Abstract: A media asset recommendation service is described. In embodiments, friends of a user are identified in a social graph that is associated with the user, and interactions with media assets by the friends of the user are compiled. Friend interaction data for each of the media assets can be determined from the media assets, the social graph, and the interactions with the media assets by the friends of the user. When a selection of one of the media assets is received from the user via a user device, the friend interaction data that is associated with the media asset selected by the user is communicated to the user device.

Abstract: Service endpoints that provide services to client devices send registration data to a service directory service. The registration data includes addresses that client devices can use to access the service endpoints along with descriptive data that describes the services that each endpoint provides, along with other information such as version numbers of applications that are supported by the service endpoints, and whether the service endpoints are running an experiment. Client devices request a manifest from the service directory service. Each request includes client metadata that describes the requesting client device. The service directory service uses the registration data and the client metadata to generate a manifest for the client device that identifies the services that the client may access and an address that the client may use to access each service.

Abstract: Client playlist generation is described. In embodiments, relationships between media assets are determined to identify similar media assets that can be included in an automatic playlist of the similar media assets. Projection vectors of the asset-to-asset relationships can be generated for each of the media assets, where a projection vector for a media asset identifies the similar media assets. The projection vectors are then communicated to a client device that utilizes the projection vectors to generate the automatic playlist for any one of the media assets that is selected as a starting media asset of the automatic playlist.

Abstract: Search techniques are described. In one or more implementations, a search service receives a search query from a user. A query context is formed for the user based on a history of content consumption that is verified for the user. A plurality of search results are generated that reference different types of content that correspond to the search query, respectively. The plurality of search results are ordered for output in a user interface according to the different types of content based at least in part on the query context.

Abstract: Similarity and relatedness content techniques are described. Responsive to an indication that an item of content is to be made available via an online marketplace, metadata is collected from one or more network services that describes the item of content. Similarity of the item of content with other items of content included in the online marketplace of the network service is calculated based at least in part on the collected metadata. A determination is made as to whether the item of content is related to the other items of content available via the online marketplace based at least in part on the collected metadata, the item of content being related to the other items of content based on membership in a common group. Results of the calculating and the determining are used in making the item of content available via the online marketplace.

Abstract: Content item recommendations are generated for users based on metadata associated with the content items and a history of content item usage associated with the users. Each content item recommendation identifies a user and a content item and includes a score that indicates how likely the user is to view the content item. Based on the content item recommendations, and constraints of one or more caches, the content items are selected for storage in one or more caches. The constraints may include users that are associated with each cache, the geographical location of each cache, the size of each cache, and/or costs associated with each cache such as bandwidth costs. The content items stored in a cache are recommended to users associated with the cache.

Abstract: Service endpoints that provide services to client devices send registration data to a service directory service. The registration data includes addresses that client devices can use to access the service endpoints along with descriptive data that describes the services that each endpoint provides, along with other information such as version numbers of applications that are supported by the service endpoints, and whether the service endpoints are running an experiment. Client devices request a manifest from the service directory service. Each request includes client metadata that describes the requesting client device. The service directory service uses the registration data and the client metadata to generate a manifest for the client device that identifies the services that the client may access and an address that the client may use to access each service.

Abstract: A system and methods are provided for automatically classifying data according to perceptual properties of the data to form a classification chain that is suited to the searching and sorting of large databases of media entities. During classification, experts assign each media entity in the training data set to one or more classes, with each class corresponding to a given subset of perceptual properties of the data. In conjunction with digital signal processing properties of the data corresponding to the perceptual properties, the classified data is then used to construct an initial classification chain. During operation, when presented with an unclassified entry, the classification chain returns an estimate of the class of the entry, as well as a confidence measure that is proportional to the level of confidence of the class assignment. Over time, as the classification chain evolves, the classification chain becomes more and more effective for quickly characterizing media entities.

Abstract: Client playlist generation is described. In embodiments, relationships between media assets are determined to identify similar media assets that can be included in an automatic playlist of the similar media assets. Projection vectors of the asset-to-asset relationships can be generated for each of the media assets, where a projection vector for a media asset identifies the similar media assets. The projection vectors are then communicated to a client device that utilizes the projection vectors to generate the automatic playlist for any one of the media assets that is selected as a starting media asset of the automatic playlist.

Abstract: An interactive media social game is described. In embodiments, a social game service administrates a networked social game based on media asset interactions. Media asset interaction data is received that identifies media assets when they are downloaded or played at a user device. The media assets are correlated with a social game entity, such as a music pet that is a token entity of the networked social game. The music pet adaptively changes according to characteristics of the media assets. The media asset interaction data is also compiled as a media asset selection history corresponding to a user that is associated with the user device. A media asset recommendation for the user can be generated based on the media asset selection history, and the media asset recommendation is communicated as a recommendation message from the music pet to the user that is associated with the user device.

Abstract: The claimed subject matter provides a system and/or a method that facilitates identifying relationships between two or more trade cards. A schema component can implement a portion of user-assisted schema to create a trade card that represents a distillation of a document with document-specific data. A relationship component can identify at least one of a hard link or a soft link associated with the trade card in connection with at least one of a network or one or more trade cards, wherein the hard link is a source that derived the trade card and the soft link is directed toward a disparate trade card with a portion of substantially similar document-specific data.

Abstract: Media playlist techniques are described. In an implementation, a communication is formed, for transmission via a network to a media provider, to request data that describes media that is available for streaming from the media provider. The playlist is created based at least in part on media that is available locally on the computer and the data received from the media provider such that the playlist references one or more of the media that are available for streaming from the media provider.

Abstract: Media recommendation techniques are described. In an implementation, a similarity value is calculated for a plurality of media using a plurality of similarity functions. A vote is assigned for each similarity value that is above a threshold that is assigned for a respective similarity function and the plurality of media is ranked based at least in part on the assigned votes. A playlist is then created based at least in part on the ranking. Media seed techniques are also described. In an implementation, a set of dissimilar candidates are calculated for a plurality of media using a similarity function in which the set of dissimilar candidates describes the media that is dissimilar in comparison with other media included in the plurality of media. A seed is selected using the set of the dissimilar candidates to create a playlist that includes at least some of the plurality of media.

Abstract: Media recommendation techniques are described. In an implementation, a similarity value is calculated for a plurality of media using a plurality of similarity functions. A vote is assigned for each similarity value that is above a threshold that is assigned for a respective similarity function and the plurality of media is ranked based at least in part on the assigned votes. A playlist is then created based at least in part on the ranking. Media seed techniques are also described. In an implementation, a set of dissimilar candidates are calculated for a plurality of media using a similarity function in which the set of dissimilar candidates describes the media that is dissimilar in comparison with other media included in the plurality of media. A seed is selected using the set of the dissimilar candidates to create a playlist that includes at least some of the plurality of media.

Abstract: An event recommendation service is described. In embodiments, selection data that correlates to media content selected by a user, location data that corresponds to a location of the user, and event data is received. A recommendation for an event that the user is likely to attend and that is proximate the location of the user can be generated by aggregating the selection data, the location data, and the event data. The recommendation can then be communicated to a user device as a calendar entry, an email message, a text message, and/or an html page for display to inform the user of the event.

Abstract: A social graph playlist service is described. In embodiments, a social graph that associates a user and friends of the user is maintained. The social graph is based on parameters that define a social relationship between the user and the friends of the user. Interaction data that identifies recently played media assets at user devices that are utilized by the friends of the user can be received. A social graph playlist that is associated with the user can be generated by determining a next media asset for the playlist from the recently played media assets. The next media asset for the playlist can then be communicated to be played at a user device that is associated with the user.

Abstract: A media asset recommendation service is described. In embodiments, friends of a user are identified in a social graph that is associated with the user, and interactions with media assets by the friends of the user are compiled. Friend interaction data for each of the media assets can be determined from the media assets, the social graph, and the interactions with the media assets by the friends of the user. When a selection of one of the media assets is received from the user via a user device, the friend interaction data that is associated with the media asset selected by the user is communicated to the user device.

Abstract: In connection with a classification system for classifying media entities that merges perceptual classification techniques and digital signal processing classification techniques for improved classification of media entities, a system and methods are provided for automatically classifying and characterizing musical consonance properties of media entities. Such a system and methods may be useful for the indexing of a database or other storage collection of media entities, such as media entities that are audio files, or have portions that are audio files. The methods also help to determine media entities that have similar consonance by utilizing classification chain techniques that test distances between media entities in terms of their properties. For example, a neighborhood of songs may be determined within which each song has a similar consonance.

Abstract: A reputation system used in a social networking service provides for recognition for its members in the form of badges that can be displayed on the members' profile pages as a way of indicating a particular status. In various illustrative examples, a “tastemaker” badge may be awarded to a member who plays or recommends to other members (through various recommendation channels such as messages, shared playlists, etc.) particular pieces of media content such as songs or videos in advance of that content becoming popular with the larger community of members in the social network. The tastemaker badge can then be placed on the member's profile page as a symbol of recognition or achievement that can help to increase that member's reputation in the social network.

Abstract: A system and method for incorporating user preferences into a document search is provided. User events are associated with one or more content tags based on a comparison of the user event with locations in a content space. The user events are tracked to provide a user event history. The user event history is analyzed to select content tags that correspond to user preferences. A search query is modified to include one or more selected content tags. The search query is then matched to documents associated with content tags contained in the search query.