateCharacter and pattern recognitionFeature vectorVariational mode decomposition
The invention discloses a high-speed train operation condition identification method, which is characterized by, to begin with, carrying out pretreatment on vibration signal data of a high-speed train; then, carrying out variational mode decomposition on different channels of monitoring data to obtain mode function characteristics; calculating fuzzy entropy correspondingly, and obtaining feature vectors of different conditions under of a plurality of channels as samples to be tested in multi-view learning; and finally, carrying out identification through a Multi-View K-means cluster. The method has the advantages of more complete knowledge learning, simpler flow and higher operability and the like, and is mainly applied to identification of operation conditions of the high-speed train.
Owner:SOUTHWEST JIAOTONG UNIV
MiRNA-disease association prediction method and device based on graph neural network fusion multi-view information
ActiveCN112784913AEasy to captureImprove forecast accuracyCharacter and pattern recognitionNeural architecturesMulti omicsPredictive methods
The invention discloses a miRNA-disease association prediction method and device based on graph neural network fusion multi-view information, and the method integrates miRNA-disease-related multi-omics data to construct a plurality of views, not only considers a plurality of homogeneous similarity networks, but also considers a heterogeneous bipartite network, extracts node features on all views in combination with a graph neural network and multi-view learning, captures dependency between global features and local features through a discriminator, and can better capture the complex nonlinear relation between miRNA and diseases.
Owner:HUNAN UNIV
Medical image classification method and device based on multi-view learning and depth supervision auto-encoder
InactiveCN112488102AImprove generalization abilityImprove classification accuracyImage enhancementImage analysisFeature vectorMedicine
The invention discloses a medical image classification method and device based on multi-view learning and a deep supervision auto-encoder, and the method comprises the following steps: 1, carrying outthe wavelet decomposition of a region of interest of a medical image, and obtaining a multi-frequency sub-band; step 2, defining each sub-band as a view, and quantitatively extracting an image omicsfeature from each view so as to obtain a multi-view feature; step 3, constructing a classification network of a deep supervision auto-encoder based on multi-view feature learning, and training the classification network based on morphological multi-view feature vectors of the image samples and classification tags thereof to obtain a trained classification model; and step 4, classifying the imageswith unknown classification labels based on the trained classification model. According to the invention, the classification accuracy of medical images can be improved.
Owner:CENT SOUTH UNIV
Bearing fault diagnosis method based on multi-view associated feature learning
PendingCN113255458AHigh precisionImprove reliabilityMachine part testingCharacter and pattern recognitionEngineeringLearning network
The invention discloses a bearing fault diagnosis method based on multi-view correlation feature learning. The method is characterized in that a vibration signal and a current signal are regarded as different views, a correlation feature learning method of a gear box bearing vibration signal and a generator current feature is designed based on multi-view learning, and the method is applied to the multi-fault diagnosis of a wind power gear box bearing. The method comprises the following steps of firstly, extracting wavelet packet sub-band time domain statistical features from the vibration and current signals to obtain an initial vibration feature space and an initial current feature space, and then inputting the vibration and current feature samples into a canonical correlation learning network in pairs to carry out correlation feature learning, so that the correlation between current and vibration signal feature mapping is maximum, and the enhanced extraction of the vibration and current features is realized. According to the method, the correlation attributes in the vibration and current signals can be learned in an unsupervised mode, the common fault feature information is obtained, the comprehensive diagnosis advantage of multiple sensing signals is fully utilized, and compared with a single signal feature method, the precision and reliability of fault diagnosis are improved.
Owner:YANSHAN UNIV
City water logging analysis method based on transfer learning
InactiveCN107025303AImprove scalabilityReduce dataData processing applicationsNatural language data processingSmall sampleSocial media
The invention discloses a city water logging analysis method based on transfer learning. The method comprises steps as follows: firstly, an area is divided according to characteristics of a city; then, features are constructed by use of social media in each area and sensor data, and the features are fused to the best by use of multi-view learning; finally, big city water logging knowledge with the rich data volume is transferred to smaller cities through transfer learning, and a water logging severity degree model for the smaller cities is trained. Different view data are fused through multi-view learning, data from other sources can be effectively extended, and meanwhile, on the basis of a transfer learning technology, the method can also have a better effect for small cities with smaller samples.
Owner:ZHEJIANG UNIV
A method and system for ship tracking
InactiveCN109102528AReduce tracking errorImprove abstractionImage enhancementImage analysisComputation complexitySize change
The invention discloses a method and system for ship tracking. The ship tracking operator STMS based on multi-view learning mechanism and sparse representation mechanism makes use of an LoG operator,a HOG operator, an LBP operator, a Canny operator and a Gabor operator to extract ship features respectively, the ship features extracted from STMS ship tracking operator are sparsely expressed by sparse representation mechanism, the STMS model is established to represent the ship tracking targets, and the LASSO penalty mechanism is introduced to balance the tracking accuracy and computational complexity of the STMS model. The tracking performance of the STMS model is optimized by using the Frobenius normal form, and the ship tracking results are obtained by solving the STMS model with the accelerated approximate gradient method. The invention has better abstractness and robustness, can effectively cope with a variety of ship tracking challenges such as illumination change, imaging visualangle change, ship occlusion and ship imaging size change, and has smaller ship tracking error and good tracking effect.
Owner:SHANGHAI MARITIME UNIVERSITY
Space-time missing data completion method and device based on multi-view learning, and medium
ActiveCN113806349AGood nonlinear relationship fitting abilityGood estimateNeural architecturesGeographical information databasesMissing dataData set
The invention discloses a space-time missing data complementing method and device based on multi-view learning, and a medium, and the method comprises the steps: constructing a data tensor comprising three dimensions of a monitoring station, monitoring time and monitoring parameters for each data missing monitoring station in a massive and multi-type environment monitoring data set; decomposing for forming the three mixed view matrixes, carrying out missing data estimation on the three constructed mixed views from local and global scales by using a matrix complete algorithm, and carrying out multi-view learning based on an artificial deep neural network to obtain a final estimation result of a monitoring parameter missing value. According to the method, missing value completion of the environment monitoring data set can be realized, and the method has important significance in the fields of marine environment monitoring, smart city development and the like.
Owner:ZHEJIANG UNIV
Multi-view learning method and system for rhesus monkey eye movement decision decoding
ActiveCN110432900AImprove generalization abilityImprove computing efficiencyDiagnostic recording/measuringSensorsNODALComputation complexity
The invention discloses a multi-view learning method and system for rhesus monkey eye movement decision decoding, and belongs to the field of multi-view decoding in intrusive brain-computer interfaces. The multi-view learning method comprises the steps that an eye movement decision decoding model comprising a characteristic node extraction network, a reinforced node extraction network and a predicting network is established; eye movement decision direction data obtained after encoding of the local field potential and action potential of a medial frontal cortex auxiliary eye area in a rhesus monkey intrusive brain-computer interface, and one-hot encoding are input into the eye movement decision decoding model for training, and thus the trained eye movement decision decoding model is obtained; and the to-be-decoded local field potential and action potential are input into the trained decoding model, and thus a decoding result of the eye movement direction is obtained. The multi-view learning method and system are less in using limitation, low in computation complexity and suitable for decoding of action potential and local field potential signals in the intrusive brain-computer interface and other various multi-view learning scenes.
Owner:北京烽火万家科技有限公司
Fraud marking method and device based on multi-view learning
The invention discloses a fraud marking method and device based on multi-view learning, and the method comprises the steps: obtaining a user data set; dividing the user data set into a marked data setand an unmarked data set; respectively training a plurality of view classifiers on the marked data set based on different views; judging and marking partial data in the unmarked data set by using theplurality of trained view classifiers; adding the marked data into a marked data set to form a new marked data set; circularly carrying out the action of the new marked data set until a preset stop condition is reached; fusing the plurality of view classifiers to generate a secondary classifier; and judging and marking the data by using the secondary classifier. According to the method, the interaction, advantage complementation and collaborative learning among views are exerted by utilizing different learning difficulties of the data in different views, so that the effect of performing fraudmarking by adopting less and lower-cost sample data is achieved.
Owner:SHANGHAI QIFU INFORMATION TECH CO LTD
Multi-task learning method and system based on feature and sample adversarial symbiosis
PendingCN111767949AImprove generalization abilityAddressing Domain Distribution DifferencesCharacter and pattern recognitionNeural architecturesSmall sampleMulti-task learning
The invention relates to the field of multi-task deep learning, and provides a multi-task learning method based on feature and sample adversarial symbiosis, which comprises the following steps: S1, randomly extracting samples of tasks, and generating common implicit features irrelevant to the field; s2, based on the common implicit characteristics generated in the step S1, generating a high-simulation sample, and taking the high-simulation sample as a task sample of the next cycle in the step S1; s3, circulating the steps S1 and S2 until the multi-task adversarial game is balanced, and generating a final high-simulation sample and a high-quality classification label. The invention further provides a multi-task learning system based on feature and sample adversarial symbiosis. According tothe method, the problems of domain distribution difference and small samples are solved, and the generalization performance of a machine learning system is greatly improved, so that a plurality of application fields of artificial intelligence are promoted to be broken through. The method is not only suitable for multi-task learning and transfer learning, but also suitable for multi-view learning and multi-modal learning.
Owner:SOUTH CHINA NORMAL UNIVERSITY
Social security character portrait method based on multi-view learning
ActiveCN110046657AAccurate mappingData processing applicationsCharacter and pattern recognitionPhysical spaceData set
The invention discloses a social security character portrait method based on multi-view learning. In social security application of multi-source identity big data, complete priori knowledge and annotation samples are difficult to obtain, a multi-view cooperative training model based on part of domain knowledge, a small number of annotation samples and a large number of unannotated samples is constructed, and accurate mapping from identity attributes to portrait labels is achieved. The method comprises the following steps: firstly, learning three attribute sub-views including a physical space,a social space and a network space and corresponding weights from ternary space identity attributes through ternary space view decomposition of a labeled sample data set; secondly, portrait label classification is conducted on unlabeled samples through a ternary space multi-view classifier, credible classified samples are generated in combination with domain knowledge voting and added to a labeledsample data set, and labeled samples are enriched; the method has an important application value in the field of social safety.
Owner:SHENZHEN RES INST OF WUHAN UNIVERISTY +1
Radio frequency fingerprint identification method based on sample multi-view learning
PendingCN114218984AImprove recognition efficiencyImprove recognition accuracyCharacter and pattern recognitionNeural architecturesData setNoise (radio)
The invention discloses a radio frequency fingerprint identification method based on multi-view learning. The radio frequency fingerprint identification method is characterized by comprising the following steps: 1) acquiring a radio frequency signal of each wireless device to be identified; 2) processing data and adding noise; 3) segmenting the data and making a data set I, a data set II and a data set III; 4) designing a complex numerical neural network sub-assembly and constructing a complex numerical neural network; 5) building a sub-neural network I; 6) building a two-dimensional convolutional neural network II and a two-dimensional convolutional neural network III; according to the method, sample data can be utilized, a computer can automatically extract signal fingerprint features, the requirement for the number of samples is lowered, meanwhile, the recognition precision at the low signal-to-noise ratio can be improved, the data feature extraction efficiency is high, and the recognition precision is high.
Owner:GUILIN UNIV OF ELECTRONIC TECH
Method for predicting dissolved oxygen in sewage biochemical treatment
PendingCN113889194AImprove accuracySolve the missing puzzleWater treatment parameter controlCharacter and pattern recognitionData setDecomposition
The invention provides a method for predicting dissolved oxygen in sewage biochemical treatment. The method comprises the steps: S1, collecting and preprocessing water quality data in sewage biochemical treatment, obtaining a data set, and performing variational mode decomposition on the factors influencing the water quality to obtain decomposition components; S2, according to the data set in the step S1, judging the relationship between the decomposition components and the dissolved oxygen, carrying out clustering analysis, and dividing the decomposition components into a tight group and a loose group; and S3, according to the data set in the step S1, carrying out multi-task learning training on the tight group, carrying out multi-view learning training on the loose group, establishing a prediction model, and carrying out dissolved oxygen prediction and auxiliary task prediction. According to the prediction method, dissolved oxygen prediction and auxiliary task prediction of related factors are combined, and the accuracy of future dissolved oxygen prediction is remarkably improved.
Owner:SHANGHAI SIIC LONGCHUANG SMARTER ENERGY TECH CO LTD
Multi-view learning algorithm based on random forest
PendingCN111798003AEnsemble learningCharacter and pattern recognitionGeneration processDecision boundary
Random forests are one of the most classical machine learning algorithms and have been widely applied. However, observation finds that although numerous multi-view data exist in reality and wide analytical studies have been obtained, it is surprising that random forest construction for a multi-view scene is very few. Only methods for solving the multi-view learning problem by using random forestsare used for generating respective random forests for each view, and then multi-view information is fused during decision making. One significant defect of the method is that the correlation among multiple views is not utilized in the construction stage of the random forest, so that information resources are undoubtedly wasted. In order to make up the defect, the invention provides an improved multi-view learning algorithm based on random forest. Specifically, view fusion is carried out in the generation process of decision trees, information interaction between views is fused into the construction stage of the decision trees, and utilization of complementary information between the views in the whole random forest generation process is achieved. In addition, decision boundaries with discrimination properties are generated for the decision tree through discriminant analysis, so that the algorithm is more suitable for classification.
Owner:NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
A multi-view learning method and system for rhesus monkey eye movement decision decoding
ActiveCN110432900BImprove generalization abilityImprove computing efficiencyDiagnostic recording/measuringSensorsComputation complexityMedicine
The invention discloses a multi-view learning method and system for rhesus monkey eye movement decision decoding, and belongs to the field of multi-view decoding in intrusive brain-computer interfaces. The multi-view learning method comprises the steps that an eye movement decision decoding model comprising a characteristic node extraction network, a reinforced node extraction network and a predicting network is established; eye movement decision direction data obtained after encoding of the local field potential and action potential of a medial frontal cortex auxiliary eye area in a rhesus monkey intrusive brain-computer interface, and one-hot encoding are input into the eye movement decision decoding model for training, and thus the trained eye movement decision decoding model is obtained; and the to-be-decoded local field potential and action potential are input into the trained decoding model, and thus a decoding result of the eye movement direction is obtained. The multi-view learning method and system are less in using limitation, low in computation complexity and suitable for decoding of action potential and local field potential signals in the intrusive brain-computer interface and other various multi-view learning scenes.
Owner:北京烽火万家科技有限公司
Loan default risk detection method based on multi-view learning and related equipment
PendingCN112651821AImprove accuracyGood effectFinanceCharacter and pattern recognitionMultiview learningBig data
The embodiment of the invention belongs to the field of big data, and relates to a loan default risk detection method based on multi-view learning; the method comprises the following steps: dividing default detection data features into a borrowing target feature view and a borrower feature view; training a first detection classifier and a second detection classifier based on the labeled samples and the rejected samples to obtain a trained first detection classifier and a trained second detection classifier; setting voting weights for the trained first detection classifier and second detection classifier according to a preset weight strategy to obtain a multi-view default risk detection model; and inputting to-be-detected target data into the multi-view default risk detection model to obtain a loan default risk detection result. By adopting the method, the loan default risk detection accuracy is improved.
Owner:PINGAN PUHUI ENTERPRISE MANAGEMENT CO LTD
Color face recognition method based on semi-supervised multi-view dictionary learning
Owner:NANJING UNIV OF INFORMATION SCI & TECH
A short-term traffic prediction method based on multi-task multi-view learning model
ActiveCN109598939BReasonable methodBalance of spatio-temporal heterogeneityDetection of traffic movementTraffic predictionTemporal heterogeneity
The invention discloses a short-term traffic forecasting method based on a multi-task multi-view learning model. The method comprises the following steps: 1, constructing a space-time data model individually for each road segment; 2, constructing a multi-core learning model; 3, constructing an objective function by adopting a multi-task multi-view feature learning model; 4, introducing a particleswarm optimization algorithm to optimize the objective function obtained in the step 3; and 5, repeating the steps 1 and 2 for any road segment to obtain input features, bringing the input features into an optimized objective function, and realizing short-term traffic prediction for any road segment. The method realizes the efficient prediction of short-term traffic, solves the problem that the spatial-temporal heterogeneity and the global prediction ability of the model can not reach equilibrium, solves the problem of parameter optimization of the model, and can be widely used in urban planning, human mobility survey, automobile navigation, emergency response, space-time accessibility analysis and traffic pollution modeling.
Owner:INST OF GEOGRAPHICAL SCI & NATURAL RESOURCE RES CAS
A Web Page Classification Method Based on Semi-Supervised Multi-view Learning
ActiveCN110413924BSolve redundancyEfficient divisionCharacter and pattern recognitionWebsite content managementWeb page categorizationThe Internet
The present invention relates to the technical field of the Internet, and more specifically, relates to a webpage classification method for semi-supervised multi-view learning, comprising: obtaining data from webpages and establishing a training set; training a classifier through a marked training set; The machine codes the marked training set and the unmarked training set to obtain the sample features; performs density clustering on the sample features to obtain the clustering results; and classifies the samples of the unmarked training set according to the clustering results. This scheme uses the labeled training set to train the classifier, adds orthogonal constraints and confrontational similarity constraints on the basis of the existing multi-view classification method, and then uses the trained classifier to perform density aggregation on all the data in the training set. Finally, the accuracy of the classifier is verified, and the above-mentioned process can be iterated multiple times to improve the classification performance of the classifier.
Owner:GUANGDONG UNIV OF PETROCHEMICAL TECH
A Method for Satisfaction Evaluation of Search Engine Users' Information Needs by Fusion of Multi-view and Semi-Supervised Learning
ActiveCN105488522BEfficiently Assess Requirements SatisfactionImprove the evaluation effectWeb data indexingCharacter and pattern recognitionEngineeringSupervised learning
The invention relates to a search engine user information demand satisfaction evaluation method that combines multi-view and semi-supervised learning. The user satisfaction model of supervised learning and six stages of evaluation, the present invention uses a small amount of labeled data and a large amount of unlabeled data to improve the performance of the evaluation model through the method of semi-supervised learning, and introduces the idea of multi-view learning to overcome the traditional Semi-supervised learning methods of views are prone to fall into the problem of local optima. The beneficial effects are: (1) the user satisfaction of search engine information needs can be effectively evaluated with a small amount of labeled data; (2) the user satisfaction model evaluation can be improved by using a small amount of labeled data and a large amount of unlabeled data performance; (3) Describe the user's search process from the perspective of behavior and time, and avoid the model from falling into local optimum through mutual learning.
Owner:ZHEJIANG HONGCHENG COMP SYST
Spatio-temporal missing data completion method, device and medium based on multi-view learning
ActiveCN113806349BGood nonlinear relationship fitting abilityGood estimateGeographical information databasesNeural architecturesMissing dataData set
Owner:ZHEJIANG UNIV
Forest fire identification method based on multi-view robust bilateral twin vector machine
PendingCN112348087AMinimize Structural RiskImprove generalization abilityCharacter and pattern recognitionSupport vector machineAlgorithm
The invention discloses a forest fire identification method based on a multi-view robust bilateral twin support vector machine. The method comprises the following steps: developing a new optimizationmodel: adopting a robust multi-view learning algorithm of a bilateral twin support vector machine: applying an MvRDTSVM to forest fire identification, and performing an experiment on a forest fire data comparing four single-view and multi-view methods by utilizing real image data, and testing robustness and generalization performance of the forest fire database; expressing the MvGSVM as an SVM type problem again, meanwhile, introducing bilateral constraints, taking an L1 norm as a distance measurement mode in a target function, and improving the robustness of the model. Because the targetfunction is non-convex and non-smooth, the invention designs a new effective iterative algorithm and theoretically proves the convergence of the algorithm, and because a series of QPP problems need to be solved in the iterative process, the calculation cost is increased, and the rapid version of the MvRDTSVM, namely the MvFRDTSVM, is further developed. By solving a series of linear equations rather than the QPP problems, the calculation speed is greatly increased, and the calculation cost is saved.
Owner:NANJING FORESTRY UNIV
Who we serve
- R&D Engineer
- R&D Manager
- IP Professional
Why Patsnap Eureka
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com