Abstract: The subject invention leverages standard probabilistic inference techniques to determine a log-likelihood for a conditional Gaussian graphical model of a data set with at least one continuous variable and with data not observed for at least one of the variables. This provides an efficient means to compute gradients for CG models with continuous variables and incomplete data observations. The subject invention allows gradient-based optimization processes to employ gradients to iteratively adapt parameters of models in order to improve incomplete data log-likelihoods and identify maximum likelihood estimates (MLE) and/or local maxima of the incomplete data log-likelihoods. Conditional Gaussian local gradients along with conditional multinomial local gradients determined by the subject invention can be utilized to facilitate in providing parameter gradients for full conditional Gaussian models.

Abstract: Systems and methods that create a classification of sentences in a language, and further construct associated local versions of language models, based on geographical location and/or other demographic criteria—wherein such local language models can be of different levels of granularity according to chosen demographic criteria. The subject innovation employs a classification encoder component that forms a classification (e.g. a tree structure) of sentences, and a local language models encoder component, which employs the classification of sentences in order to construct the localized language models. A decoder component can subsequently enable local word wheeling and/or local web search by blending k-best answers from local language models of varying demographic granularity that match users demographics. Hence, k-best matches for input data by users in one demographic locality can be different from k-best matches for the same input by other users in another locality.

Abstract: A method of inputting text is provided in which a first portion of an input string is received from a user, the first portion of the input string including at least one keystroke representing a wildcard character of the input string. A second portion of the input string is then received, with the second portion including one or more keystrokes all representing non-wildcard characters of the input string.

Abstract: The claimed subject matter provides systems and/or methods that expand input data. An interface can obtain input data and a wildcard insertion component can modify the input data to include at least one implicit wildcard inserted at an end of each intended word. Additionally, an expansion component can generate a candidate list of expanded data based at least in part on the input data including the at least one implicit wildcard utilizing a language model that provides likely expansions of wildcards. Further, the expansion component can evaluate the input data at a server side.

Abstract: The claimed subject matter provides systems and/or methods that expand input data. An interface can obtain input data and a wildcard insertion component can modify the input data to include at least one implicit wildcard inserted at an end of each intended word. Additionally, an expansion component can generate a candidate list of expanded data based at least in part on the input data including the at least one implicit wildcard utilizing a language model that provides likely expansions of wildcards. Further, the expansion component can evaluate the input data at a server side.

Abstract: Determining the near-optimal block size for incremental-type expectation maximization (EM) algorithms is disclosed. Block size is determined based on the novel insight that the speed increase resulting from using an incremental-type EM algorithm as opposed to the standard EM algorithm is roughly the same for a given range of block sizes. Furthermore, this block size can be determined by an initial version of the EM algorithm that does not reach convergence. For a current block size, the speed increase is determined, and if the speed increase is the greatest determined so far, the current block size is set as the target block size. This process is repeated for new block sizes, until no new block sizes can be determined.

Abstract: System(s) and method(s) that facilitate utilizing pre-cached disk space. Pre-cached memory space within a storage device is identified, and a subset of the pre-cached memory space is pre-populated with data so that the data can be selectively and dynamically accessed. During use of a computer (e.g., in a web-browsing session) a subset of the pre-stored data can be dynamically and selectively exposed to the user as a function of user and/or computer application state. Pre-storage of the data on pre-cached memory of the computer mitigates delayed data access (e.g., due to insufficient transmission bandwidth) thereby enhancing user computing experience. The user can utilize the device without having to distinguish between pre-cached and free memory. In other words, the operating system can present the cached memory to the user so that it appears as free memory without the user having to direct the system to do so.

Abstract: The present invention utilizes generic and user-specific features of handwriting samples to provide adaptive handwriting recognition with a minimum level of user-specific enrollment data. By allowing generic and user-specific classifiers to facilitate in a recognition process, the features of a specific user's handwriting can be exploited to quickly ascertain characteristics of handwriting characters not yet entered by the user. Thus, new characters can be recognized without requiring a user to first enter that character as enrollment or “training” data. In one instance of the present invention, processing of generic features is accomplished by a generic classifier trained on multiple users. In another instance of the present invention, a user-specific classifier is employed to modify a generic classifier's classification as required to provide user-specific handwriting recognition.

Abstract: 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.

Abstract: The invention performs handwriting recognition using mixtures of Bayesian networks. A mixture of Bayesian networks (MBN) consists of plural hypothesis-specific Bayesian networks (HSBNs) having possibly hidden and observed variables. A common external hidden variable is associated with the MBN, but is not included in any of the HSBNs. Each HSBN models the world under the hypothesis that the common external hidden variable is in a corresponding one of its states. The MBNs encode the probabilities of observing the sets of visual observations corresponding to a handwritten character. Each of the HSBNs encodes the probabilities of observing the sets of visual observations corresponding to a handwritten character and given a hidden common variable being in a particular state.

Abstract: The present invention utilizes generic and user-specific features of handwriting samples to provide adaptive handwriting recognition with a minimum level of user-specific enrollment data. By allowing generic and user-specific classifiers to facilitate in a recognition process, the features of a specific user's handwriting can be exploited to quickly ascertain characteristics of handwriting characters not yet entered by the user. Thus, new characters can be recognized without requiring a user to first enter that character as enrollment or “training” data. In one instance of the present invention, processing of generic features is accomplished by a generic classifier trained on multiple users. In another instance of the present invention, a user-specific classifier is employed to modify a generic classifier's classification as required to provide user-specific handwriting recognition.

Abstract: The present invention leverages curve fitting data techniques to provide automatic detection of data anomalies in a “data tube” from a data perspective, allowing, for example, detection of data anomalies such as on-screen, drill down, and drill across data anomalies in, for example, pivot tables and/or OLAP cubes. It determines if data substantially deviates from a predicted value established by a curve fitting process such as, for example, a piece-wise linear function applied to the data tube. A threshold value can also be employed by the present invention to facilitate in determining a degree of deviation necessary before a data value is considered anomalous. The threshold value can be supplied dynamically and/or statically by a system and/or a user via a user interface. Additionally, the present invention provides an indication to a user of the type and location of a detected anomaly from a top level data perspective.

Abstract: A method of compressing a language model is provided. A list of numerical values is generated from elements of a user input. The list of values is sorted. Differences between adjacent integer values in the list are calculated. Each calculated difference is encoded using a Golomb code.

Abstract: A method of inputting text is provided in which a first portion of an input string is received from a user, the first portion of the input string including at least one keystroke representing a wildcard character of the input string. A second portion of the input string is then received, with the second portion including one or more keystrokes all representing non-wildcard characters of the input string.

Abstract: A system and method for generating staged mixture model(s) is provided. The staged mixture model includes a plurality of mixture components each having an associated mixture weight, and, an added mixture component having an initial structure, parameters and associated mixture weight. The added mixture component is modified based, at least in part, upon a case that is undesirably addressed by the plurality of mixture components using a structural expectation maximization (SEM) algorithm to modify at the structure, parameters and/or associated mixture weight of the added mixture component. The staged mixture model employs a data-driven staged mixture modeling technique, for example, for building density, regression, and classification model(s). The basic approach is to add mixture component(s) (e.g., sequentially) to the staged mixture model using an SEM algorithm.

Abstract: The present invention leverages curve fitting data techniques to provide automatic detection of data anomalies in a “data tube” from a data perspective, allowing, for example, detection of data anomalies such as on-screen, drill down, and drill across data anomalies in, for example, pivot tables and/or OLAP cubes. It determines if data substantially deviates from a predicted value established by a curve fitting process such as, for example, a piece-wise linear function applied to the data tube. A threshold value can also be employed by the present invention to facilitate in determining a degree of deviation necessary before a data value is considered anomalous. The threshold value can be supplied dynamically and/or statically by a system and/or a user via a user interface. Additionally, the present invention provides an indication to a user of the type and location of a detected anomaly from a top level data perspective.

Abstract: The subject invention leverages the conditional Gaussian (CG) nature of a continuous variable stochastic ARMAxp time series model to efficiently determine its parametric gradients. The determined gradients permit an easy means to construct a parametric structure for the time series model. This provides a gradient-based alternative to the expectation maximization (EM) process for learning parameters of the stochastic ARMAxp time series model. Thus, gradients for parameters can be computed and utilized with a gradient-based learning method for estimating the parameters. This allows values of continuous observations in a time series to be predicted utilizing the stochastic ARMAxp time series model, providing efficient and accurate predictions.

Abstract: The present invention leverages curve fitting data techniques to provide automatic detection of data anomalies in a “data tube” from a data perspective, allowing, for example, detection of data anomalies such as on-screen, drill down, and drill across data anomalies in, for example, pivot tables and/or OLAP cubes. It determines if data substantially deviates from a predicted value established by a curve fitting process such as, for example, a piece-wise linear function applied to the data tube. A threshold value can also be employed by the present invention to facilitate in determining a degree of deviation necessary before a data value is considered anomalous. The threshold value can be supplied dynamically and/or statically by a system and/or a user via a user interface. Additionally, the present invention provides an indication to a user of the type and location of a detected anomaly from a top level data perspective.

Abstract: The invention performs handwriting recognition using mixtures of Bayesian networks. A mixture of Bayesian networks (MBN) consists of plural hypothesis-specific Bayesian networks (HSBNs) having possibly hidden and observed variables. A common external hidden variable is associated with the MBN, but is not included in any of the HSBNs. Each HSBN models the world under the hypothesis that the common external hidden variable is in a corresponding one of its states. The MBNs encode the probabilities of observing the sets of visual observations corresponding to a handwritten character. Each of the HSBNs encodes the probabilities of observing the sets of visual observations corresponding to a handwritten character and given a hidden common variable being in a particular state.

Abstract: A technique for reducing a model database for use with handwriting recognizers. The model database is processed with a tuning set to identify a set of models that would result in the greatest character recognition accuracy. If further model database reduction is desired, the technique iteratively identifies smaller models that have the least adverse effect on the error rate. The technique continues identifying smaller models until a desired model database size has been achieved.