摘要:
Techniques for creating breadcrumbs for a trail of activity are described. The trail of activity may be created by recording movement information based on inferred actions of walking, not walking, or changing floor levels. The movement information may be recorded with an accelerometer and a pressure sensor. A representation of a list of breadcrumbs may be visually displayed on a user interface of a mobile device, in a reverse order to retrace steps. In some implementations, a compass may additionally or alternatively be used to collect directional information relative to the earth's magnetic poles.
摘要:
A source wireless device is capable of transmitting items over a wireless network using a number of different wireless transfer mechanism such as, for example, Infra-Red (IR) or Bluetooth. Several potential destination wireless devices may be capable of receiving the item using at least one of the different wireless transfer mechanisms. The user selects one or more destination wireless devices to send items to using a unified user interface that lists the various potential destination wireless devices. After a user selects items to transfer as well as the destination wireless device(s), the wireless transfer mechanism appropriate for transfer to each device is then automatically identified without requiring user selection of the particular wireless transfer mechanism.
摘要:
Modulating the behavior of an agent for directing automated services such as for messaging and scheduling is disclosed. In one embodiment, a computer-implemented method inputs an action probability. The action probability has a location within a region corresponding to an option that in one embodiment can be one of: (1) inaction, (2) automatic action, and (3) engaging the user in a dialog about the user's desire for automated action, followed by action pending receipt of user approval. The method then modulates the behavior of the agent depending on the location of the action probability within the region. In one embodiment, the region in which the action probability is located is divided into two or more sub-regions, such that the behavior of the agent is modulated based on which sub-region in which it lies. In another embodiment, a continuous function is imposed onto the region in which the probability is located, where the function yields a confidence level based on the location of the action probability within the region, such that the behavior of the agent is modulated based on this confidence level.
摘要:
Directing automated services for messaging and scheduling. In one embodiment, a computer-implemented method first determines a message to analyze. The method then determines the probability that the user would like to review a calendar or to schedule an appointment based on information in the message, based on the message, and based on the scheduling probability, selects one of the following options: (1) inaction, (2) automatic action, or (3) suggested action with user approval. Upon the method selecting either the (1) automatic action option or the (2) suggested action with user approval option—the latter also in conjunction with receiving actual user approval—the method performs a scheduling action based on the message in a manner that takes into consideration the precision and confidence of the scheduling action.
摘要:
A system and method for personalizing an information classifier is provided. The system includes a default or universal classifier that is pre-trained with features that have relatively universal discriminatory relationships for multiple people and that is operable to produce a measure that a message is classified as having one of several characteristics. The system further includes a second classifier that is constructed and personalized through training via a more general search through the space of potentially discriminatory features. The second classifier, after personalization, is intended to classify information at a level exceeding that of the first classifier based on the specific preferences, habits, and desires of the user who personalizes the second classifier. The system further includes a weighting component that facilitates a combining component producing an integrated measure based on input from both the first classifier and the second classifier.
摘要:
The present invention relates to a system (10) and methodology (170, 173, 180, 400, 600, 1000) to enable a variety of information associated with one or more notification sources (26-28) to be directed to one or more notification sinks (36-38) via a notification platform architecture (10). The architecture (10) includes a context analyzer (22) for determining a user's state such as location and attentional focus, wherein the user's state is employed by a notification manager (24) to make decisions regarding what, when and how information generated by the notification sources (26-28) should be forwarded to the notification sinks (36-38), for example. These decisions can include a cost benefit analysis wherein considerations are given as to whether the benefits of notifying the user are outweighed by the costs of disrupting the user. Decision-theoretic policies (177, 180) and/or somewhat less formal heuristic policies (177) can be employed to enable the decision-making process within the notification manager (24).
摘要:
A source wireless device is capable of transmitting items over a wireless network using a number of different wireless transfer mechanism such as, for example, Infra-Red (IR) or Bluetooth. Several potential destination wireless devices may be capable of receiving the item using at least one of the different wireless transfer mechanisms. The user selects one or more destination wireless devices to send items to using a unified user interface that lists the various potential destination wireless devices. After a user selects items to transfer as well as the destination wireless device(s), the wireless transfer mechanism appropriate for transfer to each device is then automatically identified without requiring user selection of the particular wireless transfer mechanism.
摘要:
Determination of the current context of the user, such as the user's current location and attentional state, is disclosed. The determined context can be used to assist determination as to whether, when and how notifications intended for the user should be conveyed to him or her. In varying embodiments of the invention, the context is determined via one or more of: direct specification by the user; direct measurement using one or more sensors; a user-modifiable profile indicating context; one or more potentially user-modifiable rules that indicate context; and, and inferential analysis utilizing a model, such as a Bayesian network or other statistical model.
摘要:
The present invention relates to a system and methodology to enable a variety of information associated with one or more notification sources to be directed to one or more notification sinks via a notification platform architecture. The architecture includes a context analyzer for determining a user's state such as location and attentional focus, wherein the user's state is employed by a notification manager to make decisions regarding what, when and how information generated by the notification sources should be forwarded to the notification sinks, for example. These decisions can include a cost benefit analysis wherein considerations are given as to whether the benefits of notifying the user are outweighed by the costs of disrupting the user. Decision-theoretic policies and/or somewhat less formal heuristic policies can be employed to enable the decision-making process within the notification manager.
摘要:
A system and method are provided to learn and infer the time until a user will be available for communications, collaboration, or information access, given evidence about such observations as time of day, calendar, location, presence, and activity. The methods can be harnessed to coordinate communications between parties via particular modalities of interaction. The system includes a user state identifier that determines a user's state from background knowledge, the flow of time, or one or more context information sources. A data log can be employed to store information about user state changes and observational evidence to accumulate statistics and build inferential models of the availability and unavailability of users for different kinds of communication, collaboration, and information access. A forecaster is constructed from the accumulated statistics and/or learned models to enable a determination of a user's likely return, or, more generally, the probability distribution over a user's likely return to particular states of availability. The forecaster can be employed to cache information for offline access, drive displays of availability and unavailability, to send messages that include availability forecasts, and to automatically perform scheduling or rescheduling of communications.