Abstract: Real-time query expansion (RTQE) is a process of supplementing an original query with addition terms or expansion choices that are ranked according to some figure of merit and presented while users are still formulating their queries. As disclosed herein, phrases may be presented and one or more terms of a focused-on phrase may be pinned (as desirable to the user). Subsequent lists may be presented as a function of pinned terms and/or user input. In one embodiment, a placeholder may be substituted for one or more pinned terms if less than some predetermined threshold of phrases is able to be presented based upon the pinned terms and/or user input, and another list of phrases may be presented as a function of a query using fewer than all the pinned terms. The placeholder may allow out-of-index phrases to be formed, for example, based upon two or more phrases and/or terms input by the user.

Abstract: Methods and systems are disclosed for learning a regression decision graph model using a Bayesian model selection approach. In a disclosed aspect, the model structure and/or model parameters can be learned using a greedy search algorithm applied to grow the model so long as the model improves. This approach enables construction of a decision graph having a model structure that includes a plurality of leaves, at least one of which includes a non-trivial linear regression. The resulting model thus can be employed for forecasting, such as for time series data, which can include single or multi-step forecasting.

Abstract: Real-time query expansion (RTQE) is a process of supplementing an original query with additional terms or expansion choices that are ranked according to some figure of merit and presented while users are still formulating their queries. As disclosed herein, individual terms may be combined and submitted as a phrase into a query. By building the phase term-by-term, users can compositionally formulate queries while maintaining the same benefits that other RTQE interfaces offer. To promote greater flexibility in its working environment, the number of terms that are presented on a display may be reduced. In place of some terms, placeholders may be used and expanded by the user when necessary. This allows phrases to be readily presented on small displays (e.g., hand-held devices).

Abstract: Real-time query expansion (RTQE) is a process of supplementing an original query with additional terms or expansion choices that are ranked according to some figure of merit and presented while users are still formulating their queries. As disclosed herein, individual terms may be combined and submitted as a phrase into a query. By building the phase term-by-term, users can compositionally formulate queries while maintaining the same benefits that other RTQE interfaces offer. The benefits include, reducing the number of keystrokes and improving retrieval performance. To promote greater flexibility in its working environment, the number of terms that are presented on a display may be reduced. In place of some terms, placeholders may be used and expanded by the user when necessary. This allows phrases to be readily presented on small displays (e.g., hand-held devices).

Abstract: A multi-modal search query refinement system (and corresponding methodology) is provided. In accordance with the innovation, query suggestion results represent a word palette which can be used to select strings for inclusion or exclusion from a refined set of results. The system employs text, speech, touch and gesture input to refine a set of search query results. Wildcards can be employed in the search either prompted by the user or inferred by the system. Additionally, partial knowledge supplemented by speech can be employed to refine search results.

Abstract: A multi-modal search system (and corresponding methodology) that employs wildcards is provided. Wildcards can be employed in the search query either initiated by the user or inferred by the system. These wildcards can represent uncertainty conveyed by a user in a multi-modal search query input. In examples, the words “something” or “whatchamacallit” can be used to convey uncertainty and partial knowledge about portions of the query and to dynamically trigger wildcard generation.

Abstract: A multi-modal search system (and corresponding methodology) is provided. The system employs text, speech, touch and gesture input to establish a search query. Additionally, a subset of the modalities can be used to obtain search results based upon exact or approximate matches to a search result. For example, wildcards, which can either be triggered by the user or inferred by the system, can be employed in the search.

Abstract: Systems and methods are provided for parsimonious representation of large sets of multi-resolution value-item lists. A hierarchical data structure associated with the lists and conditioning variables is learnt while exploiting both semantics encoded in target variables and a notion of nearness among nodes at the same detail level in the hierarchical data structure. Such a level of description can be dictated by a depth in a tree data structure. A compression scheme that relies on (i) a similarity metric and (ii) recursive greedy pairing of value-item lists in order to promote elements populating a specific tree node upwards in the tree facilitates a parsimonious representation of the compressed hierarchic structure.

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: The present invention utilizes a cross-prediction scheme to predict values of discrete and continuous time observation data, wherein conditional variance of each continuous time tube variable is fixed to a small positive value. By allowing cross-predictions in an ARMA based model, values of continuous and discrete observations in a time series are accurately predicted. The present invention accomplishes this by extending an ARMA model such that a first time series “tube” is utilized to facilitate or “cross-predict” values in a second time series tube to form an “ARMAxp” model. In general, in the ARMAxp model, the distribution of each continuous variable is a decision graph having splits only on discrete variables and having linear regressions with continuous regressors at all leaves, and the distribution of each discrete variable is a decision graph having splits only on discrete variables and having additional distributions at all leaves.

Abstract: The present invention utilizes a discriminative density model selection method to provide an optimized density model subset employable in constructing a classifier. By allowing multiple alternative density models to be considered for each class in a multi-class classification system and then developing an optimal configuration comprised of a single density model for each class, the classifier can be tuned to exhibit a desired characteristic such as, for example, high classification accuracy, low cost, and/or a balance of both. In one instance of the present invention, error graph, junction tree, and min-sum propagation algorithms are utilized to obtain an optimization from discriminatively selected density models.

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: 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 claimed subject matter provides for character selection based on two orthogonal directional inputs associated with two input controllers. Such character selection can include a dual-input selection component comprising two independent input directional controllers, wherein activating a directional input effectuates a selection. Further included is a grouping component that can group characters into sub-groups, and map sub-groups and characters to orthogonal directional inputs related to the two input controllers, whereby activating a direction input can select a particular sub-group or character mapped to that direction.

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 relates to a system and method to facilitate data mining applications and automated evaluation of models for continuous variable data. In one aspect, a system is provided that facilitates decision tree learning. The system includes a learning component that generates non-standardized data that relates to a split in a decision tree and a scoring component that scores the split as if the non-standardized data at a subset of leaves of the decision tree had been shifted and/or scaled. A modification component can also be provided for a respective candidate split score on the decision tree, wherein the above data or data subset can be modified by shifting and/or scaling the data and a new score is computed on the modified data. Furthermore, an optimization component can be provided that analyzes the data and determines whether to treat the data as if it was: (1) shifted, (2) scaled, or (3) shifted and scaled.

Abstract: A model is constructed for an initial subset of the data using a first parameter estimation algorithm. The model may be evaluated, for example, by applying the model to a holdout data set of the data. If the model is not acceptable, additional data is added to the data subset and the first parameter estimation algorithm is repeated for the aggregate data subset. An appropriate subset of the data exists when the first parameter estimation algorithm produces an acceptable model. The appropriate subset of the data may then be employed by a second parameter estimation algorithm, which may be a more accurate version of the first algorithm or a different algorithm altogether, to build a statistical model to characterize the data.

Abstract: Mean shift is a nonparametric estimator of density which has been applied to image and video segmentation. Traditional mean shift based segmentation uses a radially symmetric kernel to estimate local density, which is not optimal in view of the often structured nature of image and more particularly video data. The system and method of the invention employs an anisotropic kernel mean shift in which the shape, scale, and orientation of the kernels adapt to the local structure of the image or video. The anisotropic kernel is decomposed to provide handles for modifying the segmentation based on simple heuristics. Experimental results show that the anisotropic kernel mean shift outperforms the original mean shift on image and video segmentation in the following aspects: 1) it gets better results on general images and video in a smoothness sense; 2) the segmented results are more consistent with human visual saliency; and 3) the system and method is robust to initial parameters.

Abstract: Match criteria are provided to specify when advertisements will be shown, for instance in a search environment. Input such as search queries can be represented in a simplified form such as an implicit and/or explicit wildcard expression. Advertisers or other entities can bid on terms such that advertisements or similar content are presented when the terms match an expansion of a simplified input. Matching ads can subsequently be displayed alone or in combination with query expansion suggestions and/or query results.

Abstract: A list of integer values is generated from n-grams of a user input. The list of integer values is sorted. Differences between adjacent integer values in the list are calculated. Each calculated difference is encoded using a Golomb code. A Golomb compressed language model is accessed to identify likely matches.