2145 results about "Statistical model" patented technology

A method and system for detecting anomalous action within a computer network is provided herein. The method starts with collecting raw data from at least one probe sensor that is associated with at least one router, switch or at least one server which are part of the computer network. Next, the raw data is being parsed and analyzed and meta-data is created from the raw data. Computer network actions are being identified based on existing knowledge about network protocols. The meta-data is associated with entities by analyzing the identified network actions and correlating between different computer network actions. Finally, creating at least one statistical model of the respective computer network said model including network actions' behavior pattern and online or batch detection of anomalous network actions associated with entities based on the statistical models.

A computer-based method and system accomplish automatic optimization of a preferred merchandising figure of merit like revenue, profit, etc. by concurrent optimization of prices and in-store promotion schedules for groups of interrelated products in a supermarket or in a chain of supermarkets. The basis of the system is a statistical model that combines effects of prices, promotion schedules and other factors that could influence demands, and produces a flexible nonparametric predictive demand function that can be optimized simultaneously in prices and promotion schedules. The system generates demand predictions for groups of interrelated products by applying data mining procedures to historical database that contains sales data along with various sales conditions, adjusting statistical predictive demand models, and then optimizing them in prices and promotion schedules. The user can interact with the system by selecting and customizing optimization and prediction scenarios provided by the system, and by requesting customized reports with results of particular optimization and prediction schemes.

A computerized method and system for evaluating customers of a financial institution using customer relationship value tags and associated customer treatment actions includes automatically analyzing information about a customer from a database of the financial institution by a customer assessment engine using a predefined statistical model to assess the value of the customer to the financial institution. At least one customer treatment action associated with the assessed value is identified by the assessment engine, and the assessment engine marks a file associated with the customer with a mark representing the assessed value and the associated customer treatment action. The marked file is then accessed by other financial systems of the financial institution, or by customer representatives of the financial institution, and the associated customer treatment action is implemented by such systems or representatives in dealing with the customer.

An apparatus and methods advantageously select content items for dynamically-generated web pages in an intelligent and virtually autonomous manner. This permits the operator of the web site to rapidly identify and respond to trends, thereby advantageously updating the web site relatively quickly and efficiently without or with less time consuming and expensive manual labor. User interaction for a plurality of users with the web site is collected in a database. For various content items, the database is mined to extract relationships between probability and references of select attributes in probability models. When a new web page is requested, attributes, which can include attributes associated with a user, are used as references to the applicable probability models of selected content items, combined with value weighting to generate expected values, and selected for use in the web page at least partially based on the expected values.

Transaction processing of online transactions at merchant sites determines the likelihood that such transactions are fraudulent, accounting for unreliable fields of a transaction order, which fields do not reliably identify a purchaser. A scoring server using statistical model uses multiple profiles associated with key fields, along with weights to indicate the degree to which the profiles identify the purchaser of the transaction.

The information management system disclosed enables caregivers to make better decisions, faster, using aggregated genetic and phenotypic data. The system enables the integration, validation and analysis of genetic, phenotypic and clinical data from multiple subjects who may be at distributed facilities. A standardized data model stores a range of patient data in standardized data classes that encompass patient profile information, patient symptomatic information, patient treatment information, and patient diagnostic information including genetic information. Data from other systems is converted into the format of the standardized data classes using a data parser, or cartridge, specifically tailored to the source system. Relationships exist between standardized data classes that are based on expert rules and statistical models. The relationships are used both to validate new data, and to predict phenotypic outcomes based on available data. The prediction may relate to a clinical outcome in response to a proposed intervention by a caregiver. The statistical models may be inhaled into the system from electronic publications that define statistical models and methods for training those models, according to a standardized template. Methods are described for selecting, creating and training the statistical models to operate on genetic, phenotypic and clinical data, in particular for underdetermined data sets that are typical of genetic information. The disclosure also describes how security of the data is maintained by means of a robust security architecture, and robust user authentication such as biometric authentication, combined with application-level and data-level access privileges.

A “Smart Grid Pricer” enables automated balancing of the supply and demand of energy supply and consumption, such as the generation and consumption of electricity between electricity providers and electricity consumers. The Smart Grid Pricer automatically computes and delivers real-time energy pricing information to consumers on behalf of energy retailers (e.g., electricity utilities) to help drive the balance of demand with supply. In various embodiments, real-time pricing is determined by using various probabilistic models to estimate overall consumer demand as a function of factors such as energy price, time of day, region, weather, etc. to compute a price that will result in an energy demand that is closely balanced to the available supply. On the consumer side, various components of the Smart Grid Pricer automatically respond to such pricing information to optimize energy consumption in accordance with a variety of automated and / or user defined rules and preferences.

Exemplary embodiments of the present invention provide a method of health insurance market risk assessment including receiving first data including demographic and cost data for members of a health insurance plan in a current market, receiving second data including demographic data for the current market, and receiving third data including demographic data for a new market. The first to third data are used to transform a distribution of the plan members to account for differences between the current and new market demographic data and to estimate probabilities of enrollment in the new market. A statistical model is learned to predict risk in the new market using the transformed distribution and the estimated probabilities. The statistical model is used to determine risk of entering the new market.

A system for analyzing driving performance data is provided. The system and method include one or more devices in electronic communication with a network, the one or more devices including one or more sensors for obtaining raw data associated with operation of a vehicle by a driver. A driving performance engine in electronic communication with the device, the driving performance engine generates one or more granular driving events by defining values of one or more driving parameters by associating the one or more driving parameters to the raw data, compares the one or more granular driving events with one or more similar previous driving events of the driver or other drivers having similar driving parameters and values thereof, normalizes the one or more granular driving events of the driver based on the comparison, and processes the one or more granular driving events using the driving performance engine and one or more statistical models to calculate a performance or risk value for the driver.

A predictive model, for example, a neural network, evaluates individual debt holder accounts and predicts the amount that will be collected on each account based on learned relationships among known variables. The predictive model is generated using historical data of delinquent debt accounts, the collection methods used to collect the debts in the accounts, and the success of the collection methods. In one embodiment, the predictive model is generated using profiles of delinquent debt accounts summarizing patterns of events in the accounts, and the success of the collection effort in each account. In another embodiment, the predictive model includes a mathematical representation of the collector's notes created during the collection period for each account.

A natural language understanding system is described which provides for the generation of concept codes from free-text medical data. A probabilistic model of lexical semantics, in the preferred embodiment of the invention implemented by means of a Bayesian network, is used to determine the most probable concept or meaning associated with a sentence or phrase. The inventive method and system includes the steps of checking for synonyms, checking spelling, performing syntactic parsing, transforming text to its "deep" or semantic form, and performing a semantic analysis based on a probabilistic model of lexical semantics. In the preferred embodiment of the invention, spell checking and transformational processing as well as semantic analysis make use of semantic probabilistic determinations.

The present invention provides a statistical modeling approach to automatic linguistic indexing of photographic images. The invention uses categorized images to train a dictionary of hundreds of statistical models each representing a concept. Images of any given concept are regarded as instances of a stochastic process that characterizes the concept. To measure the extent of association between an image and a textual description associated with a predefined concept, the likelihood of the occurrence of the image based on the characterizing stochastic process is computed. A high likelihood indicates a strong association between the textual description and the image. The invention utilizes two-dimensional multi-resolution hidden Markov models that demonstrate accuracy and high potential in linguistic indexing of photographic images.

Methods, apparatuses, and systems directed to cost-aware bandwidth management schemes that are adaptive to monitored network or application performance attributes. In one embodiment, the present invention supports bandwidth management systems that adapt to network conditions, while managing tradeoffs between bandwidth costs and application performance. One implementation of the present invention tracks bandwidth usage over an applicable billing period and applies a statistical model to allow for bursting to address increased network loading conditions that degrade network or application performance. One implementation allows for bursting at selected time periods based on computations minimizing cost relative to an applicable billing model. One implementation of the present invention is also application-aware, monitoring network application performance and increasing bandwidth allocations in response to degradations in the performance of selected applications.

Computer implemented methods and systems of processing transactions to determine the risk of transaction convert high categorical information, such as text data, to low categorical information, such as category or cluster IDs. The text data may be merchant names or other textual content of the transactions, or data related to a consumer, or any other type of entity which engages in the transaction. Content mining techniques are used to provide the conversion from high to low categorical information. In operation, the resulting low categorical information is input, along with other data, into a statistical model. The statistical model provides an output of the level of risk in the transaction. Methods of converting the high categorical information to low categorical clusters, of using such information, and other aspects of the use of such clusters are disclosed.

A method for estimating a receiver's location in a wireless communication environment having several channels. Each channel has at least one signal parameter that varies with location differently from the other channels. A set of calibration data is determined for each calibration point, each set including the location and at least one measured signal parameter for each of several channels. The calibration data serve as a basis for a statistical model of the signal parameters versus a receiver's location. A set of observed signal parameters is determined, the set including at least one signal parameter for each of several channels at the receiver's location. A location estimate approximating the location of the receiver is determined on the basis of the statistical model and the set of observed signal parameters.

The system implements a novel method for personalized filtering of information and automated generation of user-specific recommendations. The system uses a statistical latent class model, also known as Probabilistic Latent Semantic Analysis, to integrate data including textual and other content descriptions of items to be searched, user profiles, demographic information, query logs of previous searches, and explicit user ratings of items. The system learns one or more statistical models based on available data. The learning may be reiterated once additional data is available. The statistical model, once learned, is utilized in various ways: to make predictions about item relevance and user preferences on un-rated items, to generate recommendation lists of items, to generate personalized search result lists, to disambiguate a users query, to refine a search, to compute similarities between items or users, and for data mining purposes such as identifying user communities.

A system provides one or more advertisements to users in response to search queries and logs user behavior associated with user selection of the one or more advertisements. The system also logs features associated with selected ones of the one or more advertisements, or associated with the search queries. The system further uses a statistical model and the logged user behavior to estimate quality scores associated with the selected advertisements and aggregates the estimated quality scores. The system predicts the quality of another advertisement using the aggregated quality scores.

At least certain embodiments of the present disclosure include a method to identify top hits in search result based on learned user preferences. In one embodiment, an editor receives a user selection of a font to be used in a document when the user is composing the document using the editor. The editor invokes a font recommendation engine. Based on the font selected, the font recommendation engine automatically recommends a set of one or more fonts to the user according to a statistical model of font usage.

Systems and methods are disclosed for generating statistical models. Such systems and methods may utilize a database comprising data representing a plurality of variables. To generate a statistical model, a set of variables may be selected in accordance with a goal of the model. Using the database, the selected set of variables may then be applied to a plurality of statistical model types and the results from each statistical model type may be analyzed. Finally, at least one of statistical model may be identified based on the analysis of the results.

An Arabic spelling error detection and correction method for identifying real word spelling errors. The method uses a corpus of Arabic text alongside n-gram statistical techniques to detect erroneous words within the text. After identifying the erroneous word the method uses a dictionary formed from the corpus of Arabic text to retrieve candidate correction word to replace the erroneous word with. Using n-gram statistical models candidate correction words are generated and ranked in order of highest probable correction for the word. The generated and ranked correction words are assessed and the best correction word is selected. A final assessment of the correction is conducted and if the result is positive then erroneous word is replaced with the highest statistical correction.

A plurality of banks of first deposit provide checking account activity data for both transit items (checks received for deposit that need to be cleared) and incoming returns (bounced checks) to a statistical model which determines from the data the likelihood that a check from a specific checking account will be returned. This data is used to populate a database of checking accounts to be used for making check risk decisions, such as check hold policy decisions, check acceptance decisions, and open to buy decisions.

Methods for determining statistical models for predicting disease risks of a population are provided. Two types of data associated with members of the population are collected. The data may include both genetic and non-genetic types of data. A candidate statistical model is selected for calculating the disease risk. The model has a plurality of parameters and is a function of only one of the two types of data. A data weight is determined for each member of the population. Members having like data of the other type have like weights. The parameters of the model are optimized by fitting the collected data to the model taking into account of the weights.

According to an embodiment, an apparatus and method are disclosed for dynamic gesture recognition from stereo sequences. In an embodiment, a stereo sequence of images of a subject is obtained and a depth disparity map is generated from the stereo sequence. The system is initiated automatically based upon a statistical model of the upper body of the subject. The upper body of the subject is modeled as three planes, representing the torso and arms of the subject, and three Gaussian components, representing the head and hands of the subject. The system tracks the upper body of the subject using the statistical upper body model and extracts three-dimensional features of the gestures performed. The system recognizes the gestures using recognition units, which, under a particular embodiment, utilizes hidden Markov models for the three-dimensional gestures.

The present invention is a method for detecting an abnormal event for process units of a Fluidized Catalytic Cracking Unit. The method compares the operation of the process units to a statistical and engineering models. The statistical models are developed by principle components analysis of the normal operation for these units. In addition, the engineering models are based on partial least squares analysis and correlation analysis between variables. If the difference between the operation of a process unit and the normal model result indicates an abnormal condition, then the cause of the abnormal condition is determined and corrected.