Abstract: The present invention relates to queue-theoretic models for integration of automated call routing systems with human operators. Organizations are increasingly turning to spoken dialog systems for automated call routing to reduce call center costs. To maintain quality service even in cases of failure, these systems often resort to ad-hoc rules for dispatching calls to a human operator. The present invention provides queue-theoretic methods that provide a modeling and simulation capability in support of decisions about the staffing of call-handling centers based on the frequency of incoming calls and the competency of automated dialog systems. The methods include a procedure for identifying when callers should be transferred to operators. The procedure integrates models that predict when a call is likely to fail using spoken dialog features with queuing models of call center volume and service time.

Abstract: The present invention relates to a system and method to facilitate providing computer users with efficient access to electronic notifications while minimizing disruption in the context of ongoing tasks. A notification system provides visual information heralds that provide summarizations of notifications to users. An information controller considers the importance or urgency of content and a user's current and/or recent activities in decisions about the rendering and positioning of the information herald on a display screen or screens in order to facilitate user processing of the summarized notifications with minimal disruption to ongoing tasks.

Abstract: Aspects relate to detecting gestures that relate to a desired action, wherein the detected gestures are common across users and/or devices within a surface computing environment. Inferred intentions and goals based on context, history, affordances, and objects are employed to interpret gestures. Where there is uncertainty in intention of the gestures for a single device or across multiple devices, independent or coordinated communication of uncertainty or engagement of users through signaling and/or information gathering can occur.

Abstract: The present invention relates to a system and methodology to facilitate extraction of information from a large unstructured corpora such as from the World Wide Web and/or other unstructured sources. Information in the form of answers to questions can be automatically composed from such sources via probabilistic models and cost-benefit analyses to guide resource-intensive information-extraction procedures employed by a knowledge-based question answering system. The analyses can leverage predictions of the ultimate quality of answers generated by the system provided by Bayesian or other statistical models. Such predictions, when coupled with a utility model can provide the system with the ability to make decisions about the number of queries issued to a search engine (or engines), given the cost of queries and the expected value of query results in refining an ultimate answer. Given a preference model, information extraction actions can be taken with the highest expected utility.

Abstract: The present invention leverages changes in the sensed strength of radio signals at different locations to determine a device's location. In one instance of the present invention, inference procedures are used to process ambient commercial radio signals, to estimate a location or a probability distribution over the locations of a device. In another instance of the present invention, a system utilizes learning and inference methods that are applied to rank vector of signal strength vectors. Moving to such rank orderings leads to systems that bypass consideration of absolute signal strengths in location calculations. The present invention facilitates approximations for locating a device by providing a system that does not require a substantial number of available ambient signal strengths while still providing useful location inferences in determining locations.

Abstract: A system that can analyze a multi-dimensional input thereafter establishing a search query based upon extracted features from the input. In a particular example, an image can be used as an input to a search mechanism. Pattern recognition and image analysis can be applied to the image thereafter establishing a search query that corresponds to features extracted from the image input. The system can also facilitate indexing multi-dimensional searchable items thereby making them available to be retrieved as results to a search query. More particularly, the system can employ text analysis, pattern and/or speech recognition mechanisms to extract features from searchable items. These extracted features can be employed to index the searchable items.

Abstract: A route planning system comprises a receiver component that receives a request for directions between a beginning point and a destination point. An analysis component analyzes a traffic system representation that varies as context varies and outputs expected amounts of travel time between the beginning point and the destination point for multiple contexts based at least in part upon the analysis. A method is described herein that includes techniques for searching over routes and trip start times simultaneously so as to identity start times and routes associated with maximal expected value, or equivalently minimum expected cost, given preferences encoded about one or more of the leaving time, the travel time, and the arrival time.

Abstract: The present invention relates to a system and methodology to facilitate collaboration and communications between entities such as between automated applications, parties to a communication and/or combinations thereof. The systems and methods of the present invention include a service that supports collaboration and communication by learning predictive models that provide forecasts of one or more aspects of a users' presence and availability. Presence forecasts include a user's current or future locations at different levels of location precision and usage of different devices or applications. Availability assessments include inferences about the cost of interrupting a user in different ways and a user's current or future access to one or more communication channels. The predictive models are constructed from data collected by considering user activity and proximity from multiple devices, in addition to analysis of the content of users' calendars, the time of day, and day of week, for example.

Abstract: An online dialog system and method are provided. The dialog system receives speech input and outputs an action according to its models. After executing the action, the system receives feedback from the environment or user. The system immediately utilizes the feedback to update its models in an online fashion.

Abstract: A system and method for identifying ideal channels for communications based on an analysis of communication channel reliability, communicating party preferences, and communicating party contexts is provided. The system attempts to optimize the utility of a communication based on inferred or directly accessed channel reliability data, communicating party preferences and communicating party contexts. Such optimization can be achieved using reliabilities, preferences and policies concerning handling the attempted contact based on a deterministic specification or through inferring reliability, context, content and task under uncertainty by employing decision-theoretic inferences. The methods may consider channels currently available as well as channels available at later times. Thus, the service can include automated rescheduling of communications based on a consideration of forecasts of reliability and availability.

Abstract: Conversational dialog with a computer or other processor-based device without requiring push-to-talk functionality. In one embodiment, a computer-implemented method first determines that a user desires to engage in a dialog. Based thereon the method turns on a speech recognition functionality for a period of time referred to as a listening horizon. Upon the listening horizon expiring, the method turns off the speech recognition functionality.

Abstract: Provided are systems and/or methods that treat illnesses and conditions using ultrasound tuned to a resonant frequency of a target material with the assistance of computer processing. The ultrasound tuned to the resonance frequency of a target material destroys the target material without harming healthy material that surrounds the target material. A resonance frequency database can be employed to ensure that local healthy material surrounding a target has a natural resonance frequency dissimilar enough from the tuned resonance frequency.

Abstract: A system and method for online reinforcement learning is provided. In particular, a method for performing the explore-vs.-exploit tradeoff is provided. Although the method is heuristic, it can be applied in a principled manner while simultaneously learning the parameters and/or structure of the model (e.g., Bayesian network model). The system includes a model which receives an input (e.g., from a user) and provides a probability distribution associated with uncertainty regarding parameters of the model to a decision engine. The decision engine can determine whether to exploit the information known to it or to explore to obtain additional information based, at least in part, upon the explore-vs.-exploit tradeoff (e.g., Thompson strategy). A reinforcement learning component can obtain additional information (e.g., feedback from a user) and update parameter(s) and/or the structure of the model.

Abstract: Sensing, learning, inference, and route analysis methods are described that center on the development and use of models that predict road speeds. In use, the system includes a receiver component that receives a traffic system representation, the traffic system representation includes velocities for a plurality of road segments over different contexts. A predictive component analyzes the traffic system representation and automatically assigns velocities to road segments within the traffic system representation, thereby providing more realistic velocities for different contexts where only statistics and/or posted speed limits were available before.

Abstract: Systems and methods are described for constructing predictive models, based on statistical machine learning, that can make forecasts about traffic flows and congestions, based on an abstraction of a traffic system into a set of random variables, including variables that represent the amount of time until there will be congestion at key troublespots and the time until congestions will resolve. Observational data includes traffic flows and dynamics, and other contextual data such as the time of day and day of week, holidays, school status, the timing and nature of major gatherings such as sporting events, weather reports, traffic incident reports, and construction and closure reports. The forecasting methods are used in alerting, the display graphical information about predictions about congestion on desktop on mobile devices, and in offline and real-time automated route recommendations and planning.

Abstract: A sensor monitoring system is described that identifies inaccurate or degraded sensor data and corresponding sensors. The system analyzes a plurality of sensors to determine reliable sensors and isolate or reinterpret unreliable sensor data. An analysis component analyzes a traffic representation that varies as context varies and determines the probability that current sensor data is invalid. Learning and reasoning are performed from a case library of sensor failure data to build predictive models of sensor health that can provide inferences about the output of sensors. Resulting inferences about sensor reliability can be employed in base-level inferences and predictions about traffic flow.

Abstract: Various components and processes are provided to enable data processing on multiple data types where aspects of the history of user activity, attention, interest, location, or other interaction with data is determined and employed to enhance information storage and access. In one particular aspect, a data manipulation system is provided. The system includes one or more data items that are associated with one or more tags and indicate at least one user's interaction or activity with the data items. A manipulation tool that processes the data items to determine a subset of data items based at least in part on the user's interaction with the data items. Methods are described for using the manipulation tool to weight terms in an index, to compress indexes, to influence the rank of items returned in a search, to generate additional queries for data items either automatically or with user direction, or for improved presentation of data items.

Abstract: The present invention relates to a system and methodology to facilitate collaboration and communications between entities such as between automated applications, parties to a communication and/or combinations thereof. The systems and methods of the present invention include a service that supports collaboration and communication by learning predictive continuous time Bayesian models that provide forecasts of one or more aspects of a users' presence and availability. Presence forecasts include a user's current or future locations at different levels of location precision and usage of different devices or applications. Availability assessments include inferences about the cost of interrupting a user in different ways and a user's current or future access to one or more communication channels.

Abstract: An information delivery system comprises a receiver component that receives information about the movement, velocity, acceleration, and/or locations over time of a user. A computation component using a predictive model generates a probability distribution relating to one or more of when the user will next be stopped, how long the user will be stopped, how long a pattern of motion, such as walking, driving in stop and go traffic, and smooth highway motion will last, based at least in part upon signals about motion over time. The system can further comprise an alerting component that determines when to provide the user with information based at least in part upon the probability distribution over some aspect of motion or cessation of motion, and optionally the content, or tagged or inferred urgency or importance, of a message or communication.

Abstract: Described is a technology by which a structured model of repetition is used to determine the words spoken by a user, and/or a corresponding database entry, based in part on a prior utterance. For a repeated utterance, a joint probability analysis is performed on (at least some of) the corresponding word sequences as recognized by one or more recognizers) and associated acoustic data. For example, a generative probabilistic model, or a maximum entropy model may be used in the analysis. The second utterance may be a repetition of the first utterance using the exact words, or another structural transformation thereof relative to the first utterance, such as an extension that adds one or more words, a truncation that removes one or more words, or a whole or partial spelling of one or more words.