103 results about "Goal variable" patented technology

In many areas, recent developments have generated very large datasets from which it is desired to extract meaningful relationships between the dataset elements. However, to date, the finding of such relationships using prior art methods has proved extremely difficult especially in the biomedical arts. Methods for local causal learning and Markov blanket discovery are important recent developments in pattern recognition and applied statistics, primarily because they offer a principled solution to the variable / feature selection problem and give insight about local causal structure. The present invention provides a generative method for learning local causal structure around target variables of interest in the form of direct causes / effects and Markov blankets applicable to very large datasets and relatively small samples. The method is readily applicable to real-world data, and the selected feature sets can be used for causal discovery and classification. The generative method GLL-PC can be instantiated in many ways, giving rise to novel method variants. In general, the inventive method transforms a dataset with many variables into either a minimal reduced dataset where all variables are needed for optimal prediction of the response variable or a dataset where all variables are direct causes and direct effects of the response variable. The power of the invention and significant advantages over the prior art were empirically demonstrated with datasets from a diversity of application domains (biology, medicine, economics, ecology, digit recognition, text categorization, and computational biology) and data generated by Bayesian networks.

The present invention leverages machine learning techniques to provide automatic generation of conditioning variables for constructing a data perspective for a given target variable. The present invention determines and analyzes the best target variable predictors for a given target variable, employing them to facilitate the conveying of information about the target variable to a user. It automatically discretizes continuous and discrete variables utilized as target variable predictors to establish their granularity. In other instances of the present invention, a complexity and / or utility parameter can be specified to facilitate generation of the data perspective via analyzing a best target variable predictor versus the complexity of the conditioning variable(s) and / or utility. The present invention can also adjust the conditioning variables (i.e., target variable predictors) of the data perspective to provide an optimum view and / or accept control inputs from a user to guide / control the generation of the data perspective.

The present invention leverages machine learning techniques to provide automatic generation of conditioning variables for constructing a data perspective for a given target variable. The present invention determines and analyzes the best target variable predictors for a given target variable, employing them to facilitate the conveying of information about the target variable to a user. It automatically discretizes continuous and discrete variables utilized as target variable predictors to establish their granularity. In other instances of the present invention, a complexity and / or utility parameter can be specified to facilitate generation of the data perspective via analyzing a best target variable predictor versus the complexity of the conditioning variable(s) and / or utility. The present invention can also adjust the conditioning variables (i.e., target variable predictors) of the data perspective to provide an optimum view and / or accept control inputs from a user to guide / control the generation of the data perspective.

A method (and system) of predicting an unobserved target variable includes building a graphical predictive model from domain knowledge, which takes advantage of conditional independence to facilitate inference about the unobserved target variable, given observations of other variables in the graphical predictive model from a plurality of information sources.

Embodiments of the present invention provide a method comprising: providing training data for training at least one mathematical model, wherein the training data is based on past flight information of a plurality of passengers, and the training data comprises a first set of vectors and an associated target variable for each passenger in the plurality of passengers; training at least one mathematical model with the training data; and providing a second set of vectors relating to past flight information of the passenger as inputs to the trained at least one mathematical model and calculating an output of the trained at least one mathematical model based on the inputs, wherein the output represents a prediction of future flight activities of the passenger.

A method and apparatus for autonomous control of unmanned aircraft. A method includes, in a memory of a flight controller associated with an unmanned aircraft, identifying a target to be captured, the identifying comprising a plurality of target variables, identifying a type of the unmanned aircraft, selecting one or more capture routines, defining desired data parameters, and storing the plurality of target variables, the one or more capture routines and the desired data parameters in the memory as a flight path. A system includes a computing device having at least a processor, a memory and a display, the display including a graphical user interface (GUI), and an unmanned aircraft including at least a flight controller, a power supply, a propeller system, a landing gear system, a Global Positioning System (GPS) device, a camera system and a one or more sensors, the flight controller wirelessly linked to the computing device which provides flight path control information to the flight controller through the GUI.

The invention relates to an equipment insurance intelligent pricing method and system based on the Internet of Things. The method comprises the steps that insurance information of equipment is acquired, and original insurance data relevant to the insured equipment is acquired according to the insurance information, wherein the original insurance data at least comprises an equipment type, condition data of the insured equipment, equipment historical insurance data and relevant data of applicants and insurants; the original insurance data is processed to obtain a target variable and a prediction variable relevant to the insured equipment; and a premium pricing algorithm is adopted to calculate the premium of the insured equipment. Through the intelligent pricing method and system, the problem that pricing of a certain specific equipment is not available in the past is solved, and the original insurance data can be adjusted regularly or irregularly. The intelligent pricing method and system are suitable for existing operating systems, tracking systems and communication systems to extract relevant data capable of being applied to insurance.

The invention relates to a basin rainfall monitoring wireless sensor network node optimization layout method comprising the steps of correlation calculation; regression analysis; redundant site removing; additional arrangement of sites; additionally arranged site layout optimization calculation; and similarity calculation so that the optimal scheme of site layout can be obtained. The conventional site extraction method is improved, site network analysis is performed on existing rainfall site points based on the improved site extraction method, the redundant sites are removed, based on which the limitation of the road network on the feasibility of the site layout is directly considered, an optional area for additional arrangement of the sites is determined, the sites are additionally arranged in the optional area, a site layout optimization model is established with regression Kriging variance acting as a target function by comprehensively utilizing the correlation between the target variable rainfall and latitude and longitude, elevation, gradient, slope direction and other environmental variables, and the optimized layout of the rainfall monitoring sites is realized by solving the site layout optimization model in a high-performance way based on a parallel designed simulated annealing algorithm.

The invention relates to a method for establishing a complex product optimization design agent model based on a small sample, belonging to the technical field of complex product optimization design. The method comprises the following steps: setting a goal of the complex product optimization design; generating an original design scheme sample set, the sample capacity of which is S, for the complex product optimization design; generating a virtual design scheme sample set for the complex product optimization design; combining the original design scheme sample set with the virtual design scheme sample set so as to form a mixed design scheme sample set; determining the sensitivities of the target of the complex product optimization design corresponding to various decision variables, and sorting the sensitivities; establishing a three-layer BP neural network model having different input variables by taking the target variable as the output variable; and training various neural network models by taking the mixed sample set as the training sample set; and selecting the neural network model having the optimal performance as the final complex product optimization design agent model. By means of the method, the sample generation workload is reduced; and the precision of the complex product optimization design agent model is ensured.

The invention relates to a method and a system for determining a target variable to be measured in a mobile device. A first physical variable is measured with the aid of a first sensor and a second physical variable with the aid of a second sensor. The second physical variable is different to the first physical variable, or is measured using a different technique. The value of the target variable is calculated with the aid of the measurement of the first and second physical variables. An estimate for the target variable is determined with the aid of at least the measurement of the first physical variable. At least a first error estimate is determined, which depicts the accuracy of the measurement of the first physical variable. The estimate of the target variable is filtered using both the first error estimate and the measurement of the second physical variable.

The invention provides a method, a system and a terminal for predicting age characteristics based on a decision tree model. The method comprises the following steps: collecting basic data information; extracting a characteristic input variable and a target variable in the attribute of the basic data information to get sample data; dividing the sample data into a training set and a test set, inputting the training set to a decision tree model for model parameter training, and applying the model parameter training result to the test set to test the model parameter training result meeting a custom stability condition; outputting the model parameter training result meeting the custom stability condition; and regularly updating the rule of the output model parameter training result to the age prediction result of an unknown user. According to the method, the system and the terminal for predicting age characteristics based on a decision tree model provided by the invention, a prediction model is built to predict the ages of users, the portraits of users are constructed precisely, a solid data foundation is laid for marketing and other scenes, and the accuracy of age identification is improved.

An enhanced modeling tool associated with an entity may facilitate end-to-end modeling of problems in any application space. The enhanced modeling tool may collect modeling data from a variety of sources, check the collected data, find the best predictor variables for a given target variable, estimate the model, implement the model, and validate the model. The output of each of these steps may be in a standardized format to allow other steps to directly incorporate the output. An additional feature of the system may include a reporting capability that generates supporting documents related to model governance and risk compliance.

The invention discloses an industrial process soft measurement method based on an xgboost model. The method firstly performs independent re-sampling on historical data, obtains a training sample set and a verification data set respectively after preprocessing, establishes an xgboost model by using training samples, and then selects the best model parameters by cross validation to determine the soft-measurement model for the target variable. Compared with other nonlinear models, the xgboost model may greatly improve the accuracy and speed of soft measurement, and better fit the variable relationship in complex production processes.

The present invention provides a system and method for an online optimization for a non-linear cognitive wireless. The optimization is achieved by using qualitative model on the system which provides information in the form of monotonic influence diagram. The influence graph represents the state (control variables, goal variables and intermediate variables) and their dependency information using a graphical model incorporating monotonicities information. The present invention also incorporates learning features to build a compact history for reducing bad configurations online. The invention also provides a control engine which achieves the near optimum control of the cognitive wireless network by using the network state information along with the qualitative model.

The invention discloses a fault diagnosis system for electric power measurement automatic verification assembly lines. Production control strategies of assembly lines are independent from one another at present, the production efficiency is severely affected, and potential safety hazards are caused for automatic production equipment. The fault diagnosis system comprises a data acquisition and processing module, a diagnostic analysis module, a system configuration module and a function display module, wherein the diagnostic analysis module implements offline analysis service and online diagnosis service, the offline analysis service includes establishment of a system SDG model, layering of an HSDG model, classification of target variables, establishment of a determined variable PS model and establishment of a variable DKPLS SVR model to be evaluated, and fault factors are extracted and serve as nodes of the HSDG model through manual abstraction of a physical structure of hardware equipment and by combining with historical production data information. The fault diagnosis system can reduce the searching space of effective nodes, thereby accelerating the fault diagnosis speed.

The invention discloses a method for determining parameter influence factors of a cut tobacco drying process based on a partial least square (PLS). The method comprises the following steps of: selecting target variables and independent variables of process data in the cut tobacco drying process, and acquiring the historical data of the target variables and the independent variables; screening the historical data by adopting a Euclidean distance method, and rejecting abnormal data from the historical data to obtain sample data; and establishing a PLS model of the independent variables x and the target variables y by adopting the PLS, and calculating the magnitudes of influence factors of the independent variables x which influence the target variables y according to the model. The method is simple and easy to implement; and a manual experience-based judgment method is replaced by the method, so that the magnitudes of the influence factors can be obtained quickly and accurately.

Provided are systems, methods and techniques for facilitating allocation of resources based on available data estimates, using steps to: input a set of predictor variables and data values pertaining to them; generate a first transfer function that maps values for the predictor variables to values for a predefined set of factor variables; generate distribution information for the predictor variables based on the input data values; conduct a simulation using the distribution information for the predictor variables; map the simulated values for the predictor variables to information regarding the factor variables, using the first transfer function; map such information to estimation information for a set of target variables using a second transfer function; and display the estimation information, use the estimation information to recommend a purchase or sale of an asset, and / or directly purchase and / or sell an asset based on the estimation information.

The invention provides an early-warning method for gear case bearing faults of a wind turbine generator system. The method comprises the following steps: collecting SCADA operation data, selecting parameters, analyzing parameter correlation, determining target variables and analysis variables, conducting interval analysis of the target variables, conducting target variable cluster statistic analysis, and determining states of a wind turbine generator system, wherein the selected parameters include gear case bearing temperature, gear case oil temperature, average wind speed, generator rotatingspeed and active power, correlation coefficients between the parameters are calculated, interval analysis is conducted by taking the analysis variables as an x-axis and the target variables as a y-axis, cluster statistic analysis of the target variables in the same interval is conducted, and whether gear case bearing faults of the wind turbine generator system exist is judged on the basis of faultalarm threshold values. According to the technical solution of the invention, an operation method is simple and highly efficient, the cost is not increased additionally, and fewer system resources are occupied.

A risk assessment method and computing device are provided. Valuable weak variables are added to a risk assessment model, making risk assessment process more comprehensive and more stable, helping to improve the accuracy and objectiveness of risk assessment. In some implementations, the method includes: categorizing a plurality of groups of variables into a first category of variable groups and a second category of variable groups in accordance with correlations between respective groups of variables and a target variable; obtaining a plurality of risk assessment sub-models for respective groups of variables in the second category; obtaining a plurality of sub-model results for the respective risk assessment sub-models in the second category; and obtaining a comprehensive risk assessment model for evaluating the risk of the target variable based on (1) a plurality of variables in the first category and (2) the plurality of sub-model variables in the second category.

A method for optimization of operating parameters of a wind energy installation defines an upper and a lower interval limit value for a parameter to be optimized. The method includes carrying out a cycle with alternate operations of the wind energy installation with the interval limit values, with a data record in each case being produced with a target variable over a variable number of repetitions. The data records relating to the interval limit values are evaluated to form a quality measure, and the interval limit value with the poorer quality measure is identified. Then, at least the interval limit value with the poorer quality measure is replaced by shifting through a step value Δ in a direction of another interval limit value. The cycle is then repeated.

The invention provides an interface test method and system, a server and a readable storage medium. The method comprises the steps of receiving and analyzing an interface test program to obtain targetvariables to be subjected to value assignment and assignment codes of the target variables, wherein the target variables to be subjected to the value assignment include first variables and second variables; according to the assignment codes of the first variables, calling corresponding variable values from a database to be assigned to the first variables; and inputting the test program after thevalue assignment to a to-be-tested interface, and according to the assignment codes of the second variables, assigning a return value of the to-be-tested interface to the corresponding second variables. According to the interface test method and system, the server and the readable storage medium, the variables can be subjected to the value assignment automatically, so that the time and labor are saved and the interface test efficiency is improved.

The invention is suitable for the technical field of communications, and provides a logic regression model building method. An analysis method comprises the steps of obtaining sample data, and preprocessing the sample data; exporting the preprocessed sample data to an Excel document; generating a VBA task, executing the VBA task, and performing monotonicity check and adjustment on the sample datain the Excel document; and importing the sample data subjected to the monotonicity check and adjustment to a database from the Excel document, and taking the sample data as a training set for traininga logic regression model of a target variable. The monotonicity check of the sample data in the process of building the logic regression model is realized, and a visual operation interface is realized, so that the monotonicity of the sample data can be quickly checked and viewed, the monotonicity check efficiency is improved, and the stability and accuracy of the built model are improved.

The invention discloses an input data processing method and device for a categorical data mining model. The method comprises the following steps: receiving data uploaded by a user, and carrying out preprocessing on the data; converting character type data in the data into numeric data; carrying out binning treatment on every continuous variable datum; calculating a preset index value of every variable, and screening a variable which is the most relevant to a preset target variable according to the preset index; and carrying out standardizing treatment on the data. Then operations such as datamodeling and follow-up classifying scoring can be carried out by treated data. Thus, after the data uploaded by the user are received, input data of the categorical data mining model can be processedautomatically, data analyzing personnel are not required, automation of a data processing stage in a data mining process is realized, moreover, operation is simple, and the operators do not need to have professional data analyzing experiences.

The invention relates to a multi-target optimization method for numerical control processing parameters. The method comprises the steps of analyzing a processing request according to a numerical control processing process, and establishing a multi-target optimization mathematical model which represents a relationship between a definite designing variable and a to-be-optimized multi-target variable; solving the multi-target optimization mathematical model by means of a Pareto genetic algorithm, and obtaining a Pareto optimal solution set between the designing variable and the multi-target variable; analyzing relationships between multi-target variable parameters in the Pareto optimal solution set, establishing a contradiction matrix table between target variables according to a TRIZ theory, then analyzing a problem according to a TRIZ theory invention principle, and determining a set of optimal processing parameters from the Pareto optimal solution set according to the analysis result. The multi-target optimization method has advantages of performing an optimization-and-decision mode in multi-target optimization of numerical control processing parameters, overcoming a subjective defect in decision based on preference or experience, truly realizing and perfecting the optimization-and-decision solving mode, and improving optimization effect.

The invention discloses a cigarette weight control system starting position predicting method and a predicting system based on a GPR model. The predicting method comprises the following steps: acquiring target position data of a cigarette weight control system ecreteur of a cigarette maker before stopping as well as target position data after starting, and extracting characteristic variable sets and target variable sets of the target position data before stopping and the target position data after stopping; with the characteristic variable sets and the target variable sets as training samples,initializing and training the GPR model; in accordance with the target position data before stopping in the current time, obtaining a predicted value of a starting position via the constructed GPR model, calculating and obtaining a predicted residual via the predicted value of the starting position and an actual value, and further constructing a residual predicting GPR model, so that a residual predicted value is obtained; and correcting the predicted value of the starting position, so that an output value of the starting position of a weight control system is obtained. The method achieves relatively good robustness; and possible jump influence of sudden change of a production state in some time to an overall predicting result can be effectively prevented.

The invention discloses a control method for parameters self-tuning of tobacco shred production process, comprising the steps of: collecting process parameters of all the processes in the tobacco shred production process; establishing a regression model by adopting process parameters, training the regression model, validating a target variable and an argument, selecting several arguments that havea significant influence on the target variable from the target variable and the argument, screening the independent variable according to a coefficient weight, obtaining an independent variable corresponding to the weight in a preset range, performing a cross validation, taking the process parameters into the targeted regression model, and obtaining a targeted regression model of the corresponding class; when an exception occurs in the process parameters in any category, removing the exception occurred in process parameters of the current process and the output parameters in the next processthrough the targeted regression model corresponding to abnormal process parameters, thereby realizing the control process of the parameters self-tuning. According to the control method and the systemfor the parameters self-tuning of the tobacco shred production process, multi-parameter correlation model of each monitoring control parameter is obtained, and the invention can be used for global parameter optimization of tobacco shred making process.

An embodiment of the application provides a source code obfuscation method and device based on variable and code execution sequence. The method comprises: performing first obfuscation on a target variable, which requires protection, in a source code of a target application; dividing the source code into multiple code blocks based on compiling or operating redirect logic of the source code; performing second obfuscation on execution sequence of target code blocks in the multiple code blocks on the basis of the redirect logic and code block markers corresponding to the code blocks; establishinga redirect table that represents the execution sequence of the multiple code blocks, wherein the code block markers and a mapping relationship of code block addresses subjected to second obfuscation are stored in the redirect table, wherein the code block markers are used for visiting the corresponding code block addresses. The method and device have the advantages that the target variable in thesource code can be obscured and obfuscated, the execution sequence of the source code can be hidden, the ability of the source code to prevent reverse analysis can be improved, and security of information in the application can be improved.

The invention belongs to the technical field of computer software, and discloses a system and method for predicting individual core characteristics based on multi-modal data. The identification moduleis used for identifying the multi-modal data variables closely related to each core trait as input characteristics of the prediction model; the prediction model establishing module is used for establishing a prediction model with an actual prediction effect on a corresponding target variable by fusing the multi-modal characteristics through a machine learning means; and the verification module isused for keeping a prediction model with good reliability and validity as a core of the prediction system after the internal cross verification of the samples and the external verification of the cross samples, carrying a matched intelligent platform, leading out an evaluation report and visually displaying corresponding results. By using the feature set after multi-modal fusion to predict the target variable, not only the interaction among the modal data can be considered, but also the advantages of the modal indexes can be brought into full play, and the prediction effectiveness of the target variable can be increased.