Abstract: In a general aspect, an approach to query specification includes processing a query by identifying one or more terms and validating the terms using a first corpus of media elements. The result of the validation is used to form a modified query. In some examples, identifying the one or more terms is based on a second corpus of media elements, which may include a different type of media than the first corpus. In some examples, the validating of the terms includes accepting input from a user according to parts of the elements of the first corpus presented to the user.

Abstract: A method of determining structured product information for a product from a product description using a product entity graph. The product graph can include a plurality of nodes. Each of the plurality of nodes can include an entity value key, one or more entity names, and an entity name count for each of the one or more entity names. The method can include determining k-grams of the product description. The method also can include, for each k-gram of the product description, determining a matching node of the plurality of nodes of the product entity graph that corresponds to the k-gram and determining a derived entity name for the product from the one or more entity names of the matching node based at least in part on the entity name counts corresponding to the one or more entity names. Other embodiments of related systems and methods are also disclosed.

Abstract: A device for determining interest includes a storage portion configured to store, on a user-by-user basis, a co-occurrence frequency in correlation with a user, the co-occurrence frequency indicating how many times a pair of words is used in a same cluster of a first document, on a pair-by-pair basis, to which the user gained access previously; a designating portion configured to allow a person who is to conduct a search to designate a second document and any one of the users; and a determination portion configured to determine that, among the pairs, a pair which is used in a same cluster of the designated second document and which also satisfies a predetermined condition of the co-occurrence frequency corresponding to the designated user is a particular pair in the second document which is probably of high interest to the designated user.

Abstract: Embodiments of a search stemming module at a computer configured to receive a query and identify stems from the query and configured to search a secondary index for variants corresponding to the stems are disclosed. The secondary index may comprise one or more lists of stems associated with variants from a primary index; a search reconfiguration module configured to reformat the query to include variants found from the secondary index; and a search engine configured to implement a search of the primary index using the variants received from the search reconfiguration module.

Abstract: Method and a recommendation system (200) for making a media item recommendation to a user. A weighted graph is created (2:2) that comprises item nodes representing media items and edges linking the item nodes. Weights are also assigned to the edges based on weights of contributors shared by the linked item nodes. When an initial media item of interest to the user is identified (2:4), the recommendation system determines (2:5) a set of media items which are linked to the initial media item in a weighted graph. At least one of the determined set of media items is also selected based on the assigned weight of edges linking the media items to the initial media item. The recommendation system then provides (2:6) the media item recommendation with the at least one selected media item to the user.

Abstract: In embodiments of the disclosed technology, indexes, such as inverted indexes, are updated only as necessary to guarantee answer precision within predefined thresholds which are determined with little cost in comparison to the updates of the indexes themselves. With the present technology, a batch of daily updates can be processed in a matter of minutes, rather than a few hours for rebuilding an index, and a query may be answered with assurances that the results are accurate or within a threshold of accuracy.

Abstract: A computer system identifies data representing interest of users in a plurality of geographic sites, such as images of geographic sites submitted by people to public photo sharing sites. The system generates a “heat map” representing densities of the interest data. The system generates a route from a starting location to an ending location. The route includes at least one of the plurality of geographic sites, even if the route is not the maximally efficient route from the starting location to the ending location (measured, e.g., in terms of the total time required to travel the route or the total distance of the route). The system favors high-temperature sites for inclusion in the route, but ensures that the route does not deviate too far from the route having the shortest possible distance between the starting location and the ending location at the expense of including high-temperature sites.

Abstract: Creating training data for a natural language processing system may comprise obtaining natural language input, the natural language input annotated with one or more important phrases; and generating training instances comprising a syntactic parse tree of nodes representing elements of the natural language input augmented with the annotated important phrases. In another aspect, a classifier may be trained based on the generated training instances. The classifier may be used to predict one or more potential important phrases in a query.

Abstract: A method of disambiguating user intent in conversational interactions for information retrieval is disclosed. The method includes providing access to a set of content items with metadata describing the content items and providing access to structural knowledge showing semantic relationships and links among the content items. The method further includes providing a user preference signature, receiving a first input from the user that is intended by the user to identify at least one desired content item, and determining an ambiguity index of the first input. If the ambiguity index is high, the method determines a query input based on the first input and at least one of the structural knowledge, the user preference signature, a location of the user, and the time of the first input and selects a content item based on comparing the query input and the metadata associated with the content item.

Abstract: A system and method for dynamically generating a web page is disclosed. The web page includes a frame having content cells. Content for the content cells is provided based in part on business rules. A business rules engine processes, in real-time, an online interaction and determines appropriate content for the content cells. The business rules engine outputs a pointer to the content which is used to populate the content cell.

Abstract: A content program information search system and method is described. The system comprises a digital signal receiver for detecting and displaying a content signal and/or detailed content information for one or more search terms from a digital signal transmitted from a transmitter; and a server having a database for storing search terms inputted from external devices, the inputted search terms being stored by order of priority based for example on search frequency, such that the server extracts and transmits to the transmitter search terms from the first database, and the digital signal receiver displays according to the order of priority at least one search term transmitted from the transmitter. The viewer thus has search terms of high search frequency to enable the viewer to rapidly and easily search for information on a desired content program.

Abstract: A system and method adapted to facilitate efficient sorting and displaying of data in a table. An example method includes displaying a table with a first portion of data and a second portion of data; simultaneously applying a first sort rule to the first portion of data (primary database object(s)) and a second sort rule to the second portion of data (secondary database object(s)) in response to user selection of a sort option; and illustrating sorted data via the table, wherein the first sort rule is adapted to cause the first portion of data to appear in one or more predetermined locations in the table relative to the second portion of data. The second portion of data is sorted according to the second sort rule. In a more specific embodiment, primary database objects are sorted via the first sort rule and are not subject to the second sort rule.

Abstract: Disclosed are various approaches for searching for electronic information. In one embodiment, a first plurality of objects is presenting for user selection. The first plurality of objects is a first subset of a second plurality of objects that is organized by a plurality of containment categories in a data structure. A plurality of associations is stored, and each association defines a relationship between one of, two different ones of the plurality of categories, or two objects in two different ones of the plurality of containment categories. In response to user selection of one of the first plurality of objects, a second subset of the second plurality of objects is selected. Selection of the objects in the second subset is in part a function of respective weighted relevance values for associations of the selected object to each of the second plurality of objects. The second subset of objects is output.

Abstract: Profile information is maintained concerning users and files in the context of a file sharing and collaboration environment. Profile information quantifies each user's interest level in specific files, and levels of similarity between users and between files. Machine learning techniques are applied to monitored actions taken by users directed towards files, and profile information is dynamically updated in response. Natural language processing such as n-gram analysis is applied to files, and file similarity levels are increased in response to requisite amounts of common content. The event notification stream is filtered. For each specific event notification, a relevance value is quantified for each specific user, based on profile information concerning the specific user, the file to which the event is directed, and the user who undertook the event. The corresponding notification is only transmitted to specific users for whom the relevance value exceeds a predetermined threshold.

Abstract: Disclosed herein are systems, methods, and non-transitory computer-readable storage media for approximating responses to a user speech query in voice-enabled search based on metadata that include demographic features of the speaker. A system practicing the method recognizes received speech from a speaker to generate recognized speech, identifies metadata about the speaker from the received speech, and feeds the recognized speech and the metadata to a question-answering engine. Identifying the metadata about the speaker is based on voice characteristics of the received speech. The demographic features can include age, gender, socio-economic group, nationality, and/or region. The metadata identified about the speaker from the received speech can be combined with or override self-reported speaker demographic information.

Abstract: A computer-implemented method, program, and system for identifying non-self-descriptive terms in electronic documents. The computer-implemented method for identifying a non-self-descriptive term in an electronic document, includes a memory and a processor communicatively coupled to the memory and configured to execute the steps of a method. The method includes acquiring a noun included in the corpus data. The method further includes calculating a qualifying level and a qualified level in the corpus data related to each known in the corpus data. The method further includes identifying one or more nouns included in the corpus data as having a qualifying level and/or qualified level satisfying a predetermined condition. The method further includes presenting a term related to one or more of the nouns in the electronic document as a candidate for the non-self-descriptive term in the electronic document.

Abstract: User content access events pertaining to a content item, such as an ebook, audio, video file, and so on, are collected and analyzed to determine progress data, including abandonment information about when the content item, or a portion thereof, has been abandoned. Once determined, recommendations may be presented based on the progress data from similar users.

Abstract: Digital goods libraries are compared via their datasets, which include organizational data and history data. Potential comparison results include a shared multiple natural languages presence, shared genre frequency change, shared artist frequency change, shared digital good frequency change, shared outlier presence, shared similar songs presence, and/or shared recommendable goods presence. Proper subsets of datasets may be specified by playlist, genre, artist, category, date, play frequency, and/or play count, and then compared. Comparison results may be visualized, such as playlist content over time, play frequency over time, and/or play count over time. Some results include printed or spoken text such as an artist name, actor name, author name, genre name, and/or review content. Control over disclosure of library owner identity is provided. In some cases a dataset is compared with multiple datasets in a group, e.g., to help assess compatibility between a prospective member and current members of the group.

Abstract: Computer services use group preferences, as partially or wholly specified by preference lists of individual members, to identify items well suited for a designated group of users. In one embodiment, items in the movie/video rental queue (one type of preference list) of a first user are prioritized based, at least partly, on the rental queue of a second, affiliated user, to give priority to items corresponding to the collective preferences of both users. Items may also be recommended to the users based on their collective preferences. In a second embodiment, the suggestion lists of multiple members of a book club are used, optionally in combination with other member preference information, to select or recommend book titles for the club. In a third embodiment, the wish lists of two or more affiliated users are used to assist others in purchasing gifts that correspond to their collective preferences.

Abstract: A system and method for using a combination of semantic and statistical processing of input strings or other data content, such as a web page or an electronic document. In accordance with an embodiment, the system enables the injection of semantics into an otherwise statistically-based environment, by recognizing that, within various topics, certain words, combinations of words, or phrases, herein referred to as keyphrases have different weights. Some keyphrases may be relatively unique within a particular topic, or have a relatively high weighting towards that topic; whereas other keyphrases may not be unique, or may have a relatively low rating toward that topic. In accordance with an embodiment, the system allows for characterization of both (a) “sufficient” and (b) “necessary” keyphrases. A keyphrase is considered sufficient for a particular topic when, if that keyphrase is found in the input string or data content, one is likely to be in that topic (but could be in another topic).

Abstract: A system for providing interactive program guide features and other features and information related to a specific user interest or programming category in niche hubs is provided. For example, a movie lovers niche hub may provide programming features such as television program listings for movies, video-on-demand listings for movies, pay-per-view listings for movies, web site links related to movies, movie-related merchandise, movie news groups, movie chat groups, movie e-mail clubs, movie contests, movie trivia questions, movie actor interviews, movie reviews, movie channel package ordering, etc.

Abstract: A processor-implemented method, system, and/or computer program product derives and utilizes a context object to generate a synthetic context-based object. A context object for a non-contextual data object is derived by contextually searching a document that contains multiple instances of the non-contextual data object. The non-contextual data object is associated with the derived context object to define a synthetic context-based object, where the non-contextual data object ambiguously relates to multiple subject-matters, and where the context object provides a context that identifies a specific subject-matter, from the multiple subject-matters, of the non-contextual data object. The synthetic context-based object is then associated with at least one specific data store, which includes data that is associated with data contained in the non-contextual data object and the context object.

Abstract: Disclosed is a service which obtains media directly from users and from online sources, which obtains events and anniversaries from online sources, which obtains location and date information associated with photographs, which dynamically provides users with a selection of automatically curated collections of photographs based on the then-current location of the user, based on and relevant to personal and publicly recognized anniversaries and holidays (with dates obtained directly from the users and from online sources), based on specific people or locations associated with dates, events, and anniversaries, and which presents intelligently organized location-based collections which can be quickly re-organized by a user.

Abstract: There is provided a contents' relationship visualizing apparatus including a usage history acquiring unit for acquiring usage histories of multiple contents, a users' set specifying unit for specifying a content's users' set indicating users utilizing each content by use of the contents' usage histories, a similarity calculating unit for calculating a similarity between the specified users' sets by use of a similarity index, and a network creating unit for creating a network indicating a relationship between the contents by use of the calculated similarity between the users' sets.

Abstract: A method includes accessing text, identifying a plurality of terms from the text, determining a plurality of term vectors associated with the identified plurality of terms, and clustering the determined plurality of term vectors into a plurality of clusters, the plurality of clusters comprising a first and a second cluster, the first and second clusters each comprising two or more of the determined term vectors. The method further includes creating a first pseudo-document according to the first cluster, creating a second pseudo-document according to the second cluster, identifying a first set of terms associated with the first cluster using latent semantic analysis (LSA) of the first pseudo-document, identifying a second set of terms associated with the second cluster using LSA of the second pseudo-document, and combining the first and second sets of terms into a list of output terms.

Abstract: A system and method for generating feature vectors of documents in different languages are provided. The feature vectors provide scores associated with keywords defined in a base language for use by a profiler for generating or updating a user profile. The system and method use a plurality of keyword sets comprising: a base language keyword set comprising a plurality of base language keywords each associated with a respective identifier (ID); and a second language keyword set comprising a plurality of second language keywords each corresponding in meaning to a respective one of the base language keywords and associated with the ID of the corresponding base language keyword. One of a plurality of tokenizers is selected to parse a document based on the language of the document and to generate the feature vector using the keyword set of the corresponding language.

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: Systems and methods described herein can assign a confidence score to a match of unstructured descriptive information with structured reference information in a reference database. The systems and methods can take into account the structured nature of the reference information in assigning the score, thereby facilitating increased confidence in the match, and consequently, facilitating improved database organization and content identification.

Abstract: Aspects and implementations of the present disclosure are directed to systems and methods of selecting content for presentation to a viewing user. In general, in some implementations, a data processing system identifies a topic for a document requested by a user, determines a confidence score for a correlation between the topic and the document, determines a sentiment score for a sentiment towards the topic, and determines, for a candidate content item, a predicted acceptance score based on the confidence score, the sentiment score, and acceptance history data for the candidate content item. The data processing system selects a candidate content item for presentation to the user based, at least in part, on the predicted acceptance score.

Abstract: Exemplary embodiments provide systems, devices and methods that allow creation and management of lists of items in an integrated manner on an interactive graphical user interface. A user may speak a plurality of list items in a natural unbroken manner to provide an audio input stream into an audio input device. Exemplary embodiments may automatically process the audio input stream to convert the stream into a text output, and may process the text output into one or more n-grams that may be used as list items to populate a list on a user interface.

Abstract: A method includes accessing text, identifying a plurality of terms from the text, determining a plurality of term vectors associated with the identified plurality of terms, and clustering the determined plurality of term vectors into a plurality of clusters, the plurality of clusters comprising a first and a second cluster, the first and second clusters each comprising two or more of the determined term vectors. The method further includes creating a first pseudo-document according to the first cluster, creating a second pseudo-document according to the second cluster, identifying a first set of terms associated with the first cluster using latent semantic analysis (LSA) of the first pseudo-document, identifying a second set of terms associated with the second cluster using LSA of the second pseudo-document, and combining the first and second sets of terms into a list of output terms.

Abstract: Provided is a method of generating updating parameters. The method obtains search keywords used by users within a predetermined time period; counts the search keywords to obtain primary keywords, related keywords, co-search frequencies of each primary keyword and the respective related keywords being searched together, and search frequencies of the primary keywords being searched alone; computes first feature values based on the search frequencies of the primary keywords being searched alone; and then computes second feature values based on the first feature values and the co-search frequencies of the primary keywords and the respective related keywords. The second feature values serve as updating parameters for determining displaying modes of the related keywords. An apparatus of generating updating parameters, and a method and an apparatus of displaying related keywords according to the updating parameters are also provided.

Abstract: Visualizing social media conflict is provided. Textual messages by a set of human users connected via a network regarding a particular topic are collected. Active users in the set of human users authoring a number of textual messages regarding the particular topic more than a threshold number of textual messages are selected. Keywords are selected that occur more than a threshold number of times within the textual messages regarding the particular topic. A sentiment score is computed for each of the keywords occurring more than the threshold number of times within the textual messages using a keyword co-occurrence graph. A sentiment of each of the active users is determined based on the computed sentiment score of each of the selected keywords that are authored by a particular active user.

Abstract: Methods, systems, and apparatus, including computer program products are provided for ranking distinct book content items based on implicit links to other distinct book content items. The implicit links are defined based on the identification of matching features in the distinct book content items. In some implementations, the matching features are uncommon phrases in textual content of the distinct book content items. Edges representing implicit links are generated between distinct nodes representing distinct book content items in a weighted graph. Search results for distinct book content items can be ordered based on the edges connected to the distinct nodes in the weighted graph that represent the distinct book content items.

Abstract: A method for searching a database of digital media assets that has been indexed according to a set of general indexers. The method includes receiving a search query and defining specialized search conditions by identifying one or more elements of the search query corresponding to one or more specialized indexers. General search conditions are defined by identifying elements of the search query corresponding to the general indexers. A subset of the digital media assets are identified by applying the general search conditions, and the subset of the digital media assets is indexed using the identified specialized indexers. The subset of the digital media assets are ranked by applying the specialized search conditions.

Abstract: Systems and methods for identifying a named entity are described. In one described system, a computer program, such as a query system, identifies an event associated with an article, identifies a named entity within the event, and creates an implicit search query comprising the named entity.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for identifying central entities. In one aspect, a method includes obtaining candidate entities for a first resource; filtering a first entity graph whose nodes represent different entities found in a plurality of resources to remove nodes that do not correspond to a candidate entity, wherein pairs of nodes in the filtered first entity graph that are connected by an edge correspond to pairs of candidate entities that are associated with the same resource; generating a second entity graph for the first resource from the filtered first entity graph, wherein the second entity graph does not include nodes from the filtered first entity graph that are not connected to other nodes in the filtered first graph; and identifying candidate entities that are represented by nodes in the second entity graph as being central entities for the first resource.

Abstract: In accordance with various embodiments of the disclosed subject matter, methods, systems, and media for generating and prioritizing relevant content in real-time data systems are provided. In some embodiments, a method for recommending content for publishing is provided, the method comprising: receiving content from one or more content sources; determining one or more term entities associated with the content; calculating a score of the one or more terms based on a frequency, a recency, and a density at which the one or more term entities appear in one or more real-time information sources; and recommending a content item for publication to at least one of the one or more real-time information sources based on the score, wherein the content item comprises at least a portion of the received content.

Abstract: A topical search computer system identifies topics from various definitional (i.e., data) sources. The system generates a catalog of different topics from the data sources. Topics with similar names are differentiated by the system based on the context in which each topic is used. The context for a topic is represented by a context vector, which describes the co-occurrence relationships between the topic and other topics derived from the data sources. Because the system has computed a context for each topic, the system can provide improved search results responsive to user queries for information.

Abstract: A method and system to determine a category score of a social network member is described. An example system comprises a sample selector, a weight value module, a storing module, an access module, and a category score module. The sample selector selects a random sample of member profiles from the profiles maintained by an on-line social network system. The weight value module obtains respective weight values associated with various phrases present in the random sample of member profiles. The access module accesses a member profile and the weighted phrases associated with a certain category. The category score module determines a category score for the member profile based on a presence of one or more phrases from the plurality of weighted phrases in the member profile.

Abstract: A question-answering device, a question-answering method, and a question-answering program that can obtain an answer to an inputted query with high probability are described. A score calculation element 305 determines a matching degree between the group of the style and the topic of an inputted query and the group of the style and the topic of the query of question-answer pairs. A search result presentation element 306 narrows the question-answer pairs on the basis of the matching degree.

Abstract: Techniques, an apparatus and an article of manufacture identifying one or more utterances that are likely to carry the intent of a speaker, from a conversation between two or more parties. A method includes obtaining an input of a set of utterances in chronological order from a conversation between two or more parties, computing an intent confidence value of each utterance by summing intent confidence scores from each of the constituent words of the utterance, wherein intent confidence scores capture each word's influence on the subsequent utterances in the conversation based on (i) the uniqueness of the word in the conversation and (ii) the number of times the word subsequently occurs in the conversation, and generating a ranked order of the utterances from highest to lowest intent confidence value, wherein the highest intent value corresponds to the utterance which is most likely to carry intent of the speaker.

Abstract: Techniques for dynamically compiling a list of one or more solution documents are provided. The techniques include receiving one or more questions, forming one or more of the one or more questions into one or more lists, wherein forming one or more questions into one or more lists comprises expressing each question in a compressed manner using one or more sets of one or more keywords, receiving input from a solution database, and using the one or more lists and input from a solution database to dynamically compile a list of one or more solution documents in connection with the one or more questions.

Abstract: Example apparatus and methods concern dynamically expiring crowd-sourced content (CSC) in a crowd-sourced database. An example apparatus may include logic for acquiring the CSC, where the CSC is data produced by a mobile device concerning a point of interest. The example apparatus also includes logic for producing an evaluation of the CSC and logic for determining an expiration criteria based on the CSC, the evaluation, and the user. The CSC may be data about a point of interest. The evaluation may be based on the completeness, timeliness, or contents of the CSC. The expiration criteria may be established based on the evaluation of the CSC and a user profile. The expiration criteria or user profile may be manipulated based on confirmation or repudiation of the CSC by a different user or by curation of the CSC.

Abstract: An example method includes receiving text from a plurality of documents, segmenting text received text of the plurality of documents, calculating a frequency statistic for each segment of each document, determining segments of potential interest of each document based on calculated frequency statistic, calculating distances between each document of the plurality of documents based on a text metric, and storing segments of potential interest of each document and the distances in a search database. The method may further include receiving a search query and performing a search of information contained in the search database, partitioning documents of search results using the distances, for each partition, determining labels of segments of potential interest for documents of that particular partition, the labels being determined based on a plurality of frequency statistics, and providing determined labels of segments of potential interest for documents of each partition.

Abstract: A method of generating a pool of matched content pieces from an available pool of content pieces based on a selected sample, including providing at least one consumer database containing a plurality of consumer profiles with consumer preference identifying data stored on a data storage device, providing at least one brand database containing details of brand clients, each having one or more branding parameters stored on a data storage device, providing at least one content database containing a plurality of pieces of content provided by at least one content provider with each piece of content having one or more content profile identifiers stored on a data storage device, selecting a number of representative content pieces based on one or more branding parameters of a brand client to convey a brand identity, testing the representative content pieces against a set of constraints based on one or more content profile identifiers to establish that a minimum number of content piece identifiers are chosen using a da

Abstract: A location extraction component analyzes a set of travelogues to identify all of the locations mentioned therein. A co-occurrence extraction component computes co-occurrence values for the identified locations. When the identity of a specified location is received, suggested locations for the specified location are identified through the use of the co-occurrence values. A map is displayed that encompasses an area including the specified location and the suggested locations. The map might include indicators for the specified location and for each of the suggested locations. Attributes of the indicators, such as their size or color, can be modified based upon the co-occurrence value associated with the corresponding suggested location.

Abstract: In one embodiment, a system and method is illustrated including receiving a feedback request identifying a particular user, retrieving a feedback entry in response to the feedback request, the feedback entry containing a first term, building a scoring model based, in part, upon a term frequency count denoting a frequency with which the first term appears in a searchable data structure, mapping the first term to a graphical illustration based upon a second term associated with the graphical illustration such that the graphical illustration may be used to represent the second term, and generating a feedback page containing the first term and the graphical illustration. The method may include assigning a value to the first term so as to identify the first term, assigning the first term to the searchable data structure, and extracting the first term from the searchable data structure based, in part, upon an extraction rule.

Abstract: Methods and comment association systems for associating one or more comments with one or more primary electronic documents are described. In one aspect, the method comprises: identifying, at a comment association system, one or more key terms from at least a portion of the one or more primary electronic documents; identifying, at the comment association system, one or more comments associated with the identified key terms; determining, at the comment association system, whether an identified comment is sufficiently related to the one or more primary electronic documents by calculating one or more relation score for that identified comment and comparing the relation score to one or more threshold; and if the identified comment is sufficiently related to the one or more primary electronic documents, then associating the identified comment with the one or more primary electronic documents at the comment association system.

Abstract: The present application discloses a computer-assisted method for determining relevance between a first text content and a second text content, and displaying relevant contents.