165 results about "Real time classification" patented technology

A method of inspecting Web elements for real-time classification and detection of Web elements spoofing attempts, according to which trustworthy Web locations are identified for generating a database of safe zones. For each inspected element, it is checked whether or not its top frame URL is included in the database, and if it is included, the element is classified as suspected in Web elements location spoofing attempt.

A new method, system and apparatus is provided that enables muscle signals that correspond to muscle contractions to be mapped to one or more functions of an electronic device such as a prosthetic device or gaming apparatus. Muscle signals are classified in real-time from self-selected intentional movements. A self-training protocol allows users to select and label their own muscle contractions, and is operable to automatically determine the discernible and repeatable muscle signals generated by the user. A visual display means is used to provide visual feedback to users illustrating the responsiveness of the system to muscle signals generated by the user.

A classification system has a classification server that receives data for an email and determines if the email message is suspicious, legitimate but failing authentication, forwarded or fully authenticated and legitimate when the domains are owned, or not owned, by the domain owner, Email messages are categorized and presented in a report that enables the email sender to identify and fix a network, malicious traffic, and legitimate messages that have failed authentication beyond control. It also highlights where everything is going well.

The present invention discloses a sudden event emergency information mining method based on social media. The method comprises the steps of (S1) using an open platform API or a web crawler to acquire social media data, the social media data being a document set; (S2) using a MongoDB cluster to store the document set; (S3) preprocessing the document set; (S4) using an LDA to mark the preprocessed document set, and obtaining a known sample; (S5) forming a word characteristic set by all words in each document of the known sample, word frequency of each word characteristic in the document being the weight of the word characteristic in the document; (S6) constructing a short text real-time classification model; (S7) classifying real-time sudden events by using the short text classification model, and predicting themes of the sudden events; and (S8) performing information mining according to the social media data of the classified sudden events. Classification of social media short texts can be automatically and rapidly achieved, and therefore sudden event emergency information is mined.

A method is provided to classify network traffic flows in real-time using spectral analysis techniques to extract regularities inside the network traffic flows. In one embodiment of the invention, subspace decomposition on power spectral density feature vectors and minimum coding length criterion are utilized for training traffic flows of different classifications. Experimental results are shown to demonstrate the effectiveness and robustness of the invention.

Methods and computer program products are disclosed for the real-time classification of text from a region of interest within an image sample. A feature extractor extracts feature data associated with a plurality of region features from the region of interest. The plurality of region features are selected as to minimize the time necessary for feature extraction. A neural network preclassifier selects one of a plurality of associated source classes for the region of interest according to the extracted feature data. A plurality of classification systems are each associated with one of the plurality of source classes. Each of the plurality of classification systems are operative to classify individual characters within the region of interest when the associated source class of the classification system is selected.

The invention discloses a water chilling unit fault feature learning and classification method based on 1DCNN and GRU fusion, and solves the problems that an existing method consumes time and is difficult to process mass data. The method comprises the following steps: firstly, preprocessing collected continuous-time fault sequence information under different working conditions to construct a sample set; secondly, carrying out preliminary feature extraction on the sample set by adopting a dimension reduction algorithm; then, creating a 1DCNN-GRU neural network model; during training, enabling fault features of a training sample at the same moment to act as input of a time step of the network model, adopting a cross entropy loss function, model parameters are adjusted in combination with a back propagation BP algorithm, enabling Softmax regression to act as a classification algorithm, conducting iterative updating on the network model parameters, and finishing model training. The methodnot only can extract the dynamic information between the local features of the sequences and the sequences, but also can realize real-time classification diagnosis of faults.

A device and method are provided for real-time classification of covert speech. The device comprises a plurality of sensors for capturing real-time bio-signal data for brain monitoring in response to mental tasks delivered to a user, and a brain computer interface with memory storing instructions to configure a processor to perform a method of real-time classification of covert speech. The method comprises capturing real-time bio-signal data for brain monitoring in response to mental tasks delivered to a user, pre-processing the raw bio-signal data, extracting a vector of features from the raw bio-signal data, selecting features from the vector of features, building classification model to generate classified covert speech data using the selected features, and controlling a display device with visual elements based on the classified covert speech data.

The invention discloses an information entropy-based self-adaptive integrated classification method of data streams. Concept drift can be detected, and duplicate concepts can also be identified. In asystem, a new classifier is reconstructed and put into a classifier pool only when existence of a new concept is detected, the problem of duplicate training caused by duplicate concept appearance is prevented, model updating frequency is reduced, and real-time classification ability and classification effect of a model are improved. Through carrying out performance analysis comparison with classical data stream algorithms on a synthetic dataset and a real dataset, experiments show that the method of the invention can cope with multiple types of concept drift, improves anti-noise ability of theclassification model, and also has lower time cost consumption on the premise of ensuring higher classification accuracy. The method of the invention can be applied to many practical problems of sensor network anomaly detection, credit-card fraud behavior detection, weather forecasting, electricity price prediction and the like.

The invention discloses a digital drilling real-time classification method for a surrounding rock of underground engineering. The method comprises the following steps: making a field drilling test onan underground engineering field to obtain a drilling parameter, obtaining uniaxial compressive strength of a rock body through a model of relationships between drilling parameters and uniaxial compressive strength of rock bodies, obtaining a rock body breaking coefficient on the basis of a model of relationships between drilling parameters and rock body breaking coefficients, obtaining equivalentuniaxial compressive strength from the uniaxial compressive strength of the rock body and the rock body breaking coefficient by use of a uniaxial compressive strength correction formula, obtaining displacement of the surrounding rock by use of a model of relationships between equivalent uniaxial compressive strength and displacement of surrounding rocks, and further classifying the surrounding rock in real time according to the displacement of the surrounding rock. The method has the characteristic of real-time accurate prediction; the displacement of the surrounding rock is reversely deducedaccording to the drilling parameter obtained in field, thereby accurately classifying the surrounding rock in real time by use of the stability of the surrounding rock and providing a basis for design and construction of underground engineering; disturbance in a process of delivering a rock sample to a laboratory is avoided, meanwhile, the test time is shortened to a great extent and manpower andmaterial resources are saved.

The invention relates to an RNNs-based method for automatic safety checking of a short message. The method specifically comprises the steps that (1) historical data is pre-processed, wherein preprocessing comprises noise removal and Chinese word segmentation; (2) characteristics of the historical data which is already pre-processed at the step (1) are extracted, so word vectors can be generated; and (3) a classification model which integrates RNNs and Naive Bayes is used to conduct real-time classification of short message texts. According to the invention, workloads of police officers in manual checking are reduced.

A method including identifying an emotion identifier from a message using an emotion identifier table is provided. The method includes separating emotion identifier from the message, splitting a portion into multiple tokens delimited by a space or a punctuation mark, and determining a score for the message based on an emotion score in the emotion identifier table, and on at least one of the multiple tokens. The method includes adding a token from the multiple tokens to the emotion identifier table based on a contribution from the token to the message score, associating a sentiment score for the token in the emotion identifier table based on the contribution of the token to the message score, and modifying an emotion score of the token when the token is already included in the emotion identifier table based on the contribution of the token to the message sentiment score.

The invention discloses an on-site tunnel real-time surrounding rock classification system and a method. The system includes a bearing platform, the bearing platform bears a measuring ruler, a point load tester for conducting point load test on an on-site surrounding rock sample, and a geological compass for determining weak structural plane occurrence, the measuring ruler, the point load tester and the geological compass are detachably disposed on the bearing platform, and the lower part of the bearing platform is provided with moving components for driving the bearing platform to move. The setting of the whole system is convenient for moving a measurement component to the site for real-time measurement, realizes real-time classification work of tunnel surrounding rock, effectively improves the efficiency of tunnel surrounding rock classification work, improves the flexibility of classification work, and can acquire real-time feedback of rock mass classification near a tunnel face, thus being convenient for directly carrying out follow-up work.

The invention discloses garbage classification equipment and a garbage classification method based on chemical element properties as well as application. The garbage classification method comprises the following steps: carrying out pre-treatment including bag breaking, crushing, sieving and the like on garbage; putting a garbage mixture which is pre-treated on a garbage conveying device; carrying out detection analysis on garbage by using equipment for detecting the garbage on line and in real time based on the chemical element properties; receiving a signal detected by the detection equipment by a computer control system, and sending an instruction to a device for classifying the garbage in real time; and classifying the garbage. The garbage classification method is applied to garbage recycling and preparation of raw materials with the stable chemical element properties; the garbage can be recycled and resources are saved; the garbage also can be prepared into the raw materials with the stable chemical element properties; and when the raw materials are used for processes including pyrolysis, gasification, combustion and the like, and good conditions are created for stable and effective operation of the processes, so that the reduction of the garbage is realized, and the aims of recycling and energy generation are realized.

The invention discloses a brain-computer interactive image retrival system based on real-time functional magnetic resonance imaging (fMRI), so as to solve the problem of, needing to improve a brain-computer interactive image retrieval system in the prior art. The brain-computer interactive image retrieval system is used by comprising the following steps that (1) an image semantic real-time classification model, an image feature similarity real-time evaluation model and a visual attention real-time decoding model are trained by collecting fMRI data in advance; (2) a to-be-retrieved image is presented to a subject as visual stimulation, the fMRI data of the subject are collected at the same time, and the semantic class of the to-be-retrieved image is distinguished by using the image semantic classification model; (3) an image which represents the distinguished result of the semantic class and the to-be-retrieved image are presented to the subject together, and whether the semantic class which is distinguished by the image semantic classification model is correct is fed back by the subject through visual attention; (4) a retrieved image is given according to the distinguished result of the semantic class, which is fed back by the subject. The brain-computer interactive image retrieval system is significant in increasing the fMRI image retrieval accuracy.