Abstract: In some instances, an image may have dimensions that do not correspond to a slot to display the image. For example, an image content item may have dimensions that do not correspond to a content item slot. The image may be resized using seam carving to add or remove pixels of the image. A saliency map for the image may be used having saliency scores for each pixel of the image. Evaluation metrics may be used before, during, and after, seam carving to determine whether salient content is affected by the seam carving. In some instances, a seam cost threshold value may be used for adaptive step size during the seam carving. The resized image may then be outputted, such as for an image content item to be served with a resource.

Abstract: Methods, systems and apparatus, including computer program products are described for ranking content items identified by a search engine and delivering corresponding search results. In one aspect, search engine user language preference data is stored in association with user content item selection records. Analysis of the records is performed to identify content items that appeal to users having common language preferences. Query results can be ranked based on the language preference of the current user and/or the user's query and data derived from the selection records.

Abstract: Methods and apparatus for determining aspects of locations based on computing interactions of users that pertain to the locations. In some implementations, an aspect of a location may be determined based on comparing a first computing interaction measure for the location to a second measure. The first computing interaction measure may be determined based on a first group of computing interactions that pertain to the location. In some implementations, the second measure may be determined based on a second group of addition computing interactions. In some implementations, a textual summary may be generated automatically based on the first and/or second computing interaction measure, and/or based on the determined aspect.

Abstract: Methods, systems, and computer program products are provided for information presentation. In one implementation, a method is provided. The method includes generating a message, identifying suggested content to be included with the message, suggesting one or more of the suggested content items to a user, receiving a user input with respect to the one or more suggested content items, and sending the message according to the user input including one or more selected content items of the one or more suggested content items.

Abstract: A system and method for identifying contrarian terms is disclosed. The system and method includes identifying a plurality of websites describing a product or service, analyzing content in the plurality of websites, the content relating to the described product or service, identifying contrarian terms based on the website content using a processing circuit, wherein the contrarian terms comprise descriptions of a product or service, associating the identified contrarian terms with a category, within which the product or service has been categorized in at least one of the plurality of websites, and storing the contrarian terms and the associated category in a memory.

Abstract: A system and method for generating a stream of content for a channel. The channel application includes a content categorizer, a scoring engine and a channel engine. The content categorizer categorizes new content items received from heterogeneous data sources. The channel engine identifies a channel category for a user based at least in part on at least one of a historical trend and a user activity. The scoring engine queries the new content items based on the channel category and at least one other channel attribute. The scoring engine retrieves candidate content items that include the channel category and the other channel attribute. The scoring engine then generates a stream of content from the candidate content items for the channel.

Abstract: In some instances, an image may have dimensions that do not correspond to a slot to display the image. For example, an image content item may have dimensions that do not correspond to a content item slot. The image may be resized using seam carving to add or remove pixels of the image. A saliency map for the image may be used having saliency scores for each pixel of the image. Evaluation metrics may be used before, during, and after, seam carving to determine whether salient content is affected by the seam carving. In some instances, a seam cost threshold value may be used for adaptive step size during the seam carving. The resized image may then be outputted, such as for an image content item to be served with a resource.

Abstract: In some instances, an image may have dimensions that do not correspond to a slot to display the image. For example, an image content item may have dimensions that do not correspond to a content item slot. The image may be resized using seam carving to add or remove pixels of the image. A saliency map for the image may be used having saliency scores for each pixel of the image. Evaluation metrics may be used before, during, and after, seam carving to determine whether salient content is affected by the seam carving. In some instances, a seam cost threshold value may be used for adaptive step size during the seam carving. The resized image may then be outputted, such as for an image content item to be served with a resource.

Abstract: In some instances, an image may have dimensions that do not correspond to a slot to display the image. For example, an image content item may have dimensions that do not correspond to a content item slot. The image may be resized using seam carving to add or remove pixels of the image. A saliency map for the image may be used having saliency scores for each pixel of the image. Evaluation metrics may be used before, during, and after, seam carving to determine whether salient content is affected by the seam carving. In some instances, a seam cost threshold value may be used for adaptive step size during the seam carving. The resized image may then be outputted, such as for an image content item to be served with a resource.

Abstract: A system and method for transferring annotations associated with a media file. An annotation associated with a media file is indexed to a first instance of that media file. By comparing features of the two instances, a mapping is created between the first instance of the media file and a second instance of the media file. The annotation can be indexed to the second instance using the mapping between the first and second instances. The annotation can be processed (displayed, stored, or modified) based on the index to the second instance.

Abstract: A system and method for transferring annotations associated with a media file. An annotation associated with a media file is indexed to a first instance of that media file. By comparing features of the two instances, a mapping is created between the first instance of the media file and a second instance of the media file. The annotation can be indexed to the second instance using the mapping between the first and second instances. The annotation can be processed (displayed, stored, or modified) based on the index to the second instance.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining a context of the current search session. In one aspect, a method includes identifying query paths from previous search sessions, each query path defining a context and being a plurality of queries in an order in which the queries were provided for in a respective previous search session; identifying search session queries of a current search session; comparing the search session queries of the current search session to the queries in the query paths from the previous search sessions; and determining that a context of the current search session is related to a query path from the previous search sessions based at least in part on the comparison, the determining including: determining that two or more of the queries of the query path are similar to two or more of the search session queries.

Abstract: A system and method for identifying contrarian terms is disclosed. The system and method includes identifying a plurality of websites describing a product or service, analyzing content in the plurality of websites, the content relating to the described product or service, identifying contrarian terms based on the website content using a processing circuit, wherein the contrarian terms comprise descriptions of a product or service, associating the identified contrarian terms with a category, within which the product or service has been categorized in at least one of the plurality of websites, and storing the contrarian terms and the associated category in a memory.

Abstract: A first content entity and one or more associated second content entities are presented to one or more arbiters. Arbiter determinations relating to the association of at least one of the second content entities with the first content entity are received. A determination as to whether the at least one of the second content entities is to be disassociated from the first content entity based on the arbiter determinations can be made.

Abstract: Methods, systems, and apparatus, including computer program products, for assigning weights to parts of a document. In one aspect, document object is generated for a collection of documents. Weights for the respective portions of the document object model are generated, and a document is ranked in a search engine index, the ranking being based at least in part on the generated weights.

Abstract: Systems, methods, and apparatus, including computer program products, for collaborative filtering are provided. A method is provided. The method includes clustering a plurality of entities with respect to one or more latent variables in a probability distribution model of a relationship between a set of entities and a set of items, the probability distribution model comprising a probability distribution of the set of items with respect to the latent variables. The method also includes, as new items are added to the set of items, updating the probability distribution of the set of the items with respect to the latent variables, and generating an updated relationship score for an entity with respect to the set of items based on the entity's fractional membership in the clustering with respect to the latent variables and based on the updated probability distribution of the set of the items with respect to the latent variables.

Abstract: Systems, methods, and apparatus, including computer program products, for collaborative filtering are provided. In one implementation, a computer-implemented method is provided. The method includes receiving a shard of data representing a subset of a set of entities and a subset of a set of items, generating an iteration of a maximum likelihood estimate of a probability distribution model of a relationship between the set of entities and the set of items, the probability distribution model comprising a probability distribution of the set of items with respect to latent variables and a probability distribution of the latent variables with respect to the set of users, and generating statistics from results from the generating step which are passed to different shards for use in a next iteration of the maximum likelihood estimate.

Abstract: A system and method to target advertisements to a user search query using a previously entered user search query as well as the current search query. The previously entered search query can accompanying the current search query in a query referrer, or can be determined separately from the current search query. The relevance of the previous search query is determined with respect to the current search query, and if found relevant a combination of the two query terms is used to find a set of advertising keywords from which candidate advertisements to be targeted and served to the user with search results are found. The process can be extended to additional search queries.

Abstract: Methods, systems and apparatus, including computer program products are described for ranking content items identified by a search engine and delivering corresponding search results. In one aspect, search engine user language preference data is stored in association with user content item selection records. Analysis of the records is performed to identify content items that appeal to users having common language preferences. Query results can be ranked based on the language preference of the current user and/or the user's query and data derived from the selection records.

Abstract: Methods, systems and apparatus, including computer program products, for providing a diversity of recommendations. According to one method, results are identified so as to increase the likelihood that at least one result will be of interest to a user. Following the identification of a first result, second and later results are identified based on an assumption that the previously identified results are not of interest to the user. The identification of diverse results can be based on formulas that approximate the probability or provide a likelihood score of a user selecting a given result, where a measured similarity between a given object and previously identified results tends to decrease the calculated probability approximation or likelihood score for that object.