119 results about "Information Criteria" patented technology

Systems and methods are described that determine which offers for credit-based transactions (if any) to market to a person. Each offer is associated with a set of third party demographic information criteria, a set of lender demographic information criteria, and a set of lender credit data criteria. The set of third party demographic information criteria is applied to a set of offers to determine a first subset of offers. The set of lender demographic information criteria is applied to the first subset of offers to determine a second subset of offers. The set of lender credit data criteria is applied to the second subset of offers to determine a third subset of offers.

Systems and methods for creation, management and delivery of targeted marketing campaigns to a consumer based on a consumer's social check-in activity as identified by a consumer on one or more location based social networks. In one embodiment, a campaign management system receives information relating to a consumer's social check-in activity from various (heterogeneous) location based social networks in a wide variety of formats, and further normalizes such disparate information received into a standardized format. According to an aspect, the system processes the normalized information in conjunction with pre-stored marketing campaigns created by marketers for purposes of identifying consumers who satisfy one or more conditions specific to the pre-stored marketing campaigns. Campaign-related information corresponding to the pre-stored marketing campaigns generally comprise various promotional offers, discounts, coupons, etc., which are delivered to the identified consumers, who initiate some kind of action thereby fulfilling the marketing campaign.

A topic analyzing method is provided in which the number of main topics in text data which is added in time series and generation and disappearance of topics are identified in real time as needed, and features of main topics are extracted and thereby one can know a change in the content of a topic with a minimum amount of memory and processing time. There is provided a system that detects topics while sequentially reading text data in a situation where the text data is added in time series, including learning means for representing a topic generation model by a mixture distribution model and learning the topic generation model online while more-heavily discounting the older data on the basis of a timestamp of the data; and model selecting means for selecting an optimal topic generation model from among a plurality of candidate topic generation models on the basis of information criteria of the topic generation models, wherein the topics are detected as mixture components of the optimal generation model.

The invention discloses a power system fault diagnosis method comprehensively using electricity amount and timing sequence information. Firstly, power-off areas before and after a fault are compared and analyzed to determine the fault area, and a suspicious element set is formed; secondly, a weighting fuzzy timing sequence Petri net model is built, and SCADA information, electricity amount information of a WAMS and the timing sequence characteristics included in the information are integrated to form the alarm information criterion, place delay restraints and the electricity amount criterion; a Gaussian function is adopted, the confidence coefficient of alarm information is obtained in combination with timing sequence reasoning, the initial confidence coefficient of the Petri net model is calculated through array calculation, and the model is solved; finally, backward reasoning is carried out according to the fault probability of an element, and protection and judgment of maloperation and motion refusal of a disconnector are carried out. The power system fault diagnosis method is high in fault-tolerant capability, can handle the phenomena of protection and maloperation/motion refusal of the disconnector and the phenomenon that an alarm is lost or wrong, and improves the accuracy and reliability of the fault diagnosis result.

The invention provides a method for building a software production line platform based on Test Director, Harvest or Subversion, which is characterized in that: by combing the information standards, information flow and information dependency relationship during the process of software research and development; then general defects, modification and configuration management rules (such as flows, tables, fields, roles, authority and protocols) are formulated; unified bases for quality measurement and statistical analysis are established, and a set of measurement models for product quality and process quality is formed. Furthermore, an assistant software research and development synergic management system is also provided for fulfilling support; the operations of service linkage connection, rule mapping and information synchronization are performed for Test Director, Harvest or Subversion. Therefore, the invention has the advantages that: the multi-dimensional data acquisition for demand management, configuration management, alteration management, release management, project management and defect management can be completed, and a software enterprise asset library with a full life cycle is thus formed.

The invention discloses a first-arrival travel time picking method and device for a seismic signal comprising noise, and the method comprises the following steps: carrying out the decomposition of an original seismic signal s(t) comprising noise through a noise adaption complete set EMD (Empirical Mode Decomposition) method, and obtaining a series of finite IMF components from high frequency to low frequency and residual components; screening out a plurality of high-frequency IMF components according to an EIF effective index function, and carrying out the wavelet threshold denoising; carrying out the accumulation and reconstruction of the high-frequency IMF components denoised through wavelet threshold denoising and the high-frequency IMF components which are not denoised, i.e., obtaining the denoised seismic signal x(t); initially determining the first-arrival travel time t0 through an improved sliding time window energy ration method; calculating a local extreme value in a time window comprising the moment t0 through an AIC information criterion method, wherein the time corresponding to the extreme value is the precise first-arrival travel time t. According to the invention, the method can effectively improve the seismic noise inhibition effect and improve the picking precision of the first-arrival travel time, and creates the conditions for the accurate processing and explanation of the subsequent seismic data.

A computer implemented method identifying calcification in a patient image data set including blood vessels. The method includes: obtaining an image data set including voxels each having an intensity value; forming the intensity values into an intensity value group covering an intensity range; defining plural intensity thresholds across the intensity range and including its end values; for each intensity threshold, partitioning the intensity values into two sub-groups according to intensity threshold, and calculating an information criterion based on intensity threshold; generating an information criterion measure curve that plots the calculated information criteria against intensity threshold; locating a maximum in the information criterion measure curve and setting the corresponding intensity threshold as a calcification threshold; and partitioning the intensity values into two sub-groups using the calcification threshold, identifying voxels corresponding to intensity values in the sub-group above the calcification threshold as representing calcification in the patient.

The invention discloses a hidden-Markov-based Internet network delay forecasting method in the technical field of network delay forecasting. The method comprises the following steps of: acquiring observable status and an observable status sequence according to a historic delay data set and preset delay forecasting precision; clustering the historic delay data set by a K-Means clustering method, computing the error of the historic delay data set under different k values, and confirming an initial value according to the error of the historic delay data set under the different k values; estimating hidden Markov parameters under the different k values, computing a hidden Markov bayes information criterion value under each k value according to the hidden Markov parameters under the different k values, and selecting the k value, which corresponds to the minimum hidden Markov bayes information criterion value, as an optimal hidden status number k-best; and forecasting future delay according to the observable status and the optimal hidden status number k-best. According to the method disclosed by the invention, the rule of the delay data set and the characteristic of the Internet network can be exactly expressed, and the accuracy of forecasting of the future observable status is high.

A system and method for measuring various weld characteristics is presented. The system and method can comprise a means to measure penetration depth of butt welds in thin plates, for example, using laser generated ultrasounds. Superimposed line sources (SLS) can be used to generate narrowband ultrasounds. A signal processing procedure that combines wavenumber-frequency (k-ω) domain filtering and synthetic phase tuning (SPT) is used to reduce the complexity of Lamb wave signals. The reflection coefficients for different wavelengths corresponding to each wave mode can be calculated. Regression analysis that can include stepwise regression and corrected Akaike's information criterion (AIC) can be performed to build prediction models that use the reflection coefficients as predictors.

Systems and techniques for collecting information as authorized by information providers and sharing the information with information recipients according to criteria specified by the information providers. Information is collected from one or more of a variety of sources and stored in a provider profile, with the provider profile also specifying criteria for sharing the information, including payment required for sharing the information with particular categories of recipients. An exchange system is maintained allowing recipients to request or to otherwise specify needs for particular categories of information and payments to be provided by the information, and needs or requests of recipients for information are matched with criteria specified by providers, with information being transferred or used to provide results for a recipient and payment being transferred from the recipient to a provider or providers when a match between information needs and criteria for sharing information is identified.

The invention provides a wireless network flow prediction method based on a local minimax probability machine, which comprises the following steps of: calculating time delay of a time sequence by using a mutual information algorithm; calculating an embedding dimension by using a Cao method; carrying out phase space reconstruction on the time sequence by using a delay coordinate state phase space reconstruction method to obtain a phase space; calculating the number of proximal points of the phase space by using an AICi information criteria; constructing a proximal point set by using a KD-Tree-based proximal point selection algorithm; rejecting pseudo proximal points from the proximal point set by using a Ki strategy to obtain a cut proximal point set for constructing a training characteristic set of the minimax probability machine; regulating parameters of the minimax probability machine; and training a model to obtain a prediction value. The wireless network flow prediction method disclosed by the invention realizes the localization of the prediction model of the minimax probability machine and can carry out wireless network flow prediction more accurately and more timely.

The invention discloses a speech processing method. The method comprises the following steps: cutting non-speech part in the voice through end point detection to acquire a plurality of first speech fragments; performing Bayes information criterion BIC detection on the plurality of first speech fragments to acquire a speaker transition point; serving the speaker transition point as a break point tobreak the plurality of speech fragments, thereby acquiring a plurality of second speech fragments; extracting speech characteristics of the second speech signal fragments to form characteristic vector, classifying the second speech fragments; and correcting the category of the second speech fragments according to a preset keyword. Therefore, the problem that the algorithm processing effect is non-ideal for the telephone speech recording on the complex service scene by the existing speaker segmentation clustering algorithm can be improved, and an effect of accurately and quickly recognizing the speaker of the speech can be improved.

The invention relates to a data processing method for monitoring and prediction in the electrical power system, in particular to an electric load prediction method used for electric power based on factors of temperature and festivals and holidays. The electric load prediction method used for the electric power based on the factors of the temperature and the festivals and holidays comprises the steps of building a model by using historical data, distinguishing electricity utilization data in the festivals and holidays from electricity utilization data in workdays, building a judging dummy variable, building a short-term prediction model of the load by using the temperature and the dummy variable, carrying out a significance testing of independent variables of the model under the alike information criterion (AIC), and finally obtaining a short-term load prediction value by using the model. Due to the fact that the load prediction value is obtained according to the model and by considering the factors of the temperature and the festivals and holidays, prediction effect is good, and prediction precision is high.

The invention relates to a digitized personnel evaluation system based on a wireless network handheld terminal. The system is composed of a host, a wireless network device and a mobile terminal and iserected simply. A wireless network established by the wireless network device works cooperatively. With corresponding software installed in the host and the mobile terminal, standing inputting of evaluation information of the mobile terminal as well as system maintenance and data storage of the host is realized. The host completes plan designing of an evaluation item, a questionnaire compiling, evaluation issuing and management, and automatic statistic analysis and report generation of evaluation results; and on the basis of account management, real-time evaluation and anonymous evaluation and full-course recording of the operation by the maintenance staff and the like are realized. Personnel evaluation as well as instant analysis of the field report can be completed automatically, safely, accurately, objectively and efficiently with high confidentiality in a standard manner. Besides, the digitized personnel evaluation system also can be applied to scenes of exams, training, voting, research and the like and is suitable for fields like personnel evaluation, thereby providing a new application direction for the information technology and mobile network technology.

The invention relates to a shot clustering method based on the spectrum segmentation theory, which comprises the following steps: utilizing, the spectrum segmentation theory for shot clustering; extracting feature vectors of each unspecified shot; calculating similarity between each two categories according to the extracted feature vectors; then constituting each shot cluster as a weighted undirected graph; segmenting each shot category into two shot categories by a using spectrum according to the similarity between each two categories; using Bayesian information criteria to judge whether the segmentation is effective or not, the effectively segmented shot sub-categories are iteratively segmented, the ineffectively segmented shot categories are terminals; finally syncretizing the classification results after the segmentation to get the optimal shot classification number and the classification result. The invention solves the difficult problem that the optimized classification number is difficult to estimate in the clustering algorithm, and improves the recall ratio and the pertinency ratio of the clustering result by utilizing the precise classification spectrum segmentation; the proposed overall fusion operation has a function of correcting the classification errors, thereby effectively avoiding the problem of local optimum relation.