254 results about "Model application" patented technology

The invention discloses a harbor district road traffic demand predicting system which is based on a TransCAD macro simulated platform and is used to obtain harbor district road traffic generation amount in an objective year. The predicting system at least comprises a storage module, a harbor district road network model, a road network model application module, a road network loading distribution unit, an analysis evaluation module and a planning module, wherein the storage module is used to store data basis for predicting harbor district road traffic generation amount; the harbor district road network model inputs a harbor district project map into a TrarsCAD model platform through a harbor district project geographical information database so as to establish the harbor district road network model according to road traffic circulation in a harbor district; the road network model application module optimizes and selects traffic parameters by means of genetic algorithm to obtain a harbor district objective year OD matrix; the road network loading distribution unit is used for obtaining the traffic flow distribution state and traffic circulation state of the entire road network; the analysis evaluation module combines with the traffic distribution result to carry out traffic adaptability analysis evaluation on a future road network planning scheme; and the planning module is used to put forward guidance instructions and overall measures with regard to harbor district road traffic planning.

The system and method relate to detection of objects that are submerged, or partially submerged (e. g. floating), relative to a water surface. One aspect of the invention emits LIDAR fan-beam pulses and analyzes return-pulse portions to determine water-surface orientations and derive submerged-object images corrected for refractive distortion. Another defines simulated images of submerged objects as seen through waves in a water surface, prepares an algorithm for applying a three-dimensional image of the water surface in refractive correction of LIDAR imaging through waves—and also models application of the algorithm to the images, and finally specifies the LIDAR-system optics. Yet another emits nearly horizontal pulses to illuminate small exposed objects at tens of kilometers, detects reflected portions and images successive such portions with a streak-tube subsystem. Still others make special provisions for airborne objects.

An integration server system for mapping data objects on a database schema offers a strongly-typed model API, complex constraint management, and association balancing. The integration server system also has improved performance through optimized handling of ordered associations of data object and of string-valued attributes. The integration server system comprises a database schema configured to store a set of data object instances. A metadata model represents a configuration of the set of data object instances in the database schema. A model application programming interface provides a client application with access to the set of data object instances, and a metadata application programming interface provides a client application with access to the set of data object instances via the metadata model.

A system for using machine-learning to create a model for performing integrated circuit layout extraction is disclosed. The system of the present invention has two main phases: model creation and model application. The model creation phase comprises creating one or more extraction models using machine-learning techniques. First, a complex extraction problem is decomposed into smaller simpler extraction problems. Then, each smaller extraction problem is then analyzed to identify a set of physical parameters that fully define the smaller extraction problem. Next, models are created using machine learning techniques for all of the smaller simpler extraction problems. The machine learning is performed by first creating training data sets composed of the identified parameters from typical examples of the smaller extraction problem and the answers to those example extraction problems as solved using a highly accurate physics-based field solver. The system them uses the created training sets to train neural networks that will be used to model the extraction problems. Bayesian inference is used to train the neural networks models. Bayesian inference may be implemented with normal Monte Carlo techniques or Hybrid Monte Carlo techniques. After the creation of a set of models for each of the smaller simpler extraction problems, the machine-learning based models may be used for extraction.

A system for using machine learning based upon Bayesian inference using a hybrid Monte Carlo method to create a model for performing integrated circuit layout extraction is disclosed. The system of the present invention has two main phases: model creation and model application. The model creation phase comprises creating one or more extraction models using machine-learning techniques. First, a complex extraction problem is decomposed into smaller simpler extraction problems. Then, each smaller extraction problem is then analyzed to identify a set of physical parameters that fully define the smaller extraction problem. Then, for each of the smaller simpler extraction problems, complex mathematical models are created using machine learning techniques. The machine learning is performed by first creating training data sets composed of the identified parameters from typical examples of the smaller extraction problem and the answers to those example extraction problems as solved using a highly accurate physics-based field solver. Next, the system uses Bayesian inference implemented with a hybrid Monte Carlo method to train a set of neural networks for extraction problems. After the creation of a set of models for each of the smaller simpler extraction problems, the machine-learning based models may be used for extraction.

A system for using machine-learning to create a model for performing integrated circuit layout extraction is disclosed. The system of the present invention has two main phases: model creation and model application. The model creation phase comprises creating one or more extraction models using machine-learning techniques. First, a complex extraction problem is decomposed into smaller simpler extraction problems. Then, each smaller extraction problem is then analyzed to identify a set of physical parameters that fully define the smaller extraction problem. Next, models are created using machine learning techniques for all of the smaller simpler extraction problems. The machine learning is performed by first creating training data sets composed of the identified parameters from typical examples of the smaller extraction problem and the answers to those example extraction problems as solved using a highly accurate physics-based field solver. The training sets are then used to train the models. In one embodiment, neural networks are used to model the extraction problems. Bayesian inference is employed by one embodiment in order to train the neural network models. Bayesian inference may be implemented with normal Monte Carlo techniques or Hybrid Monte Carlo techniques. After the creation of a set of models for each of the smaller simpler extraction problems, the machine-learning based models may be used for extraction.

A system and method of managing computing systems and, more particularly, to a system and method for modeling applications in a network. The method includes determining dependency of components in a shared distributed computing environment and measuring consumption of the dependent components. The method further includes building a model to compare component behavior with a current load of the dependent components to optimize component performance or to find a deviation and hence a “troubled spot”. The system includes a local agent configured to perform the method and to provide data to a centralized storage.

The invention relates to the technical field of machine learning, and especially relate to a method and a system for integrated machine learning convenient for data analysis personnel to use. The method comprises the following steps: (1) data exploring; (2) data cleaning; (3) feature extraction; (4) feature selection; (5) sampling; (6) model training; (7) model optimization; (8) model combination;(9) model interpretability; (10) nature language processing. The system comprises a data processing module, a feature processing module, a model processing module, and a nature language processing module. The method and the system provide a unified algorithm modeling process for machine learning engineers, students, teachers, and machine learning fans, so that the machine learning engineers, students, teachers, and machine learning fans complete a modeling process by 20% efforts, and concentrate 80% efforts on understanding of business and model application, to deeply understand business andpreferably realize requirements of business personnel on models.

A system for using machine-learning to create a model for performing integrated circuit layout extraction is disclosed. The system of the present invention has two main phases: model creation and model application. The model creation phase comprises creating one or more extraction models using machine-learning techniques. First, a complex extraction problem is decomposed into smaller simpler extraction problems. Then, each smaller extraction problem is then analyzed to identify a set of physical parameters that fully define the smaller extraction problem. Next, models are created using machine learning techniques for all of the smaller simpler extraction problems. The machine learning is performed by first creating training data sets composed of the identified parameters from typical examples of the smaller extraction problem and the answers to those example extraction problems as solved using a highly accurate physics-based field solver. Next, the system trains a set of neural networks using the training sets. In one embodiment, Bayesian inference is used to train the neural networks that are used to model the extraction. After the creation the neural network based models for each of the smaller simpler extraction problems, the neural network based models may be used for extraction.

An open community model for information exchange enables individuals to plan and attain superior outcomes in a dynamic environment. A continually “world model” maintains information about the dynamic environment that is valuable for individuals of the open community. An open community model application publicizes a specification of relevant types of information it receives from suppliers and provides to consumers. Suppliers provide information consistent with the specification. Consumers utilize received information to reduce uncertainties or errors in the assumed environment so they can generate improved plans or improve their expected outcomes. Consumers give feedback to the open community model about the quality of information they received. A “world model” application combines information received to update and improve its understanding of the dynamic environment and its estimates of relevant parameters. The community model simplifies the creation and operation of information markets where innovative products are naturally selected and economically reinforced.

A system for using machine-learning to create a model for performing integrated circuit layout extraction is disclosed. The system of the present invention has two main phases: model creation and model application. The model creation phase comprises creating one or more extraction models using machine-learning techniques. First, a complex extraction problem is decomposed into smaller simpler extraction problems. Then, each smaller extraction problem is then analyzed to identify a set of physical parameters that fully define the smaller extraction problem. Next, models are created using machine learning techniques for all of the smaller simpler extraction problems. The machine learning is performed by first creating training data sets composed of the identified parameters from typical examples of the smaller extraction problem and the answers to those example extraction problems as solved using a highly accurate physics-based field solver. The training sets are then used to train the models. In one embodiment, neural networks are used to model the extraction problems. To train the neural network models. Bayesian inference is used in one embodiment. Bayesian inference may be implemented with normal Monte Carlo techniques or Hybrid Monte Carlo techniques. After the creation of a set of models for each of the smaller simpler extraction problems, the machine-learning based models may be used for extraction.

The invention provides a multi-cycle installment decision making method and a multi-cycle installment decision making system. The multi-cycle installment decision making method comprises the steps of: acquiring user credit investigation information of a target user according to an installment request input by a user end or an external trigger, wherein the user credit investigation information comprises anti-fraud information and credit information of the user; applying for anti-fraud scanning based on the anti-fraud information and a called anti-fraud model, so as to obtain risk rating information; applying for credit scanning based on the credit information and a called credit model, so as to obtain credit rating information; and matching the risk rating information with a risk requirement and matching the credit rating information with a credit requirement in installment decision making, so as to output matched installment decisions. The multi-cycle installment decision making method and the multi-cycle installment decision making system can extract effective user information for data processing through classifying and integrating a large amount of user information, and feed back installment decision making schemes timely and accurately.

A model application unit calculates linear prediction coefficients of a multi-step linear prediction model by using discrete acoustic signals. Then, a late reverberation predictor calculates linear prediction values obtained by substituting the linear prediction coefficients and the discrete acoustic signals into linear prediction term of the multi-step linear prediction model, as predicted late reverberations. Next, a frequency domain converter converts the discrete acoustic signals to discrete acoustic signals in the frequency domain and also converts the predicted late reverberations to predicted late reverberations in the frequency domain. A late reverberation eliminator calculates relative values between the amplitude spectra of the discrete acoustic signals expressed in the frequency domain and the amplitude spectra of the predicted late reverberations expressed in the frequency domain, and provides the relative values as predicted amplitude spectra of a dereverberation signal.

A system for using machine learning based upon Bayesian inference using a hybrid monte carlo method to create a model for performing integrated circuit layout extraction is disclosed. The system of the present invention has two main phases: model creation and model application. The model creation phase comprises creating one or more extraction models using machine-learning techniques. First, a complex extraction problem is decomposed into smaller simpler extraction problems. Then, each smaller extraction problem is then analyzed to identify a set of physical parameters that fully define the smaller extraction problem. Next, complex mathematical models are created using machine learning techniques for all of the smaller simpler extraction problems. The machine learning is performed by first creating training data sets composed of the identified parameters from typical examples of the smaller extraction problem and the answers to those example extraction problems as solved using a highly accurate physics-based field solver. Next, the system uses Bayesian inference implemented with a Monte Carlo method to train a set of neural networks for extraction problems. After the creation of a set of models for each of the smaller simpler extraction problems, the machine-learning based models may be used for extraction.

The invention discloses a system and method for monitoring loads on tires in real time in the automobile driving process. The method is characterized by including the steps of firstly, calibrating mapping relations of the loads on the tires of an automobile, pressure on the tires, temperatures of the tires and perpendicular deformation amounts of rubber tire bodies of the tires through bench tests; secondly, evaluating the loads on the tires in real time in the automobile driving process through an automobile velocity test device, a tire pressure monitoring device, a tire velocity sensor and the like. The method is mature in concept, the designed monitoring device has high test accuracy, and the vehicle model application range is wide.

The invention discloses a material constitutive model numerical analysis method taking damage accumulation effect into consideration. The method includes: step A, defining damage variable D; step B, acquiring material monotonic loading mechanical performance according to a stress-strain curve of a material; step C, calculating damage parameters beta, zeta 1 and zeta 2 according to a hysteretic curve of a material different in strain amplitude; step D, acquiring undetermined parameters m0, m1, m2 and m3 according to the hysteretic curve of the material, and building a material circulation constitutive model taking damage accumulation effect into consideration; step E, through ABAQUS finite element software, substituting related parameters of the material into a material user subprogram VUMAT to accurately simulate damage of the material under circulation load action. By the method, difficulty of accumulation damage model application at the present stage can be solved effectively, damageparameters can be specified, calculation accuracy can be improved, application of the material constitutive model taking damage accumulation effect in the finite element software is realized, and damage to the material under the circulation load action can be simulated accurately.

The invention discloses simulation of a process of discrete events or tasks having a plurality of available resources associated therewith. A database stores a plurality of models, each including a plurality of one or more entity, task, and resource parameter, and dependencies and relationships. A model application communicates with the database and is configured to receive commands from a user, to retrieve one of the plurality of models and the corresponding plurality of one or more entity, task, and resource parameter in response to a user command, to receive input data corresponding to one or more entity, task, and resource parameter from a business database system, and to generate a simulation model based on the business database system and the input data. An optimizing application in communication with the model application and is configured to receive commands from a user, to select one or more entity, task, and resource parameter of the simulation model with respect to an objective function, to define bounds of the one or more entity, task, and resource parameter selected, to generate values for the objective function based on the one or more of the entity, task, and resource parameter selected, and to generate financial performance data based on the values generated for the objective function. A server performs a simulation of the process by processing the simulation model and generates an output data file containing output data representative thereof. The objective function comprises a combination of system financial performance measures (e.g., operational margin) and process performance measures (e.g., cycle time, throughput, utilization.

The invention proposes a convolutional neural network-based human body behavior recognition method. The method comprises three stages, namely, a data acquisition and preprocessing stage, a model establishment and training stage and a model application stage. According to the method of the invention, an established convolutional neural network structure can effectively identify six kinds of daily human behaviors on the basis of simple data preprocessing and provide a good foundation for transplanting the recognition method into a smart phone with limited computing power. The method can be applied to application fields such as smart phone end-based human body daily activity recording, health tracking and health monitoring.

The invention discloses a method using processing parameter to predict control indexes during tobacco processing procedures. The method includes data collection, data processing, relativity analysis, model establishment and model application. The method includes steps as follows: using the correlation analysis method and the discrimination analysis method to perform characterization analysis to processing parameters and control results of single procedures during the tobacco production process; establishing the mathematical model between numerous processing control parameters and corresponding control indexes of key procedures; and further, representing the relevance between the procedure processing parameters and the control indexes, and applying the model to the production process, so as to realize basis-based and directional adjustment to the control parameters. The method overcomes the defect that the present procedure capability evaluation method can only perform the control capability evaluation on the single process parameters or the process indexes, and can perform relevant description evaluation between parameters and indexes and between the indexes, thereby providing powerful basis support to the index adjustment and process testing.

A deployment modeling platform enables a user to model application characteristics of target software and to associate application modeling parameters to the modeled application characteristics. A user may also model environment characteristics of a target deployment environment and to associate environment modeling parameters to the modeled deployment environment characteristics. Still further, a user may create a deployment model that associates and maps selected parameters of the modeled application characteristics of the target software to associated parameters of the modeled environment characteristics of the deployment environment, and to verify that each parameter that relates to a requirement is mapped to and is fulfilled by an associated parameter that relates to a corresponding capability to determine whether validation problems exist in order to deploy the target software in the associated deployment environment.

The invention discloses an assessment computing method of coalbed methane reserve volume at a coal mining stable region. The method comprises the following steps of: (1) obtaining effective pressure relief range of a mining stable region; (2) judging application condition of an estimation model; (3) computing surrounding rock pore volume in the mining stable region; (4) obtaining free gas concentration in the mining stable region; (5) estimating residual coal quantity in the mining stable region and residual gas content; (6) estimating the coalbed methane reserve volume in the mining stable region. The study on the effective pressure relief range in the mining stable region is capable of determining the estimation border range of the coalbed methane reserve volume in the mining stable region, the estimation on the estimation model application condition is capable of selecting effective estimation region to guarantee the accuracy of the estimation result, the computing of surrounding rock pore volume and the study on the free gas concentration are capable of accurately estimating the total quantity of the free coalbed methane in the mining stable region, and the study on the residual coal quantity in the mining stable region and residual gas content is capable of accurately estimating the total quantity of adsorption coalbed methane in the mining stable region.

Embodiments of the invention provide a technique to model applications and storage used thereby, and to monitor application service level objectives. An aspect is directed to a method for managing a computer system which includes at least one host computer and at least one storage system which are connected via a network, the at least one host having an application running thereon. The method comprises modeling an application running on a host of a computer system as an application object, which includes associating storage utilized by the application with the application object; monitoring one or more application-based storage service level objectives for the storage utilized by the application, each service level objective being defined by a user to be threshold specific to the application; and alerting the user when any of the one or more service level objectives being monitored is violated.

According to some embodiments, account information may be received in connection with a potential insurance policy. A premium indication portal processor may receive, from a risk score model application, an account score matrix for the potential insurance policy, including grade values comparing the account information with other insured policies in a risk database, along with a benchmark premium value calibrated to a target return on equity based on the account information and information in the risk database. The account score matrix may be displayed on an underwriter device, and guide indication adjustments may be received from the underwriter device for the potential insurance policy. The premium indication portal processor may then automatically calculate an adjusted premium value calibrated to the target return on equity based at least in part on the guide indication adjustments.

The present invention provides a computer-based system for evaluating risk in asset backed securities (ABS) comprising: a database containing data associated with an asset pool of an ABS; a computer having a processor for executing software and being adapted to establish a communication link with an external provider of electronic data and to receive a first data set associated with an asset pool of an ABS, the first data set including credit score data related to the asset pool; and a migratory pattern predictive model application executed by the processor and adapted to analyze at least a part of the first data set, including the credit score data, and to determine a rating concerning the relative risk associated with the ABS.

The invention provides a basin hydrologic model parameter calibration method. The method is characterized in that focusing on the key issue of conceptual model parameter calibration, the process is divided into the three steps of basic data preparation and processing, parameter cognition and parameter optimization, wherein during basic data preparation and processing, for a lumped hydrologic model, special attention needs to be paid to analysis of periodical floods, and the floods need to be classified according to different rainfall central positions and flood magnitudes to be selectively used for parameter optimization; parameter cognition aims to clarify the influence of a value change of each parameter on the hydrologic process, which is the most basic to model application; and duringparameter optimization, parameters are divided into runoff producing parameters and converging parameters for step-by-step optimization, and an appropriate objective function needs to be selected according to an expected goal at each optimization stage, which can ensure that an optimization result meets water balance while meeting flood process similarity, and optimization efficiency is improved.

The invention relates to a layered design method of a model-based automatic transmission software development platform. The software development platform is divided into three layers, namely a model application layer, a hardware abstraction layer and a bottom-layer driving layer, wherein the hardware abstraction layer is used for jointing the model application layer and the bottom-layer driving layer; an automatic transmission control strategy is characterized by modelling, emulating and testing by a model development tool, automatically generating C codes and converting the C codes into an application layer software module to form the model application layer; the model application layer carries out data exchange by using the hardware abstraction layer and realizes access of bottom-layer platform data and corresponding operation by a joint function; the bottom-layer driving layer carries out controller area network (CAN) driving, analogue / digital conversion, digital input and output, and own function and state monitoring of the platform; and the bottom-layer driving layer executes the corresponding operation by function call of the hardware abstraction layer, and processes and updates data in real time.

The invention discloses a smart precision marketing system for big data. The system comprises a big data platform, a data fusion unit, a model analysis unit, a model optimization unit and a model application unit, wherein the data fusion unit is used for collecting and cleaning multi-source heterogeneous data and performing data correlation analysis; the model analysis unit is used for deeply analyzing a data center and making a corresponding analysis model according to a business scenario; the model optimization unit is used for acquiring a model selection optimization scheme and a characteristic change scheme through monitoring, comparison and analysis for the analysis model; the model application unit is used for realizing fusion of the trained model or processed data with tools and systems of an e-commerce platform. Travellers are precisely differentiated and the special analysis models are established with a distributed deep learning algorithm through collection and fusion of traveller data on the basis of a big data technology, and the smart precision marketing system is constructed by tracking the model effect and optimizing the model continuously.