Digital Fountain, Inc.

A9.com, Inc.

Diane Hernek

Michael G Luby

John Byers

Lars Eilstrup Rasmussen

Roberto Attias

Ronen Vainish

David Kushi

Serban Simu

Adrian Perrig

Michael Walfish

Gavin Bernard Horn

Michael David Mitzenmacher

Armin D Haken

Yue Zhou

Nicholas E Matsakis

Benoit J Dupin

Steve Martin

Dimitris Margaritis

A method of serving content to multiple clients via a network is provided. Independent sessions with each of a plurality of clients are maintained, wherein the number of clients in the plurality of clients can vary over time, and wherein the start of each session and the end of each session can be independent of the start and end of other sessions. A stream of packet payloads is received, each packet payload of the stream of packet payloads including data generated from the content, wherein each packet payload in at least a subset of the stream of packet payloads includes a different set of data. Each packet payload in the stream of packet payloads is transmitted to each client of the plurality of clients in corresponding packets, wherein the packet payload transmitted to a client at any particular time is independent of the state of the corresponding session.

Multi-output packet server with independent streams

Systems, methods, and apparatus are provided for determining relevance of documents to queries. An optimized relevance function is configured to determine a relevance value of documents of a first type that are linked to documents of a second type. The relevance function is optimized to satisfy certain criteria. According to one criterion, a relevance value produced by the optimized relevance function, when invoked for documents of the first type, should have a locally maximal degree of fit to the results of the existing relevance function for the sample training documents of the second type. An assessed degree of fit of a document can be increased or decreased to arrive at an optimized relevance function that ranks the documents of the first type in a desired position relative to documents of the second type in search results. The degree of fit can be assessed by a user-provided objective function.

Ranking non-product documents

A technique for associating keywords with a set of items is provided. The present invention associates one or more keywords with terms found in the two or more characterization fields of an item. Each combination associates each keyword with at most one field. The present invention then calculates a keyword relevancy score for the set of potential keyword field combinations and identifies the keyword field combination with an optimal keyword relevancy score for the item. The method further includes displaying a list of results based on keyword relevancy scores for items.

Relevance scoring based on optimized keyword characterization field combinations

Output symbols are generated using input symbols and basis elements. A plurality of basis elements are generated. Each basis element is generated from a predetermined function of associated input symbols associated with the basis element. For each output symbol, a set of associated basis elements associated with the output symbol is determined, and a set of direct associated input symbols directly associated with the output symbol is determined. For each output symbol, the output symbol is generated from a predetermined function of the associated basis elements and the associated input symbols.

Generating high weight encoding symbols using a basis

Growing community of inventors

Oakland, CA, United States of America

Diane Hernek