Abstract: Various embodiments enable a device to perform tasks such as processing an image to recognize and locate text in the image, and providing the recognized text an application executing on the device for performing a function (e.g., calling a number, opening an internet browser, etc.) associated with the recognized text. In at least one embodiment, processing the image includes substantially simultaneously or concurrently processing the image with at least two recognition engines, such as at least two optical character recognition (OCR) engines, running in a multithreaded mode. In at least one embodiment, the recognition engines can be tuned so that their respective processing speeds are roughly the same. Utilizing multiple recognition engines enables processing latency to be close to that of using only one recognition engine.

Abstract: In one embodiment, a method comprises, by a computer system, receiving a search term, determining a search constraint for a search query, and generating a search query. The search query includes the search term and the search constraint. The search query requires that at least a first number of search results that are returned in response to the search query match both the search term and the search constraint. The search query permits at least a second number of the search results to match only the search term. The computer system provides the search query for a search against a store of data.

Abstract: The disclosure includes a system and method for providing recommendation items to users. The system includes a processor and a memory storing instructions that when executed cause the system to: receive identification input data associated with a user; identify the user using the identification input data; receive sensor data; determine a current situation associated with the user from one or more predefined situations described by predefined situation data and associated parameters; receive data describing a set of candidate items; generate a set of efficacy scores for the set of candidate items; eliminate one or more candidate items from the set of candidate items to obtain one or more remaining candidate items; rank the one or more remaining candidate items based on one or more associated efficacy scores; and provide one or more recommendation items to the user from the one or more ranked remaining candidate items.

Abstract: Systems and methods for generating dynamic hierarchical facets from business intelligence artifacts are described herein. In accordance with one aspect of the present disclosure, grouping information is extracted from one or more business intelligence artifacts or a semantic layer. Facets or categories are dynamically linked based on the extracted grouping information to generate one or more hierarchical facets. The hierarchical facets may be incorporated in search results to facilitate ease of navigation.

Abstract: Methods, systems, and techniques for keyword management are described. Some embodiments provide a keyword management system (“KMS”) configured to determine the effectiveness of multiple candidate keywords. In some embodiments, the KMS generates multiple candidate keywords based on an initial keyword. The KMS may then determine an effectiveness score for each of the candidate keywords, based on marketing information about those keywords. Next, the KMS may process the candidate keywords according to the determined effectiveness scores. In some embodiments, processing the candidate keywords includes applying rules that conditionally perform actions with respect to the candidate keywords, such as modifying advertising expenditures, modifying content, or the like.

Abstract: A system ranks results. The system may receive a list of links. The system may identify a source with which each of the links is associated and rank the list of links based at least in part on a quality of the identified sources.

Abstract: Method, system, and programs for realtime de-duplication of objects. A received object is hashed to generate a hashed object, which is then used to generate a query for an inverted index. Candidate matching objects are determined based on the query of the inverted index. From the candidate matching objects, a matched object that corresponds to the received object is determined.

Abstract: A system and method for providing at least one search result responsive to receiving a search query includes receiving the search query, locating at least one record containing at least one search query term, and displaying a text representation of each of the records containing at least one search query term. The invention further includes displaying a representation of an object in relation with the text representation for each record that includes an object comprising at least one predetermined attribute. In another embodiment, the invention provides a method for indexing records in an index of an information network, including searching the record for an indication that an object comprising a specified attribute is to be displayed with the record, creating a representation of the object; storing the representation in association with the record to which it corresponds, and making an entry for the record in the index.

Abstract: Techniques are disclosed for selecting an item to recommend to a user from a plurality of items. A first set of scores indicating degrees of association between items of the plurality of items may be calculated using a first scoring algorithm having a first scale, and a second set of scores indicating degrees of association between items of the plurality of items may be calculated using a second scoring algorithm having a second scale. The second set of scores may be adjusted to match the second scale to the first scale, and a third set of scores may be calculated for the plurality of items by combining the first set of scores and the adjusted second set of scores. An item may be selected from the plurality of items to recommend to the user based on the third set of scores.

Abstract: A computer-implemented method for ranking search results using an entity network includes: receiving an entity search query submitted for performance on an entity network, the entity search query including at least one term and indicating a query submitter, the entity network comprising entities associated with each other by relations; identifying, based on the received search query and using one or more processors, all entities in the entity network that are responsive to the at least one term; ranking the identified entities, based on relevance with regard to the query submitter and using one or more processors, the ranking performed using proximity in the entity network; and presenting the identified entities in response to the entity search query, the entities presented in an order corresponding to the ranking.

Abstract: Systems, computer program products, apparatus, and methods are described that perform operations including receiving a search query that includes a name, receiving multiple resources that have been identified by a search engine as best satisfying the search query, wherein the identified multiple resources include a resource including a plurality of images. The operations include identifying an image of the plurality of images displaying a face of the person. The image is identified based on a description associated with the image. The description is based at least in part on one or more resources included in the search results. The operations further include providing the identified image with the search results. The search results are provided as a plurality of links. Each link identifies a corresponding resource of the identified plurality of resources.

Abstract: A computer-implemented system and process for generating a relationship network is disclosed. The system provides a set of data items to be related and generates variable length data vectors to represent the relationships between the terms within each data item. The system can be used to generate a relationship network for documents, images, or any other type of file. This relationship network can then be queried to discover the relationships between terms within the set of data items.

Abstract: In one exemplary aspect, a method includes the step of obtaining a content of a content block. The content is represented as a content vector. A query is received. The query is represented as a query vector. A hierarchical sliding similarity and dissimilarity is determined for matching the content vector and the query vector, this step can include the steps of: determining a similarity measure and a dissimilarity measure for each content vector element with respect to the query vector, identifying a strong match over a sliding window of sub-terms of each content vector element; computing a sub-similarity score and a sub-dissimilarity score for each level of the convent vector element; determining a final similarity score as a combination of the strong match of some sub-vectors at different levels; and determining a final dissimilarity score as a combination of the strong match of some sub-vectors at different levels.

Abstract: A web-based application provides more accurate and clearer methods of searching, sorting, and displaying a set of images stored in a database. A first aspect of the present invention is the method by which image data is stored. Typical, content-based systems use colour information, whereas the present invention uses an image-location tagging method. A second aspect of the present invention is the method by which the set of images are sorted in relevancy. Tag data of the images allows for a new and last method of searching through an entire set. A third aspect of the present invention is the method by which the sorted images are displayed to the user. Instead of the common method of just displaying the images in a rectangular array, where each image is the same size, the web-based application positions and sizes each image based on how relevant it is.

Abstract: A computer processor accesses QA forum activity associated with users of a QA forum, wherein the QA forum activity contains historical questions, historical answers, and profiles, each having elements including QA forum information. The processor accesses historical online social activity corresponding to the users of the QA forum, each historical online social activity having elements including at least social activity information. The processor performs a first similarity calculation and selects an at least one model answerer based on the similarity between elements of a question on the QA forum, and the elements of the QA forum activity associated and corresponding to each of the users of the QA forum. Based on a second similarity calculation, the processor determines a similarity between the QA forum user elements and the at least one model answerer elements, and identifies at least one prospect answerer from the QA forum users.

Abstract: System and methods performed by a server for determining weighted influence in social networks, including: determining posts related to the topic; characterizing each post as one or more of: a reply post, a mention post, and a re-posting; generating a group of user accounts comprising any user account that authored a posting to which is replied, being mentioned in the mention post, that posted content being re-posted, and/or that authored one or more posts that are related to the topic; representing each of the user accounts as a node in the group in a connected graph and establishing an edge between one or more pair of nodes when there is a follower-followee relationship between the nodes; and for each edge between nodes, determining a weighting that is a function of one or more of: a number of mention posts, a number of reply posts, and a number of re-posts.

Abstract: A system that facilitates ranking search results returned by a search engine in response to receipt of a query is described herein. The system includes a receiver component that receives categorical metadata pertaining to an item and categorical metadata pertaining to the query and a computation component that computes at least one of a document feature pertaining to the item, a query feature pertaining to the query, or a document-query feature pertaining to the item and the query based at least in part upon one or more of the categorical metadata pertaining to the item or the categorical metadata pertaining to the query. The system also includes a ranker component that selectively places the item in a particular location in a sequence of items based at least in part upon the at least one of the document feature, the query feature, or the document-query feature.

Abstract: Systems and methods for creating image-based content based on text-based content. A data processing system receives a text-based content item based on which an image-based content item is to be created. The data processing system determines a context of the text-based content item based on the content of the text-based content item and the content of a landing page associated with the text-based content item. The data processing system determines one or more search terms from the determined context of the text-based content item. The data processing system then identifies from an image database, one or more candidate images that match at least one of the search terms determined from the context of the text-based content item. The data processing system then creates an image-based content item based on the text-based content item using at least one of the candidate images.

Abstract: To provide a contents creating device and a contents creating method capable of distributing contents having an article arrangement with high relativity to a user as distribution destination. A contents creating device 100 to create contents to be distributed within a predetermined group, including a material evaluation unit 130 configured to evaluate relativity of material data of contents to a user as contents distribution destination, and a contents editing unit 140 configured to determine an article arrangement within contents according to an evaluation, wherein contents based on the article arrangement are created. Due to this, it is possible to distribute contents having an article arrangement with high relativity to a user of the distribution destination and it is possible for the user as distribution destination to browse an article in which the user is highly interested.

Abstract: Systems and methods for enabling communication between a user and at least one entity by a communication service are disclosed. In one example, the method comprises acts of receiving a search for the at least one entity, analyzing the search, by a processor, to determine search results based on the search including the at least one entity, receiving a message from the user to the at least one entity, the message including inquiry information directed to the at least one entity, determining, by the processor, presence status associated with the at least one entity, establishing, by the processor, a communication channel between the user and the at least one entity based on the presence status, and transmitting a response to the message based on the inquiry information via the communication channel.

Abstract: The present application provides a method for content ranking for recommending content to a community of users. Recommending content to a community of users is provided by identifying data sources associated with a user of the recommendation system. Each data source comprises content items. Hyperlinks embedded in the content items from the data sources associated with the user are extracted. The hyperlinks and a set of user preferences are used to rank new time-sensitive content items associated with a plurality of data sources. A set of ranked time sensitive customized content items are presented to the user as recommended content.

Abstract: This invention relates to an advertisement machine which provides advertisements to a user searching for desired information within a data network. The machine receives, from a user, a search request including a search argument corresponding to the desired information and searches, based upon the received search argument, a first database having data network related information to generate search results. It also correlates the received search argument to a particular advertisement in a second database having advertisement related information. The search results together with the particular advertisement are provided by the machine to the user.

Abstract: There is provided a computer-implemented method and system for ranking documents. The method includes identifying a number of query-document pairs based on clickthrough data for a number of documents. The method also includes building a latent semantic model based on the query-document pairs and ranking the documents for a search based on the latent semantic model.

Abstract: Methods and apparatus to methods and apparatus to identify images in print advertisements are disclosed. An example method comprises computing a first image feature vector for a first presented image, comparing the first image feature vector to a second image feature vector, and when the first image feature vector matches the second image feature vector, storing printed-media information associated with the first presented image in a database record associated with the second image feature vector.

Abstract: Determining ranked candidate media in response to media query data corresponding to a query media includes receiving the media query data including feature data of the query media, coordinate data, and boundary data, matching the features with corresponding features of an media database using the feature data to identify features in the media database within a predetermined hamming distance in a hash table from the corresponding features of the query media to obtain matched features in the media database, determining candidate media whose number of matched features exceeds a matched feature threshold, generating a geometry similarity score between the query media and each candidate media using the feature data and the coordinate data, generating a boundary similarity score between the query media and each candidate media using the boundary data, ranking the candidate media based on the numbers of matched features, the geometry similarity scores and the boundary similarity scores.

Abstract: A method, system and computer program product for finding a diversified ranking list for a given query. In one embodiment, a multitude of date items responsive to the query are identified, a marginal score is established for each data item; and a set, or ranking list, of the data items is formed based on these scores. This ranking list is formed by forming an initial set, and one or more data items are added to the ranking list based on the marginal scores of the data items. In one embodiment, each of the data items has a measured relevance and a measured diversity value, and the marginal scores for the data items are based on the measured relevance and the measured diversity values of the data items.

Abstract: Systems, methods, and computer-readable media are provided for ranking documents. An exemplary method includes receiving a plurality of queries from a user that are related to a search topic. The queries are compared to a plurality of documents stored in a database to identify candidate documents that match one or more of the queries. The number of matching queries is determined for the candidate documents. The candidate documents are ranked, using a computer processor, for relevance to the search topic based at least on the number of matching queries for each respective candidate document.

Abstract: A computer-implemented method of determining the relevance, to a user, of documents accessible over a computer network involves analyzing information relating to a first document, identifying a second document corresponding to the first document using a computer, obtaining information that indicates the relevance of a second document, determining the relevance of the first document using the information indicating relevance of the second document. The first document may be a mobile document, while the second may be a non-mobile document.

Abstract: Two methods for measuring keyword-document relevance are described. The methods receive a keyword and a document as input and output a probability value for the keyword. The first method is a similarity-based approach which uses techniques for measuring similarity between two short-text segments to measure relevance between the keyword and the document. The second method is a regression-based approach based on an assumption that if an out-of-document phrase (the keyword) is semantically similar to an in-document phrase, then relevance scores of the in and out-of document phrases should be close to each other.

Abstract: A forensic system includes a result information receiving unit that receives result information which is a determination result of connection between a lawsuit and a document group including a predetermined number of documents, which is extracted from document data included in digital information, by a user, an element selection unit that calculates evaluation values of elements which commonly appear in the document group in each result information item from the characteristics of the elements and selects the elements on the basis of the evaluation values, a score calculation unit that calculates a score of each document in the document data from the selected elements included in each document of the document data and the evaluation values of the selected elements, and a recall ratio calculation unit that calculates a recall ratio related to the determination of the connection to the lawsuit on the basis of the score.

Abstract: Techniques are described for displaying help articles, for a web page, that are related to the context of the web page. A help article provider has access to a database of help articles, each of which are associated with target-context data that may include content identifiers for the help article and/or user profile attributes for users that the help article targets. The help article provider identifies content identifiers for a viewed web page based on one or more of: metadata and a URL for the web page. The help article provider searches the help article database using the identified content identifiers. The help article provider may filter the help articles based on an authenticated user's attributes. An ordered list of the identified help articles is displayed, with the most relevant or important help articles displayed at or near the top of the list.

Abstract: Various embodiments described herein provide users with a fast and efficient way for identifying scenes for editing purposes. At least one embodiment is a method for searching media content based on identification data embedded in an editing file. The method comprises receiving the editing file and extracting identification data from the editing file, wherein the identification data is uniquely associated with particular media content. The method further comprises searching one or a plurality of media content based on the identification data and identifying matching media content, performing a secure download of the identified media content, and performing playback of the media content based on special effects specified in the editing file, wherein the media content is unmodified during playback.

Abstract: Media content is enhanced and/or categorized through association with descriptive terms of a nomenclature that are obtained from network sources of information. In one example, a search query identifying a target media content item is received, and a search is performed based on the search query to obtain search result information for the target media content item. A schema defining a set of descriptive fields and an associated nomenclature of terms for each of the descriptive fields is referenced with regards to the search result information. The search result information is processed to identify a sampling metric for instances of the nomenclature of terms that are contained within text information of the search result for the descriptive fields. One or more suggested terms that have been selected from the nomenclature of terms for the descriptive fields are output for the target media content item based on the sampling metric.

Abstract: To increase the likelihood of a user viewing content items previously identified for presentation but not viewed by the user, a social networking system accesses a view state associated with the user that identifies content items provided to the user and provided content items viewed by the user. Based on the view state, content items previously provided to the user but not viewed by the user are identified. Additional content items whose state has changed since presentation of the content items to the user are also identified. Content items previously presented to the user but not viewed with at least a threshold relevance to the user are ranked along with the additional content items. Based on the ranking, content items are selected for presentation to the user.

Abstract: A method and apparatus for dynamically adjusting the user interface of a search engine in order to effectively communicate the improved relevancy achieved through real-time implicit re-ranking of search results is described. Real-time implicit re-ranking occurs without delay after every user action as the search is being conducted, so finding methods of immediately altering the search page without disrupting the user experience is important. Graphical icons next to search results are employed to enable generating and removing re-ranked results, referred to as “recommended” search results. Clusters based on the real-time user model are also displayed to facilitate query reformulations. Sponsored links are selected using the real-time user model along with a combination of RPC and CTR information and are displayed in a manner similar to the organic results, or used to replace the initial sponsored links altogether.

Abstract: Techniques and tools are described for augmenting search using association information. Searches can be performed using a combination of index information and association information. In some examples, index information is stored in a first data store and association information is stored in a second data store. Search queries can be received and modified using association information. Modified search queries can be executed using a combination of index information and association information. Index information can be generated by indexing a set of documents. Association information can be generated by monitoring user activity occurring between users and a set of documents.

Abstract: A system and method are disclosed for modifying a document format. In one embodiment, a structure of a first document is extracted to form a first data structure, including multiple subdocuments, each subdocument having a label assigned thereto. A table of contents listing the labels of the subdocuments is then generated. The various labels are then ordered according to the amount of text of the associated subdocument, a comparison of the text of a previous link and the text of the associated subdocument, a comparison of the text of the previous document and the text of the associated subdocument, a comparison of the text of the previous link and the text of the associated subdocument, or a combination of these.

Abstract: Techniques for providing electronic content may include: identifying members of a social network that have a characteristic in common; determining a first proportion of the social network that the members compose; determining that a first member of the social network has the characteristic; identifying first content from the members that have the characteristic in common; and outputting second content from the social network to the first member. At least some of the first content may be included in the second content, and the at least some of the first content may represent a second proportion of the second content, where the second proportion is based on the first proportion.

Abstract: Disclosed are a method for providing a time machine service based on an SNS (social network service), including: extracting keywords based on SNS information input by SNS users and collecting rich contents relating to the SNS information based on the keywords; dividing points of past, present, and future, based on a predetermined present time period and classifying the rich contents into any one of past information, present information, and future information, based on the division result; and providing the classified rich contents to the SNS users.

Abstract: A computer-implemented method includes receiving, at a computer system, a question; identifying one or more first semantic elements in the question; selecting, from one or more electronic documents, a plurality of candidate responses to the question based on comparison of the one or more first semantic elements to second semantic elements; determining completeness scores for the plurality of candidate responses, wherein each of the completeness scores indicates how completely a corresponding candidate response from the plurality of candidate responses answers the question; determining relevance scores for the plurality of candidate responses, wherein each of the relevance scores indicates how relevant a corresponding candidate response from the plurality of candidate responses is to the question; and providing, by the computer system, at least a portion of the plurality of candidate responses based, at least in part, on the completeness scores and the relevance scores.

Abstract: Embodiments are directed to providing access to supplemental query data that is relevant to a data query. In one scenario, a computer system receives an input indicating that a data query is to be processed, where the input provides various search criteria. The computer system then generates the data query according to the search criteria in the received input. The computer system further sends the data query to a second computer system which is configured to process the data query and receives the results of the data query. The results indicate that additional data elements relevant to the search criteria have been identified. The computer system then displays an indication of various steps that are to be taken in order to access the identified additional data elements.

Abstract: Some embodiments of on-the-fly pattern recognition with configurable bounds have been presented. In one embodiment, a pattern matching engine is configured based on user input, which may include values of one or more user configurable bounds on searching. Then the configured pattern matching engine is used to search for a set of features in an incoming string. A set of scores is updated based on the presence of any of the features in the string while searching for the features. Each score may indicate a likelihood of the content of the string being in a category. The search is terminated if the end of the string is reached or if the user configurable bounds are met. After terminating the search, the scores are output.

Abstract: Architecture that enables an optional display of a longer version of each subsnippet in response to user interactions such as clicking, hovering, or other suitable form of interaction. More specifically, options are provided to display additional text from a search result at the point where a subsnippet (a subsegment in a snippet that is delimited by ellipses) ends. Selecting suitable boundaries for both initial subsnippets and expanded subsnippets enables relevant information to be presented and increased readability.

Abstract: A user interface for searching records in a database is provided. Each record may contain information regarding one or more categories. The user interface may receive search criteria of one or more categories and provide search results comprising exact and/or inexact matching records. An inexact matching record may have a relevance value representing the relevance of the record to the search query, and inexact matching records may be displayed in an ordering based on the relevance values. The user interface may provide slider tools for receiving changes in a weight value or search criterion of an associated category. A new ordering of the inexact matching records may be displayed in real-time based on newly received weight values or search criterion. The user interface may also provide a mapping tool, a search-result selection function, and/or one or more drill down tools. The user interface may provide any tool or function alone or in any combination.

Abstract: A machine may be configured to access an image of an item described by a description of the item. The machine may determine an image quality score of the image based on an analysis of the image. A request for search results that pertain to the description may be received by the machine, and the machine may present a search result that references the item's image, based on its image quality score. Also, the machine may access images of items and descriptions of items and generate a set of most frequent text tokens included in the item descriptions. The machine may identify an image feature exhibited by an item's image and determine that a text token from the corresponding item description matches one of the most frequent text tokens. A data structure may be generated by the machine to correlate the identified image feature with the text token.

Abstract: In one embodiment, a system includes one or more computing systems that implement a social networking environment and are operable to access stored information including a plurality of nodes including a first set of user nodes that each correspond to a respective user and a second set of concept nodes that each correspond to a respective concept. The system may generate a match coefficient for the user and concept, representing the degree of relevance of a particular concept node to a particular user node.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for identifying one or more second documents related to one or more first documents. Strength of relationship scores between candidate documents in a group of candidate documents and each first document are determined by aggregating user selection data for users, the user selection data indicating, for each user, whether the user viewed the candidate document during a window of time after the first document is presented to the user on a search results web page in response to a query. An aggregate strength of relationship score is calculated for each candidate document from the strength of relationship scores for the candidate document. Second documents are selected from the candidate documents according to the aggregate strength of relationship scores for the candidate documents.

Abstract: An “Interactive Word Lattice” provides a user interface for interacting with and selecting user-modifiable paths through a lattice-based representation of alternative suggested text segments in response to a user's text segment input, such as phrases, sentences, paragraphs, entire documents, etc. More specifically, the user input is provided to a trained paraphrase generation model that returns a plurality of alternative text segments having the same or similar meaning as the original user input. An interactive graphical lattice-based representation of the alternative text segments is then presented to the user. One or more words of each alternative text segment represents a “node” of the lattice, while each connection between nodes represents a lattice “edge. Both nodes and edges are user modifiable. Each possible path through the lattice corresponds to a different alternative text segment. Users select a path through the lattice to select an alternative text to the original input.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for identifying one or more second documents related to one or more first documents. Strength of relationship scores between candidate documents in a group of candidate documents and each first document are determined by aggregating user selection data for users, the user selection data indicating, for each user, whether the user viewed the candidate document during a window of time after the first document is presented to the user on a search results web page in response to a query. An aggregate strength of relationship score is calculated for each candidate document from the strength of relationship scores for the candidate document. Second documents are selected from the candidate documents according to the aggregate strength of relationship scores for the candidate documents.

Abstract: A computer-implemented method is provided for enabling Internet users to interact with a graphical search interface by: generating a multi-dimensional search space having a plurality of vertices, based on the initial and related search parameters, wherein each vertex is populated with and represents one of the initial or related search parameters; generating a matrix of hyperlinks within the multi-dimensional search space, each hyperlink having coordinates within the search space defining a relevance of the hyperlink to each of the initial and related search parameters; displaying a graphical interface projecting the multi-dimensional search space, along with user elements by which the user may adjust a relative weight of one or more of the initial and related search parameters by interacting with one or more of the hyperlinks; and displaying a plurality of search results automatically updated in real time based on the user-adjusted weights of the initial and related search parameters.