Amazon Technologies, Inc.

Imdb. Com, Inc.

Jinesh Udaykumar Vora

James Hollis Wood

James Marvin Freeman, Ii

Wesley Shawn Davis

Hilliard Bruce Siegel

Adam Brent Johnson

Bradley Baron

Aaron James Dykstra

Adam Lloyd Days

Tanya Salyers

Tri M Cao

Yann Thierry Oehl

Scott Thomas Moran

Stephen Michael Brozovich

This disclosure describes various processes for collecting information about users from sources other than catalog activity data. This information can be used to generate recommendations for users with activity data deficiencies. Some example sources for this supplemental data include collaborative content sites, such as social networking sites. Social networking sites typically allow users to engage in conversations with other users through text, audio, and/or video. Conversation data collected from these sites or from other sources can be analyzed to infer user interests. A recommendation process (or other application) can use the inferred interests to supplement or take the place of catalog activity data.

Recommendations based on mined conversations

Virtual clickstream recommendations

Embodiments may include the generation and use of content source rankings for improving playback quality of content streams. In embodiments, a merchant system may collect and aggregate multiple quality metrics for different content streams that various content sources provide to various clients. The merchant system may utilize the quality metrics to generate rankings of content sources for clients. In some cases, a client itself may generate a ranking of content sources for use in the selection of content sources. The merchant system may provide such rankings to clients to improve the content stream playback experience on the client. For instance, at content acquisition time, a client may request a content stream from a content resource ranked highest on a content source ranking provided by the merchant. If the content stream degrades, the client may source subsequent portions of the content stream from another highly ranked content source of the ranking.

System and method for selection and switching of content sources for a streaming content session

A system and a method for the management of addressable network resources (e.g., Web pages, electronic documents accessed via a Uniform Resource Identifier, etc.) within a network are provided. Specifically, embodiments of addressable network resource selection systems and methods are disclosed for managing the analysis and identification of optimal links between addressable network resources. A resource selection component enables pruning of undesirable pages on the basis of various usage metrics, including user depart rate. The resource selection component may then identify optimal linking strategies by determining one or more linking scores for remaining pages based on any of a number of different page usage metrics. In other embodiments, addressable network resource selection systems and methods may weight linking scores to emphasize desired outcomes, such as encouraging exploration of a network-based site or encouraging high aggregate sales figures.

Addressable network resource selection management

Disclosed are various embodiments for identifying relationships between items included in user-created item lists and a predefined set of keywords based at least in part upon a comparison of the item list titles and the keywords. Each item lists title may be compared with the set of keywords to determine if any of the title terms in the item list title matches any of the keywords. If a match is detected, each of the items included in the item list are associated with the matching keyword. Items that are associated with a particular keyword based on the comparison may be added to an item category corresponding to the keyword.

Categorizing items using user created data

When a user receives digital content from a host over a network, such as streaming media, the user typically desires to receive a certain quality of the digital content with few or no interruptions. Techniques may be used to store and monitor a quality of digital content, a quality of the transmission of the digital content that is transmitted to a client device of a user, and/or the quality of rendering of the digital content by the client device. The techniques may provide restitution to the user when the quality of the digital content, transmission, and/or rendering thereof fails to comply with predetermined threshold values.

Digital returns

Various embodiments of a system and method for assignment and switching of content sources for a streaming content session are described. Embodiments may include the generation and use of content source rankings for improving playback quality of content streams. In embodiments, a merchant system may collect and aggregate multiple quality metrics for different content streams that various content sources provide to various clients. The merchant system may utilize the quality metrics to generate rankings of content sources for clients. The merchant system may provide such rankings to clients to improve the content stream playback experience on the client. For instance, at content acquisition time, a client may request a content stream from a content resource ranked highest on a content source ranking provided by the merchant. If the content stream degrades, the client may source subsequent portions of the content stream from another highly ranked content source of the ranking.

System and method for assignment and switching of content sources for a streaming content session

Disclosed are various embodiments for generating recommendations based at least in part on keywords associated with users. In some embodiments, among others, a system includes at least one computing device and a recommendation generator executable in the at least one computing device. The recommendation generator comprises logic that generates a plurality of pools of keywords based at least in part on a plurality of behavioral histories. Each pool corresponds to a behavioral history of a user across a plurality of domains. The recommendation generator also comprises logic that clusters at least a number of the keywords in a cluster across at least two pools including the same keyword and logic that recommends an item based at least in part on the cluster of keywords.

Customer keyword-based item recommendations

Disclosed are various embodiments for forming relationships among items to enhance user experiences in navigating a content database. A user may initiate a request to associate a keyword or label with an item. The keyword may be associated with the user to form a user-keyword association and may be associated with the item to form an item-keyword association. If the item-keyword association meets a predefined threshold, the keyword may be surfaced as a community keyword corresponding to the item.

Growing community of inventors

Issaquah, WA, United States of America

Jinesh Udaykumar Vora