摘要:
In accordance with one aspect of the invention, web workers and local storages can be extended to a cloud-based environment. This allows web workers to be executed on any of a number of different cloud platforms located in a cloud, leveraging available resources to provide a quicker and more efficient processing environment for the various web workers. The present invention also provides these functionalities in a way that is transparent to not just the user, but also to the web page developer as well, eliminating the need for the web page developer to be aware of the cloud-based environment and design the web page for use therewith.
摘要:
Techniques for assessing the cost of allocation of execution and affecting the allocation of execution are disclosed. The cost of allocation of execution between a first computing device (e.g., mobile device) and one or more computing resource providers (e.g., Clouds) can be determined during runtime of the code. A computing system can operate independently of the first computing device and a computing resource provider and provide execution allocation cost assessment. Execution allocation cost can be assessed based on execution allocation data pertaining to the first computing device and computing resource providers. Power consumption of a mobile device can be used as a factor in determining how to allocate individual components of an application program between a mobile phone and a Cloud. In an Elastic computing environment, external computing resources can be used to extend the computing capabilities beyond that which can be provided by internal computing resources.
摘要:
Techniques for allocating individually executable portions of executable code for execution in an Elastic computing environment are disclosed. In an Elastic computing environment, scalable and dynamic external computing resources can be used in order to effectively extend the computing capabilities beyond that which can be provided by internal computing resources of a computing system or environment. Machine learning can be used to automatically determine whether to allocate each individual portion of executable code (e.g., a Weblet) for execution to either internal computing resources of a computing system (e.g., a computing device) or external resources of an dynamically scalable computing resource (e.g., a Cloud). By way of example, status and preference data can be used to train a supervised learning mechanism to allow a computing device to automatically allocate executable code to internal and external computing resources of an Elastic computing environment.
摘要:
Access permission can be assigned to a particular individually executable portion of computer executable code (“component-specific access permission”) and enforced in connection with accessing the services of a service provider by the individually executable portion (or component). It should be noted that least one of the individually executable portions can request the services when executed by a dynamically scalable computing resource provider. In addition, general and component-specific access permissions respectively associated with executable computer code as a whole or one of it specific portions (or components) can be cancelled or rendered inoperable in response to an explicit request for cancellation.
摘要:
In one embodiment, a method for operating a local virtual world proxy is provided. A virtual world view is received corresponding to a virtual world client. Information about resources available to the virtual world client is obtained. Virtual representations of one or more resources available to a virtual world client are fetched. Then the virtual representations are injected into the virtual world view. The virtual world view is then forwarded to the virtual world client for display to a user of the virtual world client.
摘要:
In one embodiment, data relating to usage patterns of the user is stored, wherein the data includes information as to items which were used and the context in which they were used. A correlation table is constructed. Correlation values between each item and each context are then stored in then correlation table, wherein the correlations are used to recommend one or more of the items.
摘要:
A method for generating a portion of a query for a user is provided, the method comprising: retrieving information regarding the user's interests; and generating a portion of a query by analyzing the user's interests and the current situation of a user device operated by the user to determine at least one keyword.
摘要:
A definition of a set of context variables to be considered is received, wherein the context variables represent categories of situations in which the computer system can be. A definition of a set of activities to be considered is also received, wherein the activities represent activities that can be performed using the computer system. For each context variable to be considered, a definition of a set of states to be considered is received, wherein the states represent situations in which the computer system can be. For each context variable to be considered, a specification of first probability distributions for each corresponding state is received. For each context variable to be considered, for each activity to be considered, a specification of second probability distributions to represent the likelihood of the activity being performed in the corresponding context is received. Usage data is generated using the first second probability distributions.
摘要:
An input situation can be represented by at least a first context variable. Data that includes interest values for multiple context variables can be provided and obtained. The obtained data can include a first data pertaining to the input situation and a second data pertaining to one or more other situations. It can be determined whether the first context variable is associated with a discrete range of values or a continuous range of values. At least a portion of data pertaining to the situations can be determined to be proximate data when the first context variable is associated with a continuous range of values. Based on the input situation and the proximate data, an interest value for the first input situation can be determined as a prediction of the interest in the input situation.
摘要:
Data relating to usage patterns of the user is stored, wherein the data includes an application portion having information as to items which were used and a context portion having information as to the context in which the items were used. The data is clustered into clusters of data points and centroid are computed, wherein the centroid includes an application portion and a context portion. Clusters similar to a current context of the user are selected by comparing a data point representing the current context of the user to the context portions of one or more of the centroids. For each of one or more items, a expectation value that the user wishes to use the corresponding item is computed, based on the application portions of the selected similar clusters, wherein the expectation values are used to recommend one or more of the items.