253 results about "Predictive modelling" patented technology

Predictive modeling of consumer financial behavior, including determination of likely responses to particular marketing efforts, is provided by application of consumer transaction data to predictive models associated with merchant segments. The merchant segments are derived from the consumer transaction data based on co-occurrences of merchants in sequences of transactions. Merchant vectors represent specific merchants, and are aligned in a vector space as a function of the degree to which the merchants co-occur more or less frequently than expected. Consumer vectors are developed within the vector space, to represent interests of particular consumers by virtue of relative vector positions of consumer and merchant vectors. Various techniques, including clustering, supervised segmentation, and nearest-neighbor analysis, are applied separately or in combination to generate improved predictions of consumer behavior.

Predictive modeling of consumer financial behavior, including determination of likely responses to particular marketing efforts, is provided by application of consumer transaction data to predictive models associated with merchant segments. The merchant segments are derived from the consumer transaction data based on co-occurrences of merchants in sequences of transactions. Merchant vectors represent specific merchants, and are aligned in a vector space as a function of the degree to which the merchants co-occur more or less frequently than expected. Supervised segmentation is applied to merchant vectors to form the merchant segments. Merchant segment predictive models provide predictions of spending in each merchant segment for any particular consumer, based on previous spending by the consumer. Consumer profiles describe summary statistics of each consumer's spending in the merchant segments, and across merchant segments. The consumer profiles include consumer vectors derived as summary vectors of selected merchants patronized by the consumer. Predictions of consumer behavior are made by applying nearest-neighbor analysis to consumer vectors, thus facilitating the targeting of promotional offers to consumers most likely to respond positively.

Various examples are provided for dynamically updating or adapting predictive modeling for prediction of outcomes of interest for operating systems and processes. Embodiments of the disclosure may provide systems, apparatus, processes, and methods for generating and deploying dynamically updated predictive models. In some embodiments, the predictive model may be deployed for the purpose of predicting operational outcomes of interests in operating systems, hardware devices, machines and / or processes associated therewith prior to the operational outcomes of interest occurring. The predictions can, for example, provide sufficient time for maintenance or repairs to be scheduled and carried out to avoid the predicted operational outcome. Autonomous evaluation of features allows the predictive models to be dynamically updated in response to changes in the environment or monitored data.

A predictive modeling method may include determining a time interval of time-series data; identifying one or more variables of the data as targets; determining a forecast range and a skip range associated with a prediction problem represented by the data; generating training data and testing data from the time-series data; fitting a predictive model to the training data; and testing the fitted model on the testing data. The forecast range may indicate a duration of a period for which values of the targets are to be predicted. The skip range may indicate a temporal lag between the time period corresponding to the data used to make predictions and the time period corresponding to the predictions. The skip range may separate input data subsets representing model inputs from subsets representing model outputs, and separate test data subsets representing model inputs from subsets representing validation data.

A system and method for creating virtual “flat” customer records derived from database customer data that may be used as standardized input for analytical models. A Customer Analytic Record (“CAR”) application may be created as a database object to extract, transform, and format all of the customer data needed for customer segmentation and predictive modeling. The CAR may be a set of database “views” that are defined using virtual stored queries and enabled using capabilities of a data base management system and a structured query language. The CAR is preferably a virtual “flat” record of the customer data needed for customer analytics. The customer data may be extracted by running one or more SQL queries against the database view(s). The CAR application may dynamically calculate additional variables using predetermined transformations, including custom transformations of an underlying behavior. If additional variables are created, the CAR may be modified to include the additional variables. The CAR is preferably a dynamic view of the customer record that changes whenever any update is made to the database. The definition of the CAR provides documentation of each data element available for use in models and analytics.

Transaction-based behavioral profiling, whereby the entity to be profiled is represented by a stream of transactions, is required in a variety of data mining and predictive modeling applications. An approach is described for assessing inconsistency in the activity of an entity, as a way of detecting fraud and abuse, using service-code information available on each transaction. Inconsistency is based on the concept that certain service-codes naturally co-occur more than do others. An assessment is made of activity consistency looking at the overall activity of an individual entity, as well as looking at the interaction of entities. Several approaches for measuring consistency are provided, including one inspired by latent semantic analysis as used in text analysis. While the description is in the context of fraud detection in healthcare, the techniques are relevant to application in other industries and for purposes other than fraud detection.

A prediction engine for use in predictive modeling associated with fantasy sports leagues is provided. The prediction engine can have a database module for storing criteria associated with players; and an information modeling module in communication with the database module for generating relationships among the criteria and providing player projections based at least in part on the relationships. The relationships can be adjusted based at least in part on changes to a scoring system of the fantasy sports league.

A computerized system and method for estimating insurance loss reserves and confidence intervals using insurance policy and claim level detail predictive modeling. Predictive models are applied to historical loss, premium and other insurer data, as well as external data, at the level of policy detail to predict ultimate losses and allocated loss adjustment expenses for a group of policies. From the aggregate of such ultimate losses, paid losses to date are subtracted to derive an estimate of loss reserves. Dynamic changes in a group of policies can be detected enabling evaluation of their impact on loss reserves. In addition, confidence intervals around the estimates can be estimated by sampling the policy-by-policy estimates of ultimate losses.

A computer system-based automated loss verification system for evaluating the validity of claims filed under an insurance policy or debt protection contract is provided by this invention. Instead of requiring the claimant to contact the insurance company or lender to file the claim and provide exhaustive documentary proof of the validity of the claimed loss, the system pre-scores the relative risk of the claim using a risk assessment tool based upon predictive modeling and a number of potential risk factors. The associated automated loss verification tool uses this risk score and other information connected with the claim to assign a relative confidence level of proof of valid loss that must be satisfied before the claim can be approved through the adjudication process, and assigns a third-party supplied source or combination of sources of proof that can be automatically accessed by the system to validate the claim.

The present invention provides a decision management system to define, validate and extract data for predictive models. A system of sensors is deployed in a sample collection environment, where such sensors are used to collect data from a biological or chemical sample, with additional sensors for ambient data whose output as a form of metadata can characterize performance conditions including background ambient conditions. A format or sequence of processes is the basis for a math model to establish a logical weight to data for predictive modeling and event reporting. The present invention provides a computer or other sensor interface system with a primary sensor or sensors, network connection, and supplementary sensors to measure the conditions in which the primary data is captured. A software process allows for user inputs of data in order to establish the methods and rules for normal function.

Transaction-based behavioral profiling, whereby the entity to be profiled is represented by a stream of transactions, is required in a variety of data mining and predictive modeling applications. An approach is described for assessing inconsistency in the activity of an entity, as a way of detecting fraud and abuse, using service-code information available on each transaction. Inconsistency is based on the concept that certain service-codes naturally co-occur more than do others. An assessment is made of activity consistency looking at the overall activity of an individual entity, as well as looking at the interaction of entities. Several approaches for measuring consistency are provided, including one inspired by latent semantic analysis as used in text analysis. While the description is in the context of fraud detection in healthcare, the techniques are relevant to application in other industries and for purposes other than fraud detection.

A system and method for improving predictive modeling of an information system is disclosed, including generating dynamic representations of the business solution through predictive modeling and providing automated calibration of a predictive model against predefined performance benchmarks.

Automated generation of a merchant recommendation list is disclosed. When a financial processor obtains rich and relevant information related to consumers and merchants, collaborative filtering, profiling, clustering and predictive modeling techniques are combined to provide merchant recommendations to a consumer. The system analyzes consumer attributes which relate to target consumer attributes to create a target consumer cluster, creates associations based upon merchant attributes and the target consumer attributes and provides the feedback based on the associations.

Certain embodiments provide methods and systems by which a predictive modeling provider may provide a client with predictive healthcare analysis. The predictive healthcare analysis may be used to assess whether a beneficiary is likely to develop conditions requiring healthcare. If such conditions are identified, intervention plans are developed and implemented to address the conditions and reduce healthcare costs. Certain embodiments provide method and systems by which an integrated benefits management organization may manage a client's healthcare program. A claim audit may be performed on healthcare claims of the client's beneficiaries. A claim repricing process may be used to discount the healthcare claims of the client's beneficiaries. In at least certain embodiments, the predictive modeling provider or integrated benefits management organization are motivated to maximize the reduction in the client's healthcare costs.

Methods and systems for performing comprehensive predictive analysis and modeling in a wager gaming environment are provided. Statistical calculations for deriving predictive values for a patron behavior in a casino or gaming environment may include various criteria, including but not limited to wager gaming activities, resort / hotel usage, dining / meals, non-gaming point of sale transactions, entertainment expenditures, coupon / prize redemption, “comps” provided, etc. In some implementations, a plurality of predictive models may be used. One or more criteria may be used to evaluate and / or rank the predictive models. For example, such an evaluation and / or ranking may be based, at least in part, on a correlation coefficient or a function thereof. In some such implementations, an evaluation and / or ranking may be based, at least in part, on a comparison of the coefficient of determination, R2, corresponding to each predictive model.

The present invention discloses a prediction model-building method for computer-aided medical diagnosis. It utilizes the medical symptom detection equipment to obtain the symptom of object to be diagnosed to form symptomatic vector, then utilizes the prediction model to make treatment so as to can obtain prediction result. It also provides its concrete steps for implementing said invented prediction model-building method.

A predictive modeling system and methodology makes predictions using unstructured content as an input, either alone or in conjunction with structured content. Content transformation rules are selected for application to the unstructured content, such as emails, call center notes, and other forms of human communication, by identifying the rules that are likely to improve the performance of a predictive modeling system.

A device and method designed to carry out the computation of a wide range of topological indices of molecular structure to produce molecular descriptors, representing important elements of the molecular structure information including but not limited to molecular structure variables such as; the molecular connectivity chi indices, mXt, and mXtv; kappa shape indices, mκ and mκα; electrotopological state indices, Si; hydrogen electrotopological state indices, HESi; atom type and bond type electrotopological state indices; new group type and bond type electrotopological state indices; topological equivalence indices and total topological index; several information indices, including the Shannon and the Bonchen Trinajstic information indices; counts of graph paths, atoms, atoms types, bond types; and others.

A computerized system for augmenting data from a source database with data from a reference database to generate an augmented database that can be used for predictive modeling is disclosed. The present invention includes a method for using the Internet to obtain information, including, reading a data record stored in a field of data, searching a database for information describing the data record, condensing the information describing the data record into a value description, associating the value description with the data record, and augmenting the field of data with the value description associated with the data. The present invention also includes a computerized system for augmenting data from a source database with data from a reference database to generate an augmented database that can be used for predictive modeling and data mining to conduct searches of data in the augmented database.

A method for establishing an agricultural pedigree for agricultural products comprises the steps of: (a) Providing an open communication network accessible information storage device adapted to receive input of data relating to agricultural product production and distribution from multiple sources; (b) Inputting said data into said information storage device; (c) Storing and said data; and (d) Providing access to said data via the open communication network, wherein the information storage device is configured to be used as at least: (i) A tool for traceability of the agricultural products, (ii) A real time decision making tool, and (iii) A predictive modeling tool.

A predictive modeling system and methodology makes predictions using unstructured content as an input, either alone or in conjunction with structured content. Content transformation rules are selected for application to the unstructured content, such as emails, call center notes, and other forms of human communication, by identifying the rules that are likely to improve the performance of a predictive modeling system.

Methods and apparatus for analyzing financial data generally includes a predictive modeling system. The predictive modeling system may include an artificial agent responsive to an input data set. The artificial agent may produce an estimated data set including a market conditions data set. The market conditions data may include an estimate of at least one of liquidity of a market, strategy of a counterparty, and an effect of information leakage. The artificial agent may determine a predictability value for the estimated data set. The predictive modeling system may also include an agent factory responsive to the input data set. The agent factory may generate an artificial agent in response to the input data set.

Methods and systems are disclosed for developing a software application through requirements specification, design, effort estimation, assigning and tracking work items, development, testing, collaboration and project management. The system is structured as an integrated software development system allowing for structured, systematic and user friendly set up of requirements specification (including workflow, user interface, business rules and integrations), technical design for UI features and integrations using design patterns, automated effort estimation, automated story card and task creation. The system also provides a novel, audio-video playback capability for workflow visualization. Further, the system provides visual depiction of defects and changed items side-by-side with requirements, as well as predictive modeling of defects that allows for forecasting of defect completion rate. In the preferred embodiment, the system is delivered as a multi-tenant Software as a Service.

The computational cost of many statistical modeling algorithms is affected by the input / output (I / O) cost of accessing out-of-core training data. This is an important challenge for emerging data mining applications, where the amount of training data can be potentially enormous. A heuristic approach to this problem is described. This approach is based on constructing a simple probability model from the large training data set, and using this model to generate simulated pseudo data for some aspects of the statistical modeling procedure. This approach is illustrated in the context of building a Naive Bayes probability model with feature selection. Here, the usual algorithms would require numerous data scans over the massive training data set, but our heuristic obtains models of comparable accuracy with just two data scans.

The invention discloses a combination forecast modeling method of wind farm power by using gray correlation analysis, belonging to the technical field of wind power generation modeling. In particular, the invention is related to a weighted combination forecast method of wind power based on a least square support vector machine and an error back propagation neural network. The forecast method comprises that forecasted values of wind speed and wind direction are acquired in advance from meteorological departments while real-time output power is acquired from a wind farm data acquiring system; that the forecasted values of wind speed and wind direction and the real-time output power are inputted into a data processing module for data analyzing extraction and data normalization, and then normalized data is loaded to a database server; processed data in the database server is extracted by a combination forecast algorithm server to carry out model training and power forecast, and the wind farm sends running data to the data processing module in real time to realize rolling forecasting. The method of the invention achieves the goal of combination forecast of wind farm output in a short time. The method not only maximally utilizes advantages of two algorithms but also increases forecast efficiency by saving computing resources and shortening computing time.

The system and method for predictive modeling in a network security service described herein may provide a scalable architecture that can model information relating to any specific threat or potential threat in a network and manage routing requests relating to the threat information among various entities participating in the security service. In particular, the scalable architecture may include various distributed databases that store serialized information describing threat instances, wherein a detection service may maintain information identifying entities associated with the databases storing the serialized information. Further, the security service may include a hierarchical subscriber name service that participating entities can traverse to locate the serialized threat information in the various databases and evaluate how the threat instances may have evolved or progressed through the network.

A method and system for administering a disease management program to improve healthcare quality, reduce healthcare costs, and optimize delivery of healthcare services. A multi-condition risk assessment is conducted for all or a substantial portion of a population of program participants, and collected multi-condition risk assessment data are combined with claims data for predictive modeling of future healthcare risk and expense. Participants are risk-stratified into one or more classifications of future healthcare cost risk, and appropriate intervention or delivery of healthcare services is made based on the risk classification.

An apparatus, system and method for predictive modeling to design, evaluate and optimize ophthalmic lenses is disclosed. Ophthalmic lenses may include, for example, contacts, glasses or intraocular lenses (IOLs). The apparatus, system and method may include a design tool for designing a lens for implantation in an eye having a plurality of characteristics, a simulator for simulating performance of the lens in at least one modeled eye having the plurality of characteristics, at least one input for receiving clinical performance of the lens in the eye having the plurality of characteristics, a comparator for comparing outcomes of the clinical performance and the simulated performance, and an optimizer for optimizing a subsequent one of the outcome of the clinical performance responsive to modification of the lens in accordance with modification to the simulated performance.

A system, method and computer-readable medium provides steps of merging one or more sources of IP addresses to create a merged database comprising a plurality of IP zones, user typing each of the plurality of IP zones, enhancing each of the plurality of IP zones, receiving an input address and selecting one or more of the plurality of IP zones having a predetermined statistical relationship to the input address.