Microsoft Technology Licensing, LLC

Bandwiz Inc.

Joseph Yossi Azar

Eyal Lubetzky

Iftah Gamzu

Xiaoxin Yin

Eric J Horvitz

Doron Rajwan

Thomas Moscibroda

Dafna Shahaf

Ori Gurel-Gurevich

Many computing scenarios involve an item cache or index, comprising items corresponding to source items that may change without notice, rendering the item in the item cache or index stale. It may not be possible to guarantee the freshness of the items, but it may be desirable to reduce staleness in an efficient manner. Therefore, the refreshing of items may be prioritized by first predicting the query frequency of respective item representing the rate at which an item is retrieved from the item cache (e.g., by monitoring queries for the item), predicting an update frequency representing the rate at which the source item is updated by the source item host (e.g., by classifying the source item type), and computing a refresh utility representing the improvement in cache freshness achieved by refreshing the item. Respective items may then be prioritized for refreshing according to the computed refresh utilities.

Cache and index refreshing strategies for variably dynamic items and accesses

Techniques and systems are disclosed providing improved ranking of results to an online search-based query. One or more user types are identified for a search-based query, and may correspond to a number of user relevant results, and which user results are selected. A user profile can be determined for the respective user types for the search-based query, which may identify a proportion of the users that belong to that type, and how many results are relevant to that type. A set of relevant results can be identified for the respective user types for the search-based query, based on a number of results used by the user type. An improved ranking of the results can be determined for the search-based query, from the one or more sets of relevant results, based on user profiles and a desired user satisfaction metric for a desired number of users.

Ranking results of multiple intent queries

A computing device configured to select a channel that is open in a local environment of the computing device and to attempt to communicate with another computing device on the selected available channel is described herein. The computing device selects the available channel based on selection probabilities for a plurality of available channels, the selection probabilities defined by an exponentially decaying distribution curve.

Channel discovery protocol

Techniques and systems are disclosed providing improved ranking of results to an online search-based query. One or more user types are identified for a search-based query, which may correspond to a number of user relevant results and which ones are selected. A user profile can be determined for the respective user types for the search-based query, which may identify a proportion of the users that belong to that type, and how many results are relevant to that type. A set of relevant results can be identified for the respective user types for the query, based on a number of results used by the user type. An improved ranking of the results can be determined for the query, from the one or more sets of relevant results, based on the user profiles and a desired user satisfaction metric for a desired number of users.

A method of streaming data, comprising: receiving an input stream of data, at a transmitter; dividing the input stream into a plurality of blocks, at least one of which blocks is incomplete; generating a plurality of first packets based on at least one block of data; generating at least one second packet from at least one as yet incomplete block of data; said block comprising recently received data; transmitting said at least one first packet and at least one second packet to a receiver that can reconstruct said stream from said first packets and said second packets, said transmitting utilizing a differential protocol by which different parts of the data are transmitted at different rates, so that a receiver can join the transmission at any time and start receiving the data at a minimum delay; and generating at least one third packet from said at least one incomplete block; said at least one third packet being based at least in part on data received subsequent to data forming the basis for the at least one second packet; wherein said at least one second packet is transmitted at a higher rate than mandated by said protocol to compensate for a later repeated transmission of information carried in said at least one second packet at a lower rate than mandated by the protocol, once said at least one third packet is generated and transmitted.

Growing community of inventors

Redmond, WA, United States of America

Joseph Yossi Azar