1790 results about "Model selection" patented technology

The present invention provides a promotion pricing system and a related model for producing a value evaluation and recommendation for promotion on a targeted product so as to analyze, evaluate, improve, and design promotions to meet a user's need. The promotion pricing system generates promotion price evaluations and recommendations for each product promotion related to a target product of a user along with associated competing products from the user and competitors. The user can be an individual, an organization, a corporation, an association or any entity providing, including activities related to making, selling, resale, offering for sale, distributing and other commercial conducts, products or service or both in the stream of commerce In the preferred embodiment, the promotion pricing system of the presenting invention is comprised of modularization of the necessary analytical steps along with specifications for these modules. These modules cooperate to implement statistical market response estimation that provide statistically stable, fact-based information on customer response to a promotions. The modules further allow data capture to leverages enterprise and supply chain data sources. The modules include a product segmentation module, an incentive translation module, a customer segmentation module, a data aggregation module, a model selection module, a calibration module, an evaluation module, a constraints generation module, a cost structure module, an optimization module, a market channel performance module, and an alert module.

A speech recognition method includes a model selection step which selects a recognition model and translation dictionary information based on characteristic information of input speech and a speech recognition step which translates input speech into text data based on the selected recognition model and translation step which translates the text data based on the selected translation dictionary information.

An electronic catalog system provides an interface for users to author and post pieces of content, referred to as “blurbs,” for viewing by other users. The blurbs submitted by a particular author are made available for viewing in an author-specific blog (web log) format. Blurbs may also be obtained from external sources, such as from blogs hosted by various web sites. A personalized blurb selection component selects blurbs to present to users based on histories of catalog items selected by such users, and / or based on various other criteria. The blurbs selected for a particular user are presented within a personal log or “plog,” which may be updated daily and will typically contain entries from many different authors. User feedback provided on specific blurbs is taken into consideration by the personalized blurb selection algorithms.

An individually targeted content delivery system and methods which allows content to be delivered to at least one set top box while preserving the privacy of the set top box users. The system and methods of the present invention determine one or more user models for a set top box based on user interactions with the set top box and content to be displayed on the set top box is selected based on the user model. The present invention also allows a set top box to determine when content has been experienced by a set top box user, and to require a set top box user to experience at least one pre-defined content before other content can be viewed.

A system and method of testing and tuning a speech recognition system by providing pronunciations to the speech recognizer. First a text document is provided to the system and converted into a sequence of phonemes representative of the words in the text. The phonemes are then converted to model units, such as Hidden Markov Models. From the models a probability is obtained for each model or state, and feature vectors are determined. The feature vector matching the most probable vector for each state is selected for each model. These ideal feature vectors are provided to the speech recognizer, and processed. The end result is compared with the original text, and modifications to the system can be made based on the output text.

A method for role-based personalization of a collaborative space can include generating a collaborative space utilizing role information for an interacting user that has been defined by an underlying business process model in a workflow. For example, the step of generating a collaborative space can include parsing the workflow to extract a role model, generating a collaborative space domain model from the role model, selecting a plurality of user interface components based upon the role model, organizing the selected user interface components in the collaborative space, and rendering the collaborative space.

A promotion pricing system produces an evaluation so as to analyze, evaluate, improve, and design promotions. The promotion pricing system generates promotion price evaluations and recommendations for each promotion related to a target product, along with analyzing competing products from the same seller and its competitors. The computerized promotion pricing system includes modules for carrying out necessary analytical steps, where the modules cooperate to implement a statistical market response estimation that provide statistically stable information on customer response to promotions using either an attractive or a multiplicative model, where the model is selected dynamically. The modules include a product segmentation module, an incentive translation module, a customer segmentation module, a data aggregation module, a model selection module, a calibration module, an evaluation module, a constraints generation module, a cost structure module, an optimization module, a market channel performance module, and an alert module.

Methods and systems are disclosed for learning a regression decision graph model using a Bayesian model selection approach. In a disclosed aspect, the model structure and / or model parameters can be learned using a greedy search algorithm applied to grow the model so long as the model improves. This approach enables construction of a decision graph having a model structure that includes a plurality of leaves, at least one of which includes a non-trivial linear regression. The resulting model thus can be employed for forecasting, such as for time series data, which can include single or multi-step forecasting.

The spectral kernel machine combines kernel functions and spectral graph theory for solving problems of machine learning. The data points in the dataset are placed in the form of a matrix known as a kernel matrix, or Gram matrix, containing all pairwise kernels between the data points. The dataset is regarded as nodes of a fully connected graph. A weight equal to the kernel between the two nodes is assigned to each edge of the graph. The adjacency matrix of the graph is equivalent to the kernel matrix, also known as the Gram matrix. The eigenvectors and their corresponding eigenvalues provide information about the properties of the graph, and thus, the dataset. The second eigenvector can be thresholded to approximate the class assignment of graph nodes. Eigenvectors of the kernel matrix may be used to assign unlabeled data to clusters, merge information from labeled and unlabeled data by transduction, provide model selection information for other kernels, detect novelties or anomalies and / or clean data, and perform supervised learning tasks such as classification.

A method for monitoring health of a tool includes receiving at least one tool parameter related to the processing of a workpiece in a tool; receiving a model selection trigger; selecting a tool health model based on the model selection trigger; generating at least one predicted tool parameter based on the selected tool health model; and generating a tool health rating for the tool based on a comparison between the measured tool parameter and the predicted tool parameter. A tool health monitor includes a library of tool health models, a model selector, and a fault detection and classification unit. The model selector is adapted to receive a model selection trigger and select a tool health model based on the model selection trigger. The fault detection and classification unit is adapted to receive at least one tool parameter related to the processing of a workpiece in a tool, generate at least one predicted tool parameter based on the selected tool health model, and generate a tool health rating for the tool based on a comparison between the received tool parameter and the predicted tool parameter.

Apparatus and methods for universal node design implementing a universal learning rule in a mixed signal spiking neural network. In one implementation, at one instance, the node apparatus, operable according to the parameterized universal learning model, receives a mixture of analog and spiking inputs, and generates a spiking output based on the model parameter for that node that is selected by the parameterized model for that specific mix of inputs. At another instance, the same node receives a different mix of inputs, that also may comprise only analog or only spiking inputs and generates an analog output based on a different value of the node parameter that is selected by the model for the second mix of inputs. In another implementation, the node apparatus may change its output from analog to spiking responsive to a training input for the same inputs.

A multi-layer, fiber-reinforced composite orthopaedic fixation device having a design selected based on a desired characteristic of the orthopaedic fixation device. The design may be selected according to a model of the device, the model defining design constraints, and the design may comprise a pattern of the fiber angle for each layer. The selection of a design may be analyzed using finite element analysis to determine whether the design will comprise the desired characteristic.

The invention discloses a natural language processing-based multi-language analysis method and device. The method comprises the following steps of: selecting to input a natural language text information language category through a language detection training model; obtaining word embedding expression information of corresponding words which can be recognized by a computer through a trained word vector model, and extracting a keyword of the obtained word embedding expression information through a TF-IDF manner; calculating an article vector and a category vector of each preset category according to the keyword and a keyword weight, and calculating a similarity between an article of natural language text information and each preset category so as to determine a text classification result ofthe natural language text information; and inputting the word embedding expression information of the natural language text information into a trained convolutional neural network and a parallel-framework text emotion analysis model of a bidirectional gate circulation unit, and obtaining a final emotion tendency value through calculation. According to the method and device, the problem that traditional multi-language analysis method needs to know domain knowledges of related linguistics and needs plenty of manpower to carry out operation is solved.

Various embodiments each include at least one of systems, methods, devices, and software for environment mapping with automatic motion model selection. One embodiment in the form of a method includes receiving a video frame captured by a camera device into memory and estimating a type of motion from a previously received video frame held in memory to the received video frame. When the type of motion is the same as motion type of a current keyframe group held in memory, the method includes adding the received video frame to the current keyframe group. Conversely, when the type of motion is not the same motion type of the current keyframe group held in memory, the method includes creating a new keyframe group in memory and adding the received video frame to the new keyframe group.

A system and method for genotyping tandem repeats in sequencing data. The invention uses Bayesian model selection guided by an empirically-derived error model that incorporates properties of sequence reads and reference sequences to which they map.

A system and method are provided which dynamically adapts to a changing economic environment by selecting or synthesizing an economic model from a set of economic models based on the selected model's ability to make accurate predictions about an actual economic market. The method and system each forms a space of different economic models, forms a behavioral landscape by extracting observables from executions of the economic models, and performs model selection and composite model synthesis through optimization over the behavioral landscape.

The invention relates to a temperature rise analytical method for predicting temperature of a permanent magnet in a permanent magnet synchronous motor (PMSM), belonging to the application electrical engineering design field; the method is characterized in that: distributed heat source of a motor is analyzed by a filed-circuit compact coupling method, comprising eddy current loss in the permanent magnet, iron loss in an iron core and copper loss in armature; on the consideration of precision requirements, the coupling analysis of a magnetic field and a temperature filed can be realized by single-way coupling mode. A thermal model of the permanent magnet synchronous motor is built based on a mixing method of a novel equivalent heat network and a finite element, heat parameters are rationally selected by adopting a combining mode of experimental measurement and empirical formula, the heat transferring coefficient and cooling condition of the motor are described completely, a stator and a rotor can be systematically combined by adopting air gap joints in the heat network, the stator and rotor unified temperature rise model is formed, the difficulty of measuring air gap temperature is avoided, material parameters are adopted at the practical working temperature, so as to lead the analysis to be rational; the accurate and optical method for predicting the temperature of the permanent magnet is realized by special correction processing in experimental links; in addition, the design method is used to give suggestions for model selection of the permanent magnet material in the motor.

The invention relates to a wireless sensor network node positioning method based on a received signal strength indicator (RSSI). The precision of the traditional method is not high, and the traditional method is easily disturbed by environment. In the method, an effective RSSI value is selected by using a Gaussian distribution function model in the aspect of reading RSSI value, so that small probability events during RSSI measurement are removed to a certain extent, and the precision of RSSI value between nodes is improved; and the coordinates of an unknown node are obtained by a triangular positioning method, and the unknown node is circularly refined via the distribution probability model of the unknown node, so that a point with the maximum distribution probability is found and is used as the final positioning coordinate. In the method, the signal intensity and distance information between anchor nodes are introduced and are used as the reference; the unknown node coordinate is found out via the distribution probability model of the unknown node; the distance measurement precision and the positioning precision between the unknown node and the anchor node are improved; and the method is not easily disturbed by environment.

The invention provides an actual-measurement modeling method for a prime mover and a speed governor thereof of an electric power system, belonging to the technical field of electric power systems. The invention solves the problem that the traditional prime mover and the speed governor thereof generate larger bias of a simulated calculation result due to the use of a typical model and typical parameters. The actual-measurement modeling method for the prime mover and the speed governor thereof of the electric power system comprises the following steps of: (1) collecting data; (2) establishing mathematical models of the prime mover and the speed governor thereof; (3) carrying out model selection; (4) carrying out segmented field testing on the prime mover and the speed governor thereof; (5) carrying out segmented parameter recognization to obtain parameters of each segment; (6) screening data before the population parameters are recognized; (7) carrying out population parameter recognization; (8) carrying out model parameter verification; (9) carrying out frequency large-disturbance testing; and (10) carrying out model parameter verification. By means of the actual-measurement modeling method, the models and the parameters can be ensured to be accurately applied to BPA (Bonneviklle Power Administration), PSS / E (Power System Simulator / Engineering) electric power system analysis software, and the electric power system simulation precision is improved.

The invention discloses a method for reconstructing a three-dimensional scene of a single image, which comprises the following steps of: a image input step, inputting each frame of image in an image sequence; a feature extracting step, extracting features from a scene of a image, carrying out scene classification and object recognition based on the extracted features to acquire semantic information, and simultaneously extracting monocular geometrical information based on the extracted features to acquire the monocular geometrical information; an object detection step, carrying out an object detection based on the extracted features and with the scene classification as a reference; a three-dimensional graphic primitive model selection step, selecting a three-dimensional graphic primitive model according to a result of the object detection; and a scene three-dimensional model generation step, carrying out inference and verification on the scene three-dimensional model according to scene semantic prior, the three-dimensional graphic primitive model and the monocular geometrical information to generate a final scene three-dimensional model.

An image processing device for generating a 3-D model image of a target object included in an input image, comprises a face image input means for inputting a face image; a 3-D model input means for inputting one or a plurality of 3-D models for each of a plurality of parts; a 3-D model selection means for selecting a 3-D model for an arbitrary part of the plurality of parts based on an instruction input by an operator; a face image mapping means for mapping the face image input via the face image input means to the 3-D model selected by the 3-D model selection means and for displaying the mapped 3-D model; and an image generation means for generating a 3-D still image using the 3-D model selected by the 3-D model selection means and the face image input by the face image input means.

An optical metrology system includes model approximation logic for generating an optical model based on experimental data. By eliminating theoretical model generation, in which the fundamental equations of a test sample must be solved, the model approximation logic significantly reduces the computational requirements of the metrology system when measuring films formed on patterned base layers. The experimental model can be created by selecting an expected mathematical form for the final model, gathering experimental data, and compiling a lookup model. The lookup model can include the actual measurement data sorted by output (attribute) value, or can include “grating factors” that represent compensation factors that, when applied to standard monolithic model equations, compensate for the optical effects of grating layers.

Dynamic parameter tuning using particle swarm optimization is disclosed. According to one embodiment, a system for dynamically tuning parameters comprising a control unit; and a system for receiving parameters tuned by the control unit. The control unit receives as input a model selection and definitions, and dynamically tunes a value for each parameter by using a modified particle swarm optimization method. The modified particle swarm optimization method comprises moving particle locations based on a particle's inertia, experience, global knowledge, and a tuning factor. The control unit outputs the dynamically tuned value for each parameter.

A speech recognition method comprises model selection step which selects a recognition model based on characteristic information of input speech and speech recognition step which translates input speech into text data based on the selected recognition model.

The invention discloses a traffic state estimation device and method based on data fusion. The traffic state estimation device comprises a pattern classifying unit, a model selecting unit and a data fusing unit, wherein the pattern classifying unit is used for classifying traffic data of one or more roads into one of a plurality of preassigned patterns; the model selecting unit is used for selecting a corresponding neural network model from a plurality of neural network models according to the classified patterns; and the data fusing unit is used for inputting traffic data into the selected neural network model for carrying out data fusion and estimating a traffic condition of one or more roads. According to the invention, a multi-pattern classifying and multi-pattern modeling mechanism is introduced, aiming at each road, different patterns are respectively independently modeled, thus the precision of the single model can be improved, and different data source types of the traffic data, different characteristics and different factors of influencing the traffic conditions are fully utilized to obtain a more realistic and accurate fusion result and traffic condition estimation.

Model selection is performed. First information is obtained from a user about a presenting issue. The first information is used within a supermodel to identify an underlying issue and an associated sub model for providing a solution to the underlying issue. A Bayesian network structure is used to identify the underlying issue and the associated sub model. The sub model obtains additional information about the underlying issue from the user. The sub model uses the additional information to identify a solution to the underlying issue.

A method includes estimating structure models for a voxel(s) of a spectral image based on a noise model, fitting structure models to a 3D neighborhood about the voxel(s), selecting one of the structure models for the voxel(s) based on the fittings and predetermined model selection criteria, and de-noising the voxel(s) based on the selected structure model, producing a set of de-noised spectral images. Another method includes generating a virtual contrast enhanced intermediate image for each energy image of a set of spectral images corresponding to different energy ranges based on de-noised spectral images, decomposed de-noised spectral images, an iodine map, and a contrast enhancement factor; and generating final virtual contrast enhanced images by incorporating a simulated partial volume effect with the intermediate virtual contrast enhanced images. Also described herein are approaches for generating a virtual non-contrasted image, a bone and calcification segmentation map, and an iodine map for multi-energy imaging studies.