1036 results about "Unsupervised learning" patented technology

A method and system that dynamically ranks electronic messages based on their situational and inherent dimensions, which are judged by a set of filters. These filters evaluate the different elemental metadata constituting a message and produce a priority value based on filters relevance and importance. The system iterates through queued messages, examine the structured content for expected attributes, statistically analyze unstructured content, apply dynamically weighted rules and policies to deliver a priority ranking, and then display the message and its vital attributes in accordance with the priority ranking. The system also adaptive learns and adjusts its weighted rules and policies to permit priority ranking to change on real-time or interval-based (may be user-defined) schedule. The system includes a GUI for increasing reading and processing efficiency. The GUI performs supervised and unsupervised learning from the user's behaviors, and displays messages in accordance with their priority classification.

A system is disclosed for selectively applying a paraphrase alternation pattern to a textual input. The system includes a pattern identification component for processing a collection of data and identifying a plurality of potentially applicable paraphrase alternation patterns. A context model provides an objective frame of reference in which to compare one or more of the plurality of potentially applicable paraphrase alternation patterns to the textual input to determine whether the pattern can be applied without changing meaning. A language model provides a principled basis for determining the boundaries of the text segment to be modified.

A method of training a natural language call routing system using an unsupervised trainer is provided. The unsupervised trainer is adapted to tune performance of the call routing system on the basis of feedback and new topic information. The method of training comprises: storing audio data from an incoming call as well as associated unique identifier information for the incoming call; applying a highly accurate speech recognizer to the audio data from the waveform database to produce a text transcription of the stored audio for the call; forwarding outputs of the second speech recognizer to a training database, the training database being adapted to store text transcripts from the second recognizer with respective unique call identifiers as well as topic data; for a call routed by the call router to an agent: entering a call topic determined by the agent into a form; and supplying the call topic information from the form to the training database together with the associated unique call identifier; and for a call routed to automated fulfillment: querying the caller regarding the true topic of the call; and adding this topic information, together with the associated unique call identifier, to the training database; and performing topic identification model training and statistical grammar model training on the basis of the topic information and transcription information stored in the training database.

A system and method is disclosed for recommending items to individual users using a combination of clustering decision trees and frequency-based term mapping. The system and method of the present invention is configured to receive data based on user action, such as television remote control activity, or computer keyboard entry, and when a new item data is made available from sources such as television program guides, movie databases, deliverers of advertising data, on-line auction web sites, and electronic mail servers, the system and method analytically breaks down the new item data, compares it to ascertained attributes of item data that a user liked in the past, and produces numeric ranking of the new item data dynamically, and without subsequent user input, or data manipulation by item data deliverers, and is tailored to each individual user. A embodiment is disclosed for learning user interests based on user actions and then applying the learned knowledge to rank, recommend, and/or filter items, such as e-mail spam, based on the level of interest to a user. The embodiment may be used for automated personalized information learning, recommendation, and/or filtering systems in applications such as television programming, web-based auctions, targeted advertising, and electronic mail filtering. The embodiment may be structured to generate item descriptions, learn items of interest, learn terms that effectively describe the items, cluster similar items in a compact data structure, and then use the structure to rank new offerings. Embodiments of the present invention include, by way of non-limiting example: allowing the assignment of rank scores to candidate items so one can be recommended over another, building decision trees incrementally using unsupervised learning to cluster examples into categories automatically, consistency with “edge” (thick client) computing whereby certain data structures and most of the processing are localized to the set-top box or local PC, the ability to learn content attributes automatically on-the-fly, and the ability to store user preferences in opaque local data structures and are not easily traceable to individual users.

The invention provides an unsupervised domain-adaptive brain tumor semantic segmentation method based on deep adversarial learning. The method comprises the steps of deep coding-decoding full-convolution network segmentation system model setup, domain discriminator network model setup, segmentation system pre-training and parameter optimization, adversarial training and target domain feature extractor parameter optimization and target domain MRI brain tumor automatic semantic segmentation. According to the method, high-level semantic features and low-level detailed features are utilized to jointly predict pixel tags by the adoption of a deep coding-decoding full-convolution network modeling segmentation system, a domain discriminator network is adopted to guide a segmentation model to learn domain-invariable features and a strong generalization segmentation function through adversarial learning, a data distribution difference between a source domain and a target domain is minimized indirectly, and a learned segmentation system has the same segmentation precision in the target domain as in the source domain. Therefore, the cross-domain generalization performance of the MRI brain tumor full-automatic semantic segmentation method is improved, and unsupervised cross-domain adaptive MRI brain tumor precise segmentation is realized.

The technology disclosed relates to machine learning based anomaly detection. In particular, it relates to constructing activity models on per-tenant and per-user basis using an online streaming machine learner that transforms an unsupervised learning problem into a supervised learning problem by fixing a target label and learning a regressor without a constant or intercept. Further, it relates to detecting anomalies in near real-time streams of security-related events of one or more tenants by transforming the events in categorized features and requiring a loss function analyzer to correlate, essentially through an origin, the categorized features with a target feature artificially labeled as a constant. It further includes determining an anomaly score for a production event based on calculated likelihood coefficients of categorized feature-value pairs and a prevalencist probability value of the production event comprising the coded features-value pairs.

A method and apparatus for creating templates for electronic documents is provided. One or more attributes are extracted, using a seed template, from a first document, such as a web page. A second document that contains a particular attribute, extracted from the first document, is identified. The second document may be in a different cluster than the first document. The second document is annotated, using an extracted attribute, to create an annotated document. The second document is annotated without human intervention. A new template for the annotated document is generated. The new template facilitates extraction of information from the annotated document. The new template may be used to extract additional attributes from all documents in the cluster of documents of which the second document is a member. The process may continue over numerous iterations to generate a large number of templates in an automated fashion.

The invention discloses a semi-supervised image classification method based on a generative adversarial network, which mainly solves the problems that the existing unsupervised learning classificationprecision is low and semi-supervised learning needs a large number of accurate labels, and comprises the following implementation steps of: 1) selecting and downloading a standard image training sample and a test sample; 2) setting relevant parameters of network supervised learning, and establishing a generative adversarial network consisting of a generator network, a discriminator network and anauxiliary classifier in parallel; 3) training the generative adversarial network by using a random gradient descent method; and 4) inputting the test sample to be classified into the trained generative adversarial network model, and outputting the category of the image to be detected. The method improves the image classification precision of unsupervised learning, can obtain a very good image classification effect on a sample set only containing a small amount of accurate annotation samples, and can be used for target classification in an actual scene.

The invention discloses a track predication method based on a Gauss mixture time series model (GMTSM for short) to carry out model regression and analysis of road vehicle flow on a large quantity of vehicle historical tracks to realize vehicle track predication. The method mainly comprises the following steps: (1) carrying out unsupervised learning cluster on vehicle historical tracks through a k-means algorithm; (2) constructing a historical track probability distribution model through a Gauss mixture time series model; and (3) predicating tracks of mobile objects through a regressive processof the mixture model. An experiment result shows that the Gauss mixture time series model automatically adjusts the weight of sub-models and selects a predication track with the largest probability in a condition that road vehicle flows change suddenly.

Monitoring system and method for use with diabetic treatment management. The system includes: a communication interface configured to permit access to stored raw log data, obtained over a certain time, being indicative of glucose measurements, meals consumed and insulin delivery; and a control unit including an unsupervised learning controller configured to receive and process said raw log data and determine at least one global insulin pump setting of basal rate, correction factor, carbohydrate ratio and insulin activity curve parameters. The system may include a processing unit including a first processor for processing measured data indicative of blood glucose level and generating first processed data, a second processor including at least one fuzzy logic module which receives input parameters corresponding to the measured data, the first processed data and a reference data, and processes the data to produce a qualitative output parameter to determine whether any treatment parameter should be modified.

The invention discloses a pedestrian re-identification method and device based on unsupervised learning and a medium, and the method comprises the steps: obtaining a target image and a comparison image, and identifying whether a pedestrian exists in the target image in the comparison image through a pedestrian re-identification model based on unsupervised learning; outputting a recognition result; establishing a pedestrian re-identification model: carrying out initial training on the visual classifier according to the labeled source data set to obtain a visual classifier; learning the label-free target data set by using the vision classifier after initial training to obtain a matching probability and space-time information; obtaining a Bayesian fusion model according to the matching probability and the space-time information; carrying out similarity matching on pedestrian images in the unlabeled target data set by the Bayesian fusion model according to the comparison target pedestrian images to obtain a similarity score; sorting the similarity scores according to a preset threshold value to obtain a sorting result; when it is detected that the current model training optimization frequency is smaller than or equal to a preset optimization threshold value, performing parameter updating on the visual classifier.

The invention discloses a face unsupervised feature learning method and device based on a generative adversarial network. Through the preprocessing of a collected original face image, the image is converted into a face training image with a set size. The converted face training image is taken as training data to train a target generation network in a constructed depth convolution generative adversarial network. A generated random vector set is inputted into the trained target generation network, and a generative image set corresponding to the random vector set is obtained. The obtained generative image set is inputted into a depth regression network of a constructed depth convolutional neural network, the depth regression network is trained, and a face feature vector of the generative image set is extracted. The invention provides the face unsupervised feature learning method and device based on a generative adversarial network, a mode with the combination of DCGAN and DCNN is used to carry out unsupervised learning, the depth regression network is used to learn a reverse target generation network, the learning effect is good, and the identification accuracy is high.

The invention discloses a firework identification method and a firework identification system based on deep learning of an image. The firework identification method comprises the following steps of step 1, acquiring a label-free sample image set and a label sample image set; step 2, obtaining a label-free training data set and a label training data set; step 3, performing whitening preliminary processing on training data; step 4, based on the label-free training data subjected to the whitening preliminary processing, constructing a deep neutral network based on sparse self coding by adopting unsupervised learning, and extracting a basic image feature set of the label-free training data; step 5, convolving basic image features and pooling image data; step 6, training a Softmax classifier based on the convolved and pooled label training data set; step 7, inputting the convolved and pooled images to be identified into the trained Softmax classifier to obtain the identification result. According to the firework identification method and the firework identification system disclosed by the invention, the visual identification rate of fireworks and a similar object can be effectively improved, and automatic identification with higher precision for the fireworks can be realized.

Hybrid packet traffic flow profiling technology inspects packet headers to classify packet traffic flows using clustering models developed using unsupervised learning based on known packet traffic flows and one or more traffic classification models developed using supervised learning based on the known packet traffic flows.

The invention discloses an image super-resolution method based on SAE and sparse representation, and belongs to the field of image processing. The image super-resolution method mainly comprises an off-line training stage and a test refactoring stage, wherein in the off-line training stage, image characteristics extracted by an SAE (Sparse Auto Encoding) model are subjected to dictionary training, and a dictionary pair reflecting corresponding relations of high-resolution images and low-resolution images is established; in the test refactoring stage, low-resolution images inputted by a user are subjected to super-resolution reconstruction by the obtained dictionaries and a sparse representation method. Through the application of the image super-resolution method, unsupervised learning training is performed on original image sampling data by using the SAE model, so that the defects that manually designed operator characteristic extraction is time-consuming and strenuous and the extracted characteristics are single are avoided, meanwhile, image characteristics represented by SAE compression are directly used for training of the high-low-resolution dictionary pair, the dictionary training is facilitated, lost detail components in the images can be estimated by the sparse representation method, and higher-quality high-resolution images can be restored from the low-resolution images conveniently.

The invention belongs to the technical field of the natural language processing of computers, in particular to a construction and utilization method for a context-aware dynamic word or character vector on the basis of deep learning. The dynamic construction method for the context-aware dynamic word or character vector on the basis of the deep learning comprises the following steps of: in massive texts, through an unsupervised learning method, simultaneously learning a global feature vector of a word or character and the feature vector representation of the global feature vector when a specific context appears, and combining the global feature vector with the context feature vector, and dynamically generating word or character vector representation. By use of the method, the word or character vector dynamically constructed on the basis of the context can be applied to a natural language processing system. The method is mainly used for solving a problem that the word or character vector expresses different meanings in different contexts, i.e. the problem that one word or one character has multiple meanings can be solved. The dynamic word or character vector can be used for obviously improving the performance of various natural language processing tasks of different languages, wherein the tasks comprise Chinese word segmentation, part-of-speech tagging, naming recognition, grammatical analysis, semantic role tagging, sentiment analysis, text classification, machine translation and the like.

Techniques are described for analyzing a stream of video frames to identify temporal anomalies. A video surveillance system configured to identify when agents depicted in the video stream engage in anomalous behavior, relative to the time-of-day (TOD) or day-of-week (DOW) at which the behavior occurs. A machine-learning engine may establish the normalcy of a scene by observing the scene over a specified period of time. Once the observations of the scene have matured, the actions of agents in the scene may be evaluated and classified as normal or abnormal temporal behavior, relative to the past observations.

The invention belongs to the technical field of image processing, and discloses an image conversion method and system based on a generative adversarial network and a ResNts technology. The image conversion method comprises the following steps of firstly, designing an enhanced high-resolution dual GAN image conversion algorithm by utilizing a dual learning method combining two GAN confrontation mechanisms and a norm loss function, and realizing image conversion based on unsupervised learning by adopting an unlabeled data set through a model; secondly, realizing the conversion from source distribution to target distribution by introducing constraint conditions for reconstructing a consistency loss function, and then reconstructing the source distribution; and finally, adding a stable normalization layer in the discriminator. Compared with the resolution of images generated by Pix2pixGAN, CycleGAN and DualGAN, the average values of the ERGAN algorithmPSNR/SSIM provided by the invention are improved by 16%/35%, 2%/9% and 4%/6% respectively.

Described is a method and system for embedding unsupervised learning into three critical processing stages of the spatio-temporal visual stream. The system first receives input video comprising input video pixels representing at least one action and at least one object having a location. Microactions are generated from the input image using a set of motion sensitive filters. A relationship between the input video pixels and the microactions is then learned, and a set of spatio-temporal concepts is learned from the microactions. The system then learns to acquire new knowledge from the spatio-temporal concepts using mental imagery processes. Finally, a visual output is presented to a user based on the learned set of spatio-temporal concepts and the new knowledge to aid the user in visually comprehending the at least one action in the input video.

The invention discloses a Chinese financial news text classification method based on a convolutional neural network. The method is mainly divided into four parts including word vector training, text preprocessing, neural network model training and news classification. Large-scale financial news corpora are used for training to obtain a broadly common financial-category word vector model through anunsupervised learning method, the word vector is effectively imported into the training of the convolutional neural network, and the statistical information of the model is increased through a methodof dynamically regulating the word vector. The adopted convolutional neural network has a simple structure, and can show excellent performances by aiming at a small sample set, so that a Chinese financial news text classification problem can be effectively solved, and effectiveness of the convolutional neural network in processing a text classification problem can be fully proved.