124 results about "Data set" patented technology

Various embodiments of the present invention provide systems and methods for data processing. For example, a variable iteration data processing system is disclosed that includes a first detector, a second detector, a decoder and a unified memory buffer. An input data set is received by the first detector that performs a data detection and provides a first detected data set. The decoder receives a derivative of the first detected data set and performs a decoding operation that yields a decoded data set. In some cases, the derivative of the first detected data set is an interleaved version of the first detected data set. The decoded data set is written to a unified memory buffer. The first decoded data set is retrievable from the unified memory buffer and a derivative thereof is provided to the second detector. In some cases, the derivative of the decoded is a de-interleaved version of the decoded data set. The second detector is operable to perform a data detection on the derivative of the decoded data set and to provide a second detected data set that is written to the unified memory buffer.

The invention provides a deep question and answer method, which comprises the following steps of: 1, receiving question data; 2, carrying out a first retrieval in a knowledge base database, wherein the knowledge base database comprises factual information in encyclopedia; and 3, carrying out a third retrieval in a self-defining data set, wherein the third step comprises the steps of: 3.1, classifying the questions by a machine study method on the basis of a sentence pattern model set, and determining the sentence pattern mode of questions; and 3.2, obtaining second type keywords through the questions matched with the sentence pattern mode, and retrieving the self-defining data set through the second type keywords. Corresponding to the method, the invention also provides a medical retrieval system which comprises an input module, a first retrieval module and a third retrieval module, wherein the first retrieval module is used for retrieving the knowledge base database, the knowledge base database comprises the factual information in encyclopedia, and the third retrieval module is used for retrieving the self-defining data set.

The invention discloses an automatic classification method for electrocardiogram signals. The method is achieved according to the following steps of firstly, obtaining electrocardiogram signals of a human body, conducting filtering on the electrocardiogram signals, and detecting R waves of the electrocardiogram signals where filtering is conducted; secondly, establishing a data set after the R waves are detected, wherein the data set is composed of multiple sets of cardiac beat data, and each set of cardiac beat data has a label; thirdly, establishing a sparse automatic coding deep learning network; fourthly, training the sparse automatic coding deep learning network step by step; fifthly, inputting the to-be-measured cardiac beat data into the sparse automatic coding deep learning network according to the network weight, obtained in the fourth step, of the first hidden layer, the network weight, obtained in the fourth step, of the second hidden layer and the network weight, obtained in the fourth step, of the softmax classifier so as to obtain cardiac data which are output in a classified mode. The sparse automatic coding deep learning network is applied to the classification of the cardiac beat data, and by means of the autonomous leaning capacity and the deep characteristic excavation characteristic of the sparse automatic coding deep learning network, deeper characteristics of signals are extracted, and the cardiac beat data are classified.

A cloud computing system is described herein that enables fast processing of queries over massive amounts of stored data. The system is characterized by the ability to scan tens of billions of data items and to perform aggregate calculations like counts, sums, and averages in real-time (less than three seconds). Ad hoc queries are supported including grouping, sorting, and filtering without the need to predefine queries by providing highly efficient loading and processing of data items across an arbitrarily large number of processors. The system does not require any fixed schema, thus the system supports any type of data. Calculations made to satisfy a query may be distributed across a large number of processors to parallelize the work. In addition, an optimal blob size for storing multiple serialized data items is determined, and existing blobs that are too large or too small are proactively redistributed or coalesced to increase performance.

Disclosed is a method and data mining system for generating association rules from a data stream. An embodiment of the invention provides a method of generating association rules from a data stream, which is a non-limited data set composed of transactions continuously generated. The method includes: when itemsets included in the generated transactions and the counts of the itemsets are managed using a prefix tree and each node of the prefix tree has information on the count of a specific itemset corresponding to the node and a specific item, updating the information of a node corresponding to the itemset or adding a new node on the basis of the itemset included in the generated transaction and the count of the itemset; comparing the support of the itemset corresponding to each of the nodes of the prefix tree with a minimum support, which is a predetermined threshold value, to select frequent itemsets; and visiting all or some of the nodes corresponding to the selected frequent itemsets, and generating the association rule on the basis of the information of each of the visited nodes.

The invention discloses a bidirectional LSTM network-based vehicle behavior identification method and system; the method comprises the following steps: invoking offline acquired traffic video data; detecting and tracking vehicles in the traffic video data, and extracting vehicle driving tracks; extracting features from preprocessed vehicle driving tracks, and building a training data set and a test data set; using a bidirectional LSTM recurrent neural network to make model training for the training data set, and forming a vehicle behavior identification model; inputting the test data set intothe vehicle behavior identification model to make precision assignment; inputting the traffic video data acquired online in real time in the vehicle behavior identification model with the precision assignment, identifying vehicle behaviors, and outputting an identification result. The method uses the vehicle behavior identification model based on the bidirectional LSTM network to identify the vehicle driving tracks in the traffic video data, thus determining the vehicle driving behaviors with a high identification precision.

The invention discloses a remote sensing image scene classification method based on an improved residual network, which comprises the following steps of (a) collecting and/or downloading scene imagesto obtain a remote sensing scene classification data set; (b) performing enhancement processing on the images in the remote sensing scene classification data set, and randomly dividing the images in the remote sensing scene classification data set into a training set and a test set; (c) improving the residual network structure, training the improved residual network by using the images in the training set, continuously optimizing the residual network structure by training, and testing the accuracy of the residual network by using the images in the test set after the training is completed; and(d) carrying out scene classification on the remote sensing images by using the trained residual network structure. According to the method, the scene classification is carried out on the remote sensing image by utilizing the improved residual network, the network is more suitable for obtaining the remote sensing image data set with large difficulty and complex background textures by adjusting thestructure of the residual network, and the classification accuracy is improved.

An ultrasonic imaging method includes activating a transmit aperture within a multi-element transducer array, transmitting one or more ultrasonic beams along scan direction(s) that span the region of interest, for each transmit event, receiving ultrasound echoes from each element of a receive aperture, grouping the receive channel echo data into two or more sets corresponding to different receive sub-apertures, combining each sub-aperture data set to generate partially focused echo-location data for one or more reconstruction lines, and storing all the sub-aperture echo data sets during a storage period in a format that can be retrieved for later analysis. A method includes, during a post-storage period, retrieving stored sub-aperture data, combining the sub-aperture data to form one or more selected reconstruction lines, processing echo data to extract motion information from one or more sample positions along the selected reconstruction lines, and displaying an image representative of the processed motion information.

A system that facilitates semi-automatically producing an output relating to a multi-dimensional data object comprises an interface component that receives responses to one or more questions and produces commands that are a function of the responses. An execution engine executes the commands and manipulates the multi-dimensional data object in accordance with the responses. For example, the multi-dimensional object can relate to one or more of a data cube, a dimension, a data source, an attribute, a property, and a mining model.

The invention relates to a deep neural network target detection method integrated with a pixel-level attention mechanism. The deep neural network target detection method comprises the following steps:collecting a training sample image; converting data in the collected image data set into a format required by training the deep convolutional neural network through preprocessing, and performing dataenhancement to improve the training effect of the neural network; designing a deep convolutional neural network structure integrated with a pixel-level attention mechanism, the network structure being used for receiving an input image and outputting bounding box regression information and category information to an object in the image: the network structure comprising two parts, one part being abasic network for preliminarily extracting features; the other part being a plurality of cascaded convolutional layers or downsampling layers added at the tail end of the basic network and being usedfor extracting convolutional feature maps with different scales and fusing the multi-scale feature maps with a pixel-level attention mechanism; and training.

The invention discloses a multi-mode collaborative learning method and apparatus based on a neural network. The method comprises the following steps: S1 capturing short video data to establish a shortvideo data set, wherein the short video data comprise short videos published by a user historically and subsidiary information thereof; S2 extracting multi-mode features of the short video data in the short video data set, and performing information complement for the multi-mode features; S3 sequentially taking any modal feature in the modal features as a main mode and using the modal features asauxiliary modes, calculating the relationship weights of the main mode and the auxiliary modes, respectively classifying identical feature information and supplementary feature information, and performing information recombination on the supplementary feature information and the identical feature information enhanced by the neural network; and S4 establishing an attention mechanism network, distinguishing the degree of contribution of each piece of recombined mode feature information a scene category to which the short videos belong, and performing short video scene classification by using the neural network.

A method of detecting a target in a scene comprises the steps of generating a comparison of the value of data elements in first and second data sets, the data elements corresponding to returns from the same part of the scene and setting a detection threshold value for the comparison so as to detect targets entering said that part of scene The method provides improved target detection in the presence of clutter.

A first data set is transmitted to a client device across a plurality of wireless communication networks, each network of the plurality transmitting a corresponding portion of the first data set. A second data set is received from the client device.

A system and method for implementing a neural network based vehicle dynamics model are disclosed. A particular embodiment includes: training a machine learning system with a training dataset corresponding to a desired autonomous vehicle simulation environment; receiving vehicle control command data and vehicle status data, the vehicle control command data not including vehicle component types or characteristics of a specific vehicle; by use of the trained machine learning system, the vehicle control command data, and vehicle status data, generating simulated vehicle dynamics data including predicted vehicle acceleration data; providing the simulated vehicle dynamics data to an autonomous vehicle simulation system implementing the autonomous vehicle simulation environment; and using data produced by the autonomous vehicle simulation system to modify the vehicle status data for a subsequent iteration.

Four first electrodes are attached on the vicinity of a lower right end of a right clavicle, the vicinity of a lower left end of a left clavicle, the vicinity of a position on a right anterior axillary line at the level of a right lowermost rib, a the vicinity of a position on a left anterior axillary line at the level of a left lowermost rib of a living body, so as to correspond to limb leads of a standard 12-lead electrocardiogram (ECG). Two second electrodes are attached on such positions of the living body that correspond to a lead V2 and a lead V4 of chest leads of the standard 12-lead ECG. A first ECG data set corresponding to leads I and II of the standard 12-lead EGG with the first electrodes. A second ECG data set including the leads V2 and V4 with the second electrodes. An instantaneous electromotive force vector (a heart vector) is calculated based on the first and second ECG data sets, and predetermined first lead vectors of the leads I, II, V2 and V4. A third ECG data set including leads V1, V3, V5 and V6 of the chest leads is calculated based on the heart vector-and predetermined second lead vectors of the leads V1, V3, V5 and V6. A fourth EGG data set corresponding to leads III, aVR, aVL and aVF of the standard 12-lead EGG based on the first ECG data set The standard 12-lead EGG is derived based on the first to fourth ECG data sets.

The invention discloses an emotion recognition method and system based on a deep learning model and a long-short memory network, The method comprises the following steps: data preprocessing and data set partitioning of EEG signals are performed to construct a network model, wherein the network model comprises a picture reconstruction model composed of a variational encoder and an emotion recognition model composed of a long-short memory network; the network model comprises an image reconstruction model composed of a variational encoder and a short-long memory network; The objective function isconstructed according to the network model. The network model is trained by training set, and the objective function is optimized by Adam optimizer in neural network, and the trained network model isobtained. Using the cross-test set to cross-test the trained network model, determining the super-parameters of the network model, and obtaining the final network model; and using the final network model to visualize the seed data and perform emotion recognition. The invention relies on the data artificial intelligence method to learn the collected EEG signal space and time complex structure, reduce the subjective factors in the prediction, and improve the prediction accuracy.

A video encoder comprises a video frame receiver (101) connected to a processor (103) deriving relative frames from the received video frames and predicted frames. The processor is connected to a Discrete Fourier Transform (DCT) processor (105) which again is connected to a quantiser (107) for generating quantised spatial frequency coefficients for the relative frame. The output of the quantiser (107) is fed to a splitter that splits the data subset having low frequency components and a second data subset having frequency components. The first subset is used in the encoding loop comprising an inverse quantiser (111), inverse DCT processor (113), motion compensation processor (115, 117) and predicted frame processor (104). Hence, the encoding loop is simplified by only considering a reduced data set for each frame. A transmitter (119) transmits the video data as a progressively scalable stream for both the first and second data subsets.

A system has a reduced sensitivity to Single Event Upset and/or Single Event Transient(s) compared to traditional logic devices. In a particular embodiment, the system includes an input, a logic block, a bias stage, a state machine, and an output. The logic block is coupled to the input. The logic block is for implementing a logic function, receiving a data set via the input, and generating a result f by applying the data set to the logic function. The bias stage is coupled to the logic block. The bias stage is for receiving the result from the logic block and presenting it to the state machine. The state machine is coupled to the bias stage. The state machine is for receiving, via the bias stage, the result generated by the logic block. The state machine is configured to retain a state value for the system. The state value is typically based on the result generated by the logic block. The output is coupled to the state machine. The output is for providing the value stored by the state machine. Some embodiments of the invention produce dual rail outputs Q and Q′. The logic block typically contains combinational logic and is similar, in size and transistor configuration, to a conventional CMOS combinational logic design. However, only a very small portion of the circuits of these embodiments, is sensitive to Single Event Upset and/or Single Event Transients.

The invention relates to a drug use warning information processing method. The method is characterized by comprising the steps of 1, receiving drug use medical order information generated by a first terminal; 2, extracting drug use information and disease condition information from the drug use medical order information; 3, querying a reasonable drug use rule library by utilizing the drug use information and the disease condition information, and when the drug use information does not conform to the reasonable drug use rule library, executing the step 4; and 4, generating a drug use unreasonable rule data set according to the reasonable drug use rule library, wherein the drug use unreasonable rule data set contains a drug use unreasonableness value; and 5, generating drug use warning information, wherein the drug use unreasonableness value is marked in the drug use warning information. According to the drug use warning information processing method provided by the invention, the unreasonable drug use situation can be avoided, so that the drug use safety of patients is improved.

The invention discloses a network user behavior prediction system, and the system comprises a data collection and storage module, a data preprocessing module, a user network behavior analysis module and a data presentation module, wherein the data collection and storage module, the data preprocessing module, the user network behavior analysis module and the data presentation module are connected sequentially. The data collection and storage module is used for collecting and storing the useful mobile Internet data of a user through collection equipment. The data preprocessing module is used for carrying out the data clearing and cleaning of the useful data, filtering the data comprising noise and abnormality, forming an effective data set of user's behavior analysis, and enabling the effective data set to be transmitted to the user network behavior analysis module. The user network behavior analysis module is used for carrying out the arrangement and analysis of the effective data set, carrying out the analysis of the behaviors of the user, and outputting a user's behavior analysis result. The data presentation module is used for presenting the user's behavior analysis result to the user. The system supports the analysis and mining of a large amount of mobile Internet data of the user, and is good in prediction effect.

The invention relates to a fraud detection method and device for video data, computer equipment and a storage medium. The method comprises the following steps: acquiring to-be-detected video data; extracting image data of each video frame from the to-be-detected video data, and dividing the image data into a plurality of image data sets according to the time sequence of each video frame, the imagedata sets which comprises image data corresponding to continuous video frames; inputting each image data set into a pre-trained image feature extraction model to obtain an image feature vector; extracting voice data from the to-be-detected video data, and obtaining a voice feature vector of the voice data; performing cascade splicing on the image feature vector and the voice feature vector to obtain a multi-modal feature vector; and inputting the multi-modal feature vector into a pre-trained fraud detection model to obtain a fraud detection result corresponding to the to-be-detected video data output by the fraud detection model. By adopting the method, the characteristic information amount can be increased, the comprehensiveness and diversity of the characteristic information are improved, and the accuracy of video data fraud detection is effectively improved.

The invention relates to a text classification method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining to-be-classified text data, so as to obtain to-be-classified data; inputting the to-be-classified data into the target text classification model for classification to obtain a classification result; outputting the classification result to a terminal so as to display the classification result on the terminal; wherein the target text classification model is obtained by extracting a vector set from input text data, generating a label and then combining to form a training data set for training. According to the invention, the label is generated in a manner of automatically generating the label for the input text data; combining the generated label with an initial vector set; text data labels are corrected in an iterative mode, the initial text classification model is trained again, the training data quality is improved, the early-stage manual label labeling cost is reduced, and the requirement for a large amount of labeled data in a text classification task is quickly responded to, so that the text classification model is quickly established, and the text classification efficiency is improved.