Abstract: Described herein is a technology that facilitates efficient large-scale similarity-based retrieval. In several embodiments documents, images, and/or other multimedia files are compactly represented and efficiently indexed to enable robust search using a long-query in a large-scale corpus. As described herein, these techniques include performing decomposition of a file, e.g., a document or document-like representation. The techniques use dimension reduction to obtain three parts, topic-related words (major semantics), document specific words (minor semantics), and background words, representing the major semantics in a feature vector and the minor semantics as keywords. Using the techniques described, file vectors are matched in a topic model and the results ranked based on the keywords.

Abstract: A method and system for calculating the importance of persons based on interpersonal relationships and prioritizing communications based on importance of participants in the communications is provided. A prioritization system identifies relationships between persons and identifies the importance of a person to other persons based on these relationships. After the prioritization system identifies the importance of persons, the prioritization system can prioritize communications based on the importance of the senders or recipients.

Abstract: Described herein is a technology that facilitates learning interests for advertising based on automated analysis of images. In several embodiments a person's interests are automatically learned based on the person's photographs for targeted advertising. Techniques are described that facilitate automatically detecting a user's interest from images and suggesting user-targeted ads. As described herein, these techniques include computer-annotating images with learned tags, performing topic learning to obtain an interest model, and performing advertisement matching and ranking based on the interest model.

Abstract: A classification system trains a classifier to classify blocks of the web page into various classifications of the function of the block. The classification system trains a classifier using training web pages. To train a classifier, the classification system identifies the blocks of the training web pages, generates feature vectors for the blocks that include a linguistic feature, and inputs classification labels for each block. The classification system learns the coefficients of the classifier using any of a variety of machine learning techniques. The classification system can then use the classifier to classify blocks of web pages.

Abstract: A method and system for determining the contribution of a document within a hierarchy of documents based on the contribution of descendant documents is provided. The contribution system provides a hierarchy of documents that specifies the ancestor/descendant relations between documents. For each document of a hierarchy, the contribution system determines the contribution of each document factoring in the contribution of descendant documents. The contribution may be the relevance of a document to a topic, a feature of a document, and so on.

Abstract: The invention provides a stable concentrated dairy liquid, such as concentrated milk, with improved flavor, color, and mouthfeel, and a method of production thereof. The method utilizes specific thermal treatments to produce the stable concentrated dairy liquid to which a stabilizer and mouthfeel enhancer are added. The resulting products have a sterilization value Fo of at least 5 that is also resistant to gelling and browning during high temperature sterilization and is also resistant to gelling and browning during storage for greater than six months. The method balances such thermal treatments with addition of stabilizer and enhancer to achieve the desired flavor/mouthfeel and sterilization and to achieve reduced level of soluble protein in the concentrated milk prior to concentration to resist gelation and minimize browning. Moreover, such processing may be utilized in a milk that is concentrated to a factor of 2.7 fold or higher and contains at least 8.5 percent protein.

Abstract: In community mining based on core objects and affiliated objects, a set of core objects for a community of objects are identified from a plurality of objects. The community is expanded, based on the set of core objects, to include a set of affiliated objects. According to one aspect, a model of a community of objects is obtained by grouping a first collection of a plurality of objects into a center portion, and grouping a second collection of the plurality of objects into one or more concentric portions around the center portion. The groupings of the first and second collections of the objects are identified as the community of objects.

Abstract: A method and system for determining relevance of a document having text and images to a text string is provided. A scoring system identifies image text associated with an image of the document. The scoring system calculates an image score indicating relevance of the image text to the text string. The image score may be used in many applications, such as searching, summary generation, and document classification, image search, and image classification.

Abstract: A large web page is analyzed and partitioned into smaller sub-pages so that a user can navigate the web page on a small form factor device. The user can browse the sub-pages to find and read information in the content of the large web page. The partitioning can be performed at a web server, an edge server, at the small form factor device, or can be distributed across one or more such devices. The analysis leverages design habits of a web page author to extract a representation structure of an authored web page. The extracted representation structure includes high level structure using several markup language tag selection rules and low level structure using visual boundary detection in which visual units of the low level structure are provided by clustering markup language tags. User viewing habits can be learned to display favorite parts of a web page.

Abstract: A method and system for searching for content relating to a target or query image by identifying duplicate images with associated content is provided. An image search system identifies visual parts of objects within the target image based on analysis of two or more versions of the target image. The image search system identifies visual parts based on analysis of the versions. The image search system then identifies images of an image database that have visual parts that are similar to the visual parts of the target image. The image search system may rank the identified images based on their likelihood of being duplicates of the target image and provide their associated content as the search result ordered according to the ranking of the images.

Abstract: Mining geographic knowledge using a location aware topic model is provided. A location system estimates topics and locations associated with documents based on a location aware topic (“LAT”) model. The location system generates the model from a collection of documents that are labeled with their associated locations. The location system generates collection level parameters based on an LDA-style model. To generate the collection level parameters, the location system estimates probabilities of latent topics, locations, and words of the collection. After the model is generated, the location system uses the collection level parameters to estimate probabilities of topics and locations being associated with target documents.

Abstract: A method and system for adapting search results of a query to the information needs of the user submitting the query is provided. A search system analyzes click-through triplets indicating that a user submitted a query and that the user selected a document from the results of the query. To overcome the large size and sparseness of the click-through data, the search system when presented with an input triplet comprising a user, a query, and a document determines a probability that the user will find the input document important by smoothing the click-through triplets. The search system then orders documents of the result based on the probability of their importance to the input user.

Abstract: A method and system for ranking content and providing a user interface for viewing the content is provided. The content system ranks content in a search result based on a combination of relevance of the content to a query and quality of the content. The content system may derive the quality of content by analyzing ratings provided by various content forums. The content system may use metadata provided by a content forum when searching for content that matches a query. The content system generates a rank score that combines the relevance and quality of the content and ranks the content according to the rank scores.

Abstract: A method and system for labeling object information of an information page is provided. A labeling system identifies an object record of an information page based on the labeling of object elements within an object record and labels object elements based on the identification of an object record that contains the object elements. To identify the records and label the elements, the labeling system generates a hierarchical representation of blocks of an information page. The labeling system identifies records and elements within the records by propagating probability-related information of record labels and element labels through the hierarchy of the blocks. The labeling system generates a feature vector for each block to represent the block and calculates a probability of a label for a block being correct based on a score derived from the feature vectors associated with related blocks. The labeling system searches for the labeling of records and elements that has the highest probability of being correct.

Abstract: A method and system for detecting events based on query-page relationships is provided. The event detection system detects events by analyzing occurrences of query-page pairs generated from a user selecting the page of the pair from a search result for the query of the pair. The event detection system may identify semantic and temporal similarity between query-page pairs. The event detection system then identifies clusters of query-page pairs that are semantically and temporally similar.

Abstract: A system for automatically classifying an object of a target image is provided. A classification system provides a collection of classified images along with a classification of the dominant object of the image. The classification system attempts to classify the object of a target image based on similarity of the target image to the classified images. To classify a target image, the classification system identifies the classified images of the collection that are most similar to the target image based on similarity between salient points of the target image and the classified images. The classification system selects a classification associated with the classified images that are most similar to the target image as a classification for the object of the target image.

Abstract: A content service network for providing content-oriented services over the Internet or similar networks comprises a service delivery overlay having a plurality of application servers and a content delivery network overlay having a plurality of service-enabled proxies in content delivery paths between content providers and content consumers. The service delivery overlay and the content delivery network collaborate to provide content-oriented processing, such as adaptive video delivery, content personalization, language translation, etc. The content service network accepts subscriptions from content providers and content consumer. For each subscription, service instructions including service binding data for binding the subscribed services with the subscriber's identity are generated and distributed to the service-enabled proxies.

Abstract: A method and system for generating a detector to detect a dominant color of an image is provided. A dominant color system trains a detector to classify colors as being dominant colors of images. The dominant color system trains the detector using a collection of training images. To train the detector, the dominant color system first identifies candidate dominant colors of the training images. The dominant color system then extracts features of the candidate dominant colors. The dominant color system also inputs an indication of dominance of each of the candidate dominant colors. The dominant color system then trains a detector to detect the dominant color of images using the extracted features and indications of dominance of the candidate dominant colors as training data.

Abstract: A method and system for distributed training of a hierarchical classifier for classifying documents using a classification hierarchy is provided. A training system provides training data that includes the documents and classifications of the documents within the classification hierarchy. The training system distributes the training of the classifiers of the hierarchical classifier to various agents so that the classifiers can be trained in parallel. For each classifier, the training system identifies an agent that is to train the classifier. Each agent then trains its classifiers.

Abstract: A method and system for generating a ranking function using a fidelity-based loss between a target probability and a model probability for a pair of documents is provided. A fidelity ranking system generates a fidelity ranking function that ranks the relevance of documents to queries. The fidelity ranking system operates to minimize a fidelity loss between pairs of documents of training data. The fidelity loss may be derived from “fidelity” as used in the field of quantum physics. The fidelity ranking system may use a learning technique in conjunction with a fidelity loss when generating the ranking function. After the fidelity ranking system generates the fidelity ranking function, it uses the fidelity ranking function to rank the relevance of documents to queries.