2434 results about "Data classification" patented technology

A system and method for classifying tissue by application of discriminant analysis to spectral data. Spectra are recorded as amplitudes at a series of discrete wavelengths. Pluralities of reference spectra are recorded for specimens having known conditions. The reference spectra are subjected to discriminant analysis to determine wavelength regions of interest for the analysis. A plurality of amplitudes are selected for the analysis, and are plotted in an N-dimensional space. For each plurality of reference spectra corresponding to a specific known condition, a characteristic point is determined and plotted, the characteristic point representative of the known condition. A test spectrum is recorded from a test specimen, and the plurality of amplitudes corresponding in wavelength to the wavelength regions of interest are selected. A characteristic point in N-dimensional space is determined for the test spectrum. The distance of the characteristic point of the test spectrum from each of the plurality of characteristic points representative of known conditions is determined. The test specimen is assigned the condition corresponding to the characteristic point of a plurality of reference spectra, based on a distance relationship with at least two distances, provided that at least one distance is less than a pre-determined maximum distance. In some embodiments, the test specimen can comprise human cervical tissue, and the known conditions can include normal health, metaplasia, CIN I and CIN II / III.

A system and method for accessing actual human personal preference and opinion data, whose unique demographic parameters are defined by an individual Internet user. The system includes a computer linked to a specialized database that allows individual Internet users to supply and access human personal preference data according to trait and preference variable data input to the system. The data input to the system includes the personal traits and personal preferences of individuals, including feelings, opinions, issues and purchasing or commercial preferences. A data correlating program provides data sorting parameters that are defined by selected degrees of human trait and preference variables. Users, or those accessing the database for customized data results from the overall database, may, in turn, supply information to the database, or build a data profile, should they indicate that the data they seek pertains to individuals similar to themselves. With the present system and method, focus group data, or consumer preference sampling may be accomplished with immediacy and unrivaled relevance, as only users and members of selected relevance to the data solicitation might be contacted. Accordingly, discreet trait and preference specific groups of individuals that would be virtually impossible to locate and sample with typical focus group approaches are capable of being located and sampled with the present system and method.

The present invention relates to systems and methods that determine intent for received data (e.g., email, voice, graphics . . . ) and respond to the data based on the intent. The systems and methods employ various combinations of features based on shallow and deep linguistic analysis (e.g., semantic and syntactic) to yield very high accuracy. The systems and methods analyze and categorize received data to locate data that can include intent. This data can be further refined by extracting features related to the intent. The features can be utilized by a classifier to determine the intent. If the intent warrants a response, the data are further scrutinized and reformulated to generate a description that is indicative of the intent. The reformulation can include representing the features in a logical form, transforming the form and generating a description of the intent that can be presented to a user visually and / or audibly.

The system and method facilitate the control and management of information and actions related to the computerized scanning, categorization, processing, storage, and retrieval of paper documents and electronic document images for the purposes of creating electronic data; data entry; data capture; data classification and categorization; and document-, file-, and records-management. The systems and methods facilitate the conversion of paper documents into electronic images and into more useful forms.

The invention discloses a three-dimensional city modeling method based on BIM and GIS, comprising the following steps: data measurement, data processing, establishing BIM model, establishing three-dimensional GIS model, model fusion, municipal simulation, element separation and optimization, model rendering and display; through BIM, the precise height of the building can be easily obtained, appearance size and interior space information, thus through combining BIM and GIS, buildings are modeled first, and then the architectural spatial information is shared with the surrounding geographic environment, and applied to urban 3D GIS analysis, so that the cost of building spatial information can be reduced. At that same time, after the data is collected, point cloud and aerial photograph data are calculated and simulated by data classification comparison, so that the elevation information of the ground object is perfect, the elevation precision of the building model is high, and the roof ismore fine. The invention does not need a large amount of manual interaction modeling, and the data are collected by the unmanned aerial vehicle and the unmanned vehicle, and the efficiency is high, the production period is short, and the timeliness is strong.

A computer system and method for capture, managing and presenting data obtained from various often unrelated postings via the Internet for examination by a user. This system includes a scraping module having one or more scraping engines operable to scrape information data sets from listings on the corporate sites and web sites, direct feeds, and other sources, wherein the scraping module receives and stores the scraped listing information data sets in a database. The system also has a management platform coordinating all operation of and communication between the sources, system administrators and processing modules. The processing modules in the platform include scraping management module analyzing selected scraped data stored in the database, and a categorization module that examines and categorizes each data set stored in the database into one or more of a predetermined set of categories and returns categorized data sets to the database.

The invention relates to an infrared behavior identification method based on adaptive fusion of an artificial design feature and a depth learning feature. The method comprises: S1, improved dense track feature extraction is carried out on an original video by using an artificial design feature module; S2, feature coding is carried out on the extracted artificial design feature; S3, with a CNN feature module, optic flow information extraction is carried out on an original video image sequence by using a variation optic flow algorithm, thereby obtaining a corresponding optic flow image sequence; S4, CNN feature extraction is carried out on the optic flow sequence obtained at the S3 by using a convolutional neural network; and S5, a data set is divided into a training set and a testing set; and weight learning is carried out on the training set data by using a weight optimization network, weight fusion is carried out on probability outputs of a CNN feature classification network and an artificial design feature classification network by using the learned weight, an optimal weight is obtained based on a comparison identification result, and then the optimal weight is applied to testing set data classification. According to the method, a novel feature fusion way is provided; and reliability of behavior identification in an infrared video is improved. Therefore, the method has the great significance in a follow-up video analysis.

A system and method for use in data analysis are provided. The system comprises a data processing utility configured to receive and process input data, comprising: plurality of neural network modules capable for operating in a training mode and in a data processing mode in accordance with the training; a network training utility configured for operating the neural network modules in the training mode utilizing selected set of training data pieces for sequentially training of the neural network modules in a cascade order to reduce an error value with respect to the selected set of the training data pieces for each successive neural network module in the cascade; and an abnormality detection utility configured for sequentially operating said neural network modules for processing input data, and classifying said input data as abnormal upon identifying that all the neural network modules provide error values being above corresponding abnormality detection thresholds.

A system and method of monitoring and predicting tool overlay settings comprise generating current lot information, generating historical data, categorizing (binning) the historical data into discrete exposure field size ranges, and predicting current lot tool overlay settings based on the current lot information and historical data. The method monitors the overlay errors during each lot pass through each lithographic process operation. Moreover, the method uses a feedback sorting criteria to monitor the tool overlay settings. Furthermore, the current lot information comprises lithographic field dimensions, wherein the lithographic field optics distortion data is derived from the current lithographic process tool. Additionally, the historical data comprises same-bin lithographic field size dimensions of previous lots, which statistically means the data is derived from the same (or similar) bin of like lots, on the current lithographic process tool.

The invention discloses a desensitization system used for big data. The desensitization system comprises a database type management module, a system management module, a data source management module, a sensitive data discovery module, a desensitization task management module, a desensitization configuration management module, a multi-stage management module, a big data desensitization analysis module and a sensitive data classification module, wherein the database type management module manages the type of desensitization data; the system management module constructs a role for a desensitization system and carries out function authorization and management on the role; the data source management module provides data source management for the desensitization system; the sensitive data discovery module automatically discovers the sensitive data in the data source and submits the sensitive data to the user for regulation and confirmation; the desensitization task management module configures and manages the desensitization task of a big data desensitization system; the desensitization configuration management module manages and defines the sensitive data and carries out desensitization algorithm and desensitization strategy management; the multi-stage management module realizes the state monitoring, the strategy issuing and the data synchronization of multi-stage desensitization equipment; the big data desensitization analysis module carries out real-time analysis on collected desensitization log data to guarantee the security of the desensitization system; and the sensitive data classification module carries out characteristic analysis on the sensitive data to classify the sensitive data. The invention also discloses a desensitization method used for the big data.

The invention relates to the technical field of computers, in particular to a computer server cluster log monitoring method and monitoring platform. The monitoring method includes step A of monitoringmassive machine code commands generated by a main board during the running of each server in a computer server cluster, and using a real-time stream data collection frame in a big data technology tocollect the machine code commands generated by the main board in real time; step B of performing data classification and transfer operations on the collected data, including real-time filtering, processing and storage; step C of performing real-time analysis and display of the data processed in step B to obtain real-time analysis results and store the results; step D of performing analysis by combination with the real-time analysis results and stored historical analysis results, obtaining potential abnormal results of the computer server, and performing early warning and notification. The computer server cluster log monitoring method and monitoring platform have the advantages of real time, expandability, automation, high availability and rich API extension.

A method of managing a storage area of a memory device in a memory system is provided. A first data is received. The first data has a logical address to be written to the memory device having a plurality of memory blocks. The first data is classified into one of a hot data and a cold data based on an update frequency of the first data. A memory block is defined into a first storage area and a second storage area based on an amount of charge loss of a memory cell in the memory block. A memory cell of the first storage area has charge loss greater than a memory cell of the second storage area. The logical address of the first data is converted to a physical address of the memory device according to a result of the classifying of the first data. The first data is written to a memory cell having the physical address of the memory device.

System and method for optimization of a design associated with a response function, using a hybrid neural net and support vector machine (NN / SVM) analysis to minimize or maximize an objective function, optionally subject to one or more constraints. As a first example, the NN / SVM analysis is applied iteratively to design of an aerodynamic component, such as an airfoil shape, where the objective function measures deviation from a target pressure distribution on the perimeter of the aerodynamic component. As a second example, the NN / SVM analysis is applied to data classification of a sequence of data points in a multidimensional space. The NN / SVM analysis is also applied to data regression.

Disclosed is a system and method for the analysis of event data that enables analysts to create user specified datasets in a dynamic fashion. Performance, equipment and system safety, reliability, and significant event analysis utilizes failure or performance data that are composed in part of time-based records. These data identify the temporal occurrence of performance changes that may necessitate scheduled or unscheduled intervention like maintenance events, trades, purchases, or other actions to take advantage of mitigate or compensate for the observed changes. The criteria used to prompt a failure or performance record can range from complete loss of function to subtle changes in performance parameters that are known to be precursors of more severe events. These specific criteria applied to any explicit specific application and this invention is relevant to this type of data taxonomy and can be applied across all areas in which event data may be collected.

The invention discloses a method for determining the abnormal transaction process of an electronic commodity. The method comprises the steps of: for each electronic commodity, extracting the historical transaction data of the commodity within specified time duration in the past from the historical transaction data recorded in a commodity transaction database; according to each transaction feature which is predetermined to be extracted, extracting a corresponding transaction feature value from the extracted historical transaction data of the commodity within specified time duration in the past; according to a data classification principle corresponding to the corresponding conditions which are satisfied by the extracted transaction feature values simultaneously, taking the volume of transactions of the commodity predicted by one type of historical transaction data corresponding to the data classification principle as the predicted volume of transactions of the commodity within the specified time duration in the past; extracting the actual volume of transactions of the commodity within the specified time duration in the past from the historical transaction data recorded in the commodity transaction database; and determining whether the transactions of the commodity within the specified time duration in the past are abnormal according to the predicted volume of transactions and the actual volume of transactions of the commodity within the specified time duration in the past.

The invention discloses a public bicycle internet of things system, comprising a perception layer, a centralized control layer, a data processing layer, a system service layer and a system management layer. The perception layer is used for acquiring public bicycle equipment and citizen bicycle lease IC card information; the centralized control layer finishes the network building, the information receiving and transmitting of the perception layer, and the data conversion processing of the information; the data processing layer is used for carrying out creating, management, data classification and processing on a system database; the system service layer is used for processing all application operations related to users of a public bicycle system; and the system management layer is used for processing all operations related to management personnel and system maintenance personnel of the public bicycle system. According to the public bicycle internet of things system, the system scale can be configured dynamically, a distributed processing strategy is adopted for system data processing, a bicycle parking locking column and a bicycle branch management box have independent offline processing capability, so that the communication pressure on a system background is greatly reduced, and the system reliability, extensibility, applicability and fault-tolerant ability are greatly improved.

A system and method for using bi-directional conversation data to improve signal presence detection are disclosed. The detector module is adapted to communicate with a signal enhancement module. The detector module collects data from a transmit direction of the connection and a receive direction of a data connection. The collected data from the transmit and the receive direction is used to classify at least one of data in the transmit direction and data in the receive direction. Responsive to the classification, the signal enhancement module enhances data in one of the transmit direction and the receive direction. Hence, data classification accuracy is improved by using data from both the transmit and receive directions. In one embodiment, the detector module applies a voice activity detection module (VAD) process to detect the presence or absence of voice data in the collected data.

The present invention discloses a multiple-UAV operation information display system which comprises a data receiving unit for receiving flying state data and image data from airborne sensors and classifies and stores into a storage unit corresponding to each UAV, then a processing unit calculates and generates an integrated information display set according to the flying state information, terrain database information of each UAV in the storage unit and the control options of a control unit, and a display unit combines a display parameter and displays the integrated information display set in the whole. According to the system, the relative positions of multiple UAV can be displayed in an integrated way, the relative relationship of UAVs flying at a low altitude and terrain and obstacles on the ground is displayed in a highlighted way, the state sensing ability of control personnel is effectively enhanced, and the system is especially for a multiple-unmanned-helicopter cluster application.

Methods and systems for learning, recognition, classification and analysis of real-world cyclic patterns using a model having n oscillators, with primary frequency ω1, ω2, . . . , ωn. The state of the oscillators is evolved over time using sensor observations, which are also used to determine the sensor characteristics, or the sensor observation functions. Once trained, a set of activity detection filters may be used to classify sensor data as being associated with an activity.

The invention discloses an advanced persistent threat (APT) attack detection method based on deep belief network-support vector data description. A deep belief network (DBN) is used for feature dimension-reduction and excellent feature vector extraction; and support vector data description (SVDD) is used for the data classification and detection. At a DBN training state, the feature dimension-reduction is performed by using the DBN model after obtaining a standard data set; a low-level restricted Boltzmann machine (RBM) receives simple representation transmitted from the low-level RBM by usingthe high-level RBM so as to learn more abstract and complex representation after performing the initial dimension-reduction, and back propagation of a back propagation (BP) neural network is used forrepeatedly adjusting a weight value until the data with excellent feature is extracted. The data processed by the DBN is divided into a training set and a testing set, and the data set is provided for the SVDD to perform training and identification detection, thereby obtaining the detection result. The attack detection method disclosed by the invention is suitable for the unsupervised attack datadetection with large data size and high-dimension feature, is fit for the APT attack detection and can obtain an excellent detection result.

A method of classifying a digital document may include: identifying, by an automated data processor, a request for access to the digital document for a first user; determining user identifying information for the first user; obtaining, by the automated data processor, according to the user identifying information a first user characteristic including organizational affiliation of the first user or a job function of the first user; generating, by the automated data processor, based on the first user characteristic, a digital document classification for the digital document; associating the digital document classification with the digital document, by embedding the document classification in the digital document or logging the document classification in a log identifying the digital document. A user access determination for the digital document may be made according to the associated digital document classification.

Methods, systems and deep learning engines for content and context aware data classification by business category and confidentiality level are provided. The deep learning engine includes a feature extraction module and a classification and labelling module. The feature extraction module extracts both context features and document features from documents and the classification and labelling module is configured for content and context aware data classification of the documents by business category and confidentiality level using neural networks.

The invention relates to a course field multi-modal document classification method based on a cross-modal attention convolutional neural network. Multi-modal document data in a course field is preprocessed; an attention mechanism and a dense convolutional network are combined, a convolutional neural network based on cross-modal attention is provided, and image features with sparsity can be constructed more effectively; a text feature-oriented attention mechanism-based bidirectional long-short-term memory network is provided, so that text features locally associated with image semantics can beefficiently constructed; and cross-modal grouping fusion based on an attention mechanism is designed, so that the local association relationship between the images and the texts in the document can belearned more accurately, and the accuracy of cross-modal feature fusion is improved. Under the data set of the same course field, compared with an existing multi-modal document classification model,the method has better performance, and the accuracy of multi-modal document data classification is improved.

The invention relates to a real-time land cluster survey method. The method comprises the following steps: with an office data service platform as a center, firstly sending the filed data acquired by virtue of GPS / PDA (global positioning system / personal digital assistant) and GPS (global positioning system) data used for monitoring and positioning in a field data acquisition subsystem to a real-time communication service system in a standard data definition format by virtue of the TCP / IP (transmission control protocol / internet protocol) network protocol; classifying the obtained data by the office data server; and issuing spatial data and attribute data in the whole database by a service terminal, and browsing and analyzing to generate a report for use of a user, thus the real-time cluster survey monitoring on the land is realized finally.

The invention discloses a polarization SAR data classification method and system based on a mixed classifier. The polarization SAR data classification method includes the steps of obtaining polarization SAR data initial polarization characteristics in different classes, adopting a decision tree classifier to select the polarization characteristics for classification from the initial polarization characteristics, being based on the polarization characteristics for classification, and adopting an SVM classifier to classify polarization SAR data. According to the polarization SAR data classification method and system, the advantages of the decision tree classifier and the advantages of the SVM classifier are synthesized, the classification accuracy reaches the level of the SVM classifier, the classification calculation efficiency is equivalent to that of the decision tree classifier, and the polarization SAR data classification method and system are of great significances in polarization SAR data classification.

The invention relates to a TSN and non-TSN interconnection oriented industrial heterogeneous network scheduling method, which belongs to the field of industrial Ethernet and comprises the following steps that: in a network deployment stage, a user performs configuration management on a TSSDN controller according to application requirements through a terminal application program; the TSSDN controller performs data classification according to a delay demand; the TSSDN controller carries out scheduling priority management on the data; the TSSDN controller calculates the shortest path and schedules the priorities; the SDN switch performs flow table matching,and judges the congestion condition of the queue; and the feedback adjustment to queue congestion is carried out. According to the method,high-efficiency and high-quality scheduling is carried out in the industrial heterogeneous network with the TSN and the non-TSN interconnected, and low-delay and deterministic transmission of different types of data in the network is guaranteed.