225 results about "Multiple classification" patented technology

A system and method that automatically, transparently and securely controls software execution by identifying and classifying software, and locating a rule and associated security level for executing executable software. The security level may disallow the software's execution, restrict the execution to some extent, or allow unrestricted execution. To restrict software, a restricted access token may be computed that reduces software's access to resources, and / or removes privileges, relative to a user's normal access token. The rules that control execution for a given machine or user may be maintained in a restriction policy, e.g., locally maintained and / or in a group policy object distributable over a network. Software may be identified / classified by a hash of its content, by a digital signature, by its file system or network path, and / or by its URL zone. For software having multiple classifications, a precedence mechanism is provided to establish the applicable rule / security level.

A method of organizing a set of recognition models of known objects stored in a database of an object recognition system includes determining a classification model for each known object and grouping the classification models into multiple classification model groups. Each classification model group identifies a portion of the database that contains the recognition models of the known objects having classification models that are members of the classification model group. The method also includes computing a representative classification model for each classification model group. Each representative classification model is derived from the classification models that are members of the classification model group. When a target object is to be recognized, the representative classification models are compared to a classification model of the target object to enable selection of a subset of the recognition models of the known objects for comparison to a recognition model of the target object.

In embodiments of the present invention improved capabilities are described for identifying a first classification scheme associated with product attributes of a first grouping of products, identifying a second classification scheme associated with product attributes of a second grouping of products, and receiving a record of data relating to an item, the classification of which is uncertain. It may also involve receiving a dictionary of attributes associated with products and assigning the item to at least one of the classification schemes based on probabilistic matching among the attributes in the classification schemes, the attributes in the dictionary of attributes and the known attributes of the item.

A method for sharing a data set including multiple classifications of security between one or more entities and / or between one or more organizations. Public-keys for use in selectively encrypting the data set with multiple classifications of security are made available in a public-key table. A private-key corresponding to the public-key is used to decrypt the selectively encrypted data set. Public-keys available in the public-key table are re-used to selectively encrypt data sets as appropriate. Public / private-key pairs also may be generated as needed for use in encrypting a document with multiple classifications of security. A single data set may be further encrypted with additional classifications of security as needed. A data set thus may be made available to various entities and / or organizations over a common repository. Symmetric and other encryption techniques also may be used.

The invention discloses a multi-model fused short text classification method. The multi-model fused short text classification method comprises a learning method and a classification method. The learning method comprises the following steps: carrying out word segmentation and filtration on short text training data to obtain a word set; calculating the IDF value of each word; calculating the TFIDF values of all the words and constructing a text vector VSM; and carrying out text learning on the basis of a vector space model, and constructing an ontology tree model, a keyword overlapping model, a naive Bayesian model and a support vector machine model. The classification method comprises the following steps: carrying out word segmentation and filtration on a to-be-classified short text; generating a text vector on the basis of the support vector machine model; respectively classifying by using the ontology tree model, the keyword overlapping model, the naive Bayesian model and the support vector machine model to obtain single model classification results; and fusing the single model classification results to obtain a final classification result. According to the method disclosed in the invention, multiple classification modes are fused and the short text classification correctness is improved.

The invention discloses a coastal city time sequence land utilization information extracting method. The method comprises the following steps: acquiring a remote-sensing image Landsat, and preforming atmospheric correction on the same; constructing a remote-sensing classification feature index database by selecting a group of remote-sensing classification features; acquiring data elevation image DEM data to obtain elevation data and slope data; constructing a decision rule of single-classification feature index or multiple classification feature indexes according to different land utilization types of the coastal city based on a multi-feature decision tree model, classifying the coastal city land utilization step by step according to the rule, and finally determining various branches of the decision tree, detecting the time sequence remote-sensing image change, and distinguishing a mistaken classification land type and a missed classification land type, wherein the method further comprises the content of two parts: evaluating classification precision, and outputting the land utilizationclassification map extracted based on the decision tree model. By use of the extracting method disclosed by the invention, the coastal city land utilizationclassification precision can be greatly improved, and a key problem in the coastal city land utilizationclassification is solved.

The invention provides an egg embryo classification based on deep learning. The method comprises the following steps of: collecting embryo images of five days, and dividing the embryo images into three classes of samples including a normal embryo, a suspension embryo and a non-insemination embryo; preprocessing the embryo images, extracting the ROI (Region Of Interest) of each image, and carrying out normalization on the sizes of the images; combining with a transfer learning method, using a CNN (Convolutional Neural Network) model pre-trained by an AlexNet classical network in ImageNet to carry out fine tuning training on a target set; and utilizing the trained model to distinguish an image to be detected. Compared with the prior art, the egg embryo classification is characterized in that the problems of the multiple classifications of the CNN model on an aspect of a small-scale egg embryo data set can be successfully solved, the egg embryo classification is high in accuracy, and the engineering requirements of the detection and the classification of egg embryo survival can be met.

The invention provides a user data classification method. The method comprises the steps that user data characteristics are generated; a tagged data set and a untagged data set of user data are generated according to tagging rules; a positive sample tagged data set P and an unknown sample data set U of one classification in multiple classifications are built according to the tagged data set and the untagged data set; a classifier is generated according to the positive sample tagged data set P, the unknown sample data set U and the corresponding user data characteristics; the classifier is used for determining whether the user data in the untagged data set belongs to the classification or not. Accordingly, the user data is classified through an improved positive sample tag learning algorithm, the method is suitable for characteristic extraction of people, people having the similar life stages in a system are mined, and then electric business advertisements achieving precise people orientation are provided.

The invention discloses a medical insurance fraud detection method based on multiple features. The method comprises: aimed at treatment histories of all patients suffered from the same disease in medical insurance declaration data, in combination with medicine classification knowledge, adopting probability statistics, mixture Gaussian modeling, feature fusion and other techniques to extract multiple secondary feature data with higher distinction degree; then, vectorizing the treatment histories of the patients based on the secondary feature data; and then, after carrying out clustering analysis on treatment data marked with 'normal', adopting a non-linear support vector machine classification technique to establish multiple classification hyperplanes for each type of normal treatment data subjected to clustering and treatment data marked with 'fraud', so that the fraud detection can be carried out on non-marked medical insurance data. The method can be used for quickly and effectively detecting the fraud data existent in the medical insurance data, and has relatively high accuracy.

The invention discloses a method of face recognition. Firstly, known and classified face samples are trained for modeling via pairs of training modules to produce a support vector machine classifier capable of classifying and recognizing unknown face samples, then, features of the unknown face samples are extracted via recognition modules, and to-be-recognized recognition samples after being extracted out features are transferred to a trained support vector machine classifier to judge the faces which category belong to. The invention provides the method of face recognition based on nuclear stochastic mapping and support vector machines, and the method of nuclear stochastic mapping is initially used for extracting more effective face features in the process of extraction of the face features. In addition, the support vector machine classifier used for dichotomy classification is employed in the problem of face multiple classification, a pair of multi-strategies are employed to convert multi-problems into two-problems, and the invention not only can largely reduce feature dimensions for recognizing faces, but also can remarkably increase the recognition capability for the faces.

A trusted relabeler system and software is provided that allows for the simultaneous reclassification of multiple files to multiple classification labels (security domains or computer networks) through automation of the multi-person review process. Roles, used to break down the requirements of the multi-person review process, dictate what specific function a user can perform. There are three roles used: Submitter, Processor, and Reviewer. The Submitter is allowed to submit one or more documents for reclassification, while the Processor is allowed to submit one or more documents for reclassification, perform an initial review and acceptance of the document, and assign a specific Reviewer for the document. The Reviewer is allowed to perform final verification and, upon acceptance, allows the document to be reclassified appropriately. Users are assigned one or more roles and a set of source and destination classification labels inside the security policy of a specific user that are bounded (restricted) by the clearance (maximum classification) assigned to a user on the trusted operating system on which the trusted relabeler system is implemented. The bundled documents to be reclassified are virus scanned, file typed, and searched for “dirty words” indicative of a particular classification level. An integrity seal is generated to verify that the files are not tampered with between processing by the Processor and the Reviewer. Files are upgraded or downgraded only if all processes are completed successfully.

A concentrated opposite blowing multiple classification NOx combustion method of the protection environmental technology area. The invention is that along the furnace height direction fall into the main combustion chamber, reducing area and burnout area, in the main area the wind fall into two shades in a level direction by firebox , a strong wind rip into the furnace burning side, after the promotion is reversed to change, and the secondary wind cut into the furnace, joint with light wind and form rotary stability aerodynamic field, so most of the wind pulverized coal is led to the furnace central region, ultimately in the heater form a light wind wrap a strong wind, a second air parcels light wind, and wind wrap powder situation, a second wind and light wind in the near wall region of water wall form oxidizing atmosphere, away from water wall region the pulverized coal burning. The invention achieves ultra-low NOx emission, while the burning area prevents to slag and ensure the high combustion efficiency etc. characteristics.

The invention discloses a short text classification method and apparatus. The method comprises the steps of performing word segmentation preprocessing on to-be-classified short texts and obtaining an extended word of each word obtained by word segmentation; obtaining weight values of each word and the extended word of each word according to a pre-constructed lexical item set; according to the weight values, obtaining a probability of each type that a short text belongs to by utilizing a plurality of SVM classification models; and determining the type that the short text belongs to according to a preset probability classification model. According to the short text classification method, the problem of short text characteristic sparsity is solved, the complexity due to the adoption of multiple classification models is effectively lowered, and actual application requirements are better met.

The invention provides a video play method and device and a computer readable storage medium. The method comprises that trailer play request is received; according to the trailer play request, a target user as well as a target video to which a trailer to be played belongs is identified; a preference classification label of the target user is identified from multiple classification labels of the target video; a video segment matching the preference classification label of the target user is obtained from target videos; and the video segment is output. Different from that the same video introduction information including a title and poster of the video is displayed in all clients, the video segment matching the preference classification label of the user of the target video is output, videoinstruction information can be pushed to different terminals in a customized way, the hit rate of the target video in different terminal users is improved, and the rate of changing from the video segment to the whole video is improved.

The invention discloses a motion recognition and fatigue detection method and system based on gait information. The method includes the steps that current gait information of a user is acquired; the current gait information is subjected to data processing, a current motion behavior of the user and multiple classification results about whether the user is currently in a fatigue state or not are recognized according to multiple preset motion recognition and fatigue classification models, and the preset motion recognition and fatigue classification models are generated in the mode that the gait information generated when the user performs multiple motion behaviors in a non-fatigue state and the fatigue state serves as sample data, and the sample data is applied to a voting classification algorithm, based on integrated learning, in machine learning to be trained; the current motion behavior of the user and the final result about whether the user is currently in the fatigue state or not are acquired according to the classification results based on the principle that minorities submit to majorities. By means of the method, the injury risk caused by the user fatigue state under different motions is reduced, and the accuracy of motion recognition and fatigue detection is improved.

The invention provides a fruit and vegetable images classifying and identifying method and system based on model integration. The method comprises the steps that the fruit and vegetable images to be identified are respectively input in each nerve network model of a plurality of nerve network models after training, so that each nerve network model outputs the identification result of the fruit andvegetable images to be identified; integrating the identification result output by each nerve network model, and obtaining the final identification result of the fruit and vegetable images to be identified. The invention is advantageous in that the classifying features of a plurality of network models can be fully developed, and the same sample can undergo multiple classifications of different network models, and finally a classifying identification result after optimization grouping can be generated, which is always more reliable and more accurate than the classifying identification result generated by a single network model.

A multi-camera vision system and method of monitoring. In one embodiment imaging systems provide object classifications with cameras positioned to receive image data from a field of view to classify an object among multiple classifications. A control unit receives classification or position information of objects and (ii) displays an image corresponding to a classified object relative to the position of the structure. An embodiment of a related method monitors positions of an imaged object about a boundary by continually capturing at least first and second series of image frames, each series comprising different fields of view of a scene about the boundary, with some of the image frames in the first series covering a wide angle field of view and some of the image frames in the second series covering no more than a narrow angle field of view.

The invention discloses a user type determination method and a device. The method comprises steps that the behavior characteristic information of to-be-identified users of a live channel is extracted, and the behavior characteristic information is taken as input of at least one pre-determined classification engine to acquire an output value of each classification engine; through a preset algorithm, output values of all the classification engines are calculated to acquire characteristic values of the behavior characteristic information; when the characteristic values of the behavior characteristic information are in a first preset scope, the to-be-identified users are determined to be normal users; when the characteristic values are in a second preset scope, the to-be-identified users are determined to be non-normal users. Through the method, user type determination accuracy is improved through modes of mutually-independent operation of multiple classification engines and integrated result analysis.

A system and method for generating target area information. The system comprises a first processor effective to receive first information of a first classification level and a second processor effective to receive second information of a second classification level distinct from the first classification level. A cross domain processor is in communication with the first and second processors. The second processor is effective to receive a request from a requesting entity about a region of interest and interrogate a first sensor regarding the request. The second processor is further effective to receive first information from the first sensor and send the request through the cross domain processor to the first processor. The first processor is effective to interrogate a second sensor regarding the request and receive second information from the second sensor. The first processor is further effective to filter the second information based on the first and second classification levels to produce filtered information and send the filtered information through the cross domain processor to the second processor. The second processor is further effective to integrate the first information and the filtered information to produce integrated information and send the integrated information to the requesting entity.

The invention provides a signal classification method. The signal classification method includes: acquiring a plurality of data units from signals to be classified, extracting m feature parameters from each data unit to form a short-term feature vector, selecting a group of data units from the data units, subjecting K short-term feature vectors corresponding to the group of data units to multiple statistical treatments so as to obtain a long-term feature vector comprising a group of statistical vectors Ys, s ranging from 1 to S, subjecting a j-th element of the short-term feature vectors from X1 to Xp to s-th statistical treatment so as to obtain a j-th element of the vectors Ys, and obtaining n long-term feature vectors corresponding to n groups of data units; sending the n long-term feature vectors into n classifiers obtained by pre-training according to a preset principle so as to obtain multiple classification results; and obtaining a final classification result from the multiple classification results according to a preset decision fusion mechanism. The invention further provides a signal classification device.

An improved system and method for classifying tags of content using a hyperlinked corpus of classified web pages is provided. An anchor text index may be searched to find anchor texts that may match text of the tag, documents referenced by the matching anchor texts may be found, and the documents referenced by the matching anchor texts may be grouped to disambiguate multiple classifications that result from matching the anchor texts with the categories of the reference documents. To resolve ambiguity between multiple classifications, weighted classifications may be used where each document may be assigned a positive weight for a mapping to a category to indicate the confidence of the classification of the document to the category. The classification for the grouping of the documents referenced by the matching anchor texts with greatest frequency may be selected and output as the classification for the tag.

A method is disclosed for performing multiple classification of an image simultaneously using multiple classifiers, where information between the classifiers is shared explicitly and is achieved with a low-rank decomposition of the classifier weights. The method includes applying an input image to classifiers and, more particularly, multiplying the extracted input image features by |Σ| embedding matrices Ŵc to generate a latent representation of d-dimensions for each of the |Σ| characters. The embedding matrices are uncorrelated with a position of the extracted character. The step of applying the extracted character to the classifiers further includes projecting the latent representation with a decoding matrix shared by all the character embedding matrices to generate scores of every character in an alphabet at every position. At least one of the multiplying the extracted input image features and the projecting the latent representation with the decoding matrix are performed with a processor.

The invention discloses a cross-linked cable partial discharge mode recognition method based on a parameter optimization SVM (Support Vector Machine) algorithm. In order to avoid problems of a low neural network convergence rate, over learning and the like or low recognition precision due to improper SVM parameter section in traditional mode recognition, while an M-ary classification theory is introduced to extend the SVM algorithm with stronger generalization and learning abilities into multiple classifications of classifiers, an improved genetic algorithm is used for optimizing a penalty factor of each sub classifier and kernel function parameters, and thus the optimal parameter SVM classification model is built. Results show that each defect recognition rate is more than 95% when the optimal SVM serves as the classifier, and whether to optimize the parameters, the SVM general recognition ability is better than that of an RBF neural network.

The embodiment of the invention provides a crawler recognition and processing method and a related device. The method includes the steps that webpage access record data is read, and feature fields in the access record data are counted according to user identifiers to generate a detection dataset; the generated detection dataset and a pre-marked training dataset are input into multiple classifiers for classification learning, and multiple classification results corresponding to first detection data in the detection dataset in the multiple classifiers are obtained; according to the number of classification results belonging to crawlers in the multiple classification results and the total number of the classification results, the classification probability that the first user identifier corresponding to the first detection data belongs to crawlers is determined. According to the method and the device, crawlers are recognized through learning with a lot of statistical data, and the adverse effect of real-time recognition on system throughput is avoided; besides, when crawlers are updated, the crawlers can be effectively recognized by adjusting the feature fields, and the expandability of crawler recognition is improved.