Abstract: A question answering system that determines whether a question is off-topic by performing the following steps: (i) receiving, by a question answering system, a set of documents; (ii) identifying topical subset(s) for each document of the set of documents using named entity recognition, where each topical subset relates to a corresponding topic; (iii) assigning a set of topic score(s) for each topical subset using natural language processing, where each topic score relates to a corresponding probability associated with the respective topical subset under a probabilistic language model; and (iv) determining, based, at least in part, on the topic score(s) corresponding to the topical subset(s), whether a question input into the question answering system is off-topic.

Abstract: A database search method and system utilize user community defined facets and facet values for refining searches. The system provides access to a database having a plurality of records in respective categories of information. Each record has one or more facets to the respective category of information, usable in connection with a search as suggested additional parameters for further refining search terms or guiding user navigation of the database. The facets and facet values are defined by a computer network community of users over time and through use of the network community portal. Another embodiment is an advertising engine that displays targeted advertisements to the user based on refined search. Another embodiment is a method that utilizes refined search to help the user with navigation of a site (e.g., website or other computer network site) as a component of a GUI.

Abstract: From the content of a document, a factual entity that relates to the content of the document is determined. Content for a knowledge panel is requested. A knowledge panel is a user interface element that provides a collection of content related to the factual entity. The contents of the knowledge panel is received for contemporaneous display on the user device with the content of the document.

Abstract: In some examples, analysis metadata may be generated for received digital assets based on analysis of content of the digital assets. As one example, a service computing device may receive and store a digital asset and first metadata for the digital asset. The service computing device may determining a content category of content of the digital asset, and may analyze the content of the digital asset based at least in part on the content category to obtain at least one analysis result as additional metadata for the digital asset. The service computing device may store the analysis result in association with the digital asset and may store analysis metadata in association with the first metadata.

Abstract: A dynamic attribute search engine system can be implemented to execute faceted searches using dynamic facets. In embodiments, a dynamic attribute search engine system includes one or more product databases, a data ingestor, an index, a query service, and a faceted searcher. The dynamic attribute search engine system is programmed to dynamically determine facets for a faceted search in response to dominant categories identified for each search query.

Abstract: Presented is a method by which large amounts of data may be sorted and downloaded. In the presented invention, two keys are selected which may be associated with all records in a designated distributed, relational, object-oriented, graph or other database, and the records are sorted first by means of the first and second keys to generate a first file, and subsequently by means of the second and first keys to generate a second file. It is determined which of the resultant files contains a larger number of groups, and the file containing the larger number of groups is selected for loading, preferably by means of a plurality of data load streams. Further presented is means by which the data load streams may be executed in parallel with one another while minimizing contention for shared resources. The method additionally helps to increase the amount of parallelism that may be deployed in support of database search operations.

Abstract: Social graph based information recommendation engines, devices, systems and methods are described where information of interest can be retrieved and provided to a user based on sensor input and profile or preference information about the user or about a person other than the user.

Abstract: A topology manager implemented on a data partition of a distributed computing system can be notified that a destination computing node in the distributed computing system is not responding to a communication request. Upon determining that the destination computing node is dead and/or has a loss of communication with one or more, and optionally a majority of other computing nodes in a plurality of computing nodes of the distributed computing system, the topology manager can retire the destination computing node and cause a load balancing of replicas of data partitions in the distributed computing system to compensate for loss of the retired computing node.

Abstract: Collecting and distributing information related to recent content publication activity of an instant messaging (IM) user provides other users in a network with timely, relevant information about people known to the user or within the same social network. A user participating in a social network can quickly and efficiently perceive new information related to other users (referred to as co-users) in a social network by reviewing the co-users' recent content publication activity. A user may be made able to do so without requiring the co-user to send a communication directly to the user regarding the new facts or new content, and also without requiring the user to actively browse or request information about the co-user.

Abstract: A data storage method and apparatus. The method includes: determining a same column attribute of at least two data objects that are to be stored in a KeyValue type distributed database; determining a format of a row identifier of each data object of the at least two data objects, where the row identifier format includes the same column attribute and a data object identifier; determining a row identifier value of each data record of each data object according to the determined row identifier format of each data object; and storing each data record and the row identifier value of each data record, where the row identifier value of each data record is used as a primary index. The data storage method and apparatus in embodiments of the present invention can improve data query efficiency.

Abstract: A method for operating a system for creating and retrieving contextual links between user interface objects involves displaying a navigation options menu as an overlay, above a content source, within a user interface, in response to receiving an input control through a UI controller for a UI object, generating an annotation entry for the UI object, generating a contextual link between a first UI object and a second UI object, and displaying an adjacently positioned link counter and a visible indicator around the UI object through operation of a presentation modeler, in response to detecting at least one correlation for the UI object in the correlation table.

Abstract: Aspects include methods, systems, and computer programs to manage user access to a collaborative effort system. The method includes providing, by a processor, a user interface allowing a first user access to the collaborative effort system, responsive to the first user utilizing the user interface, generating a plurality of threads within the collaborative effort system, responsive to the first user utilizing the user interface, authorizing a second user access to at least one thread of the plurality of threads, and generating an icon displayed in the user interface to indicate that the second user is authorized to access the at least one thread. Authorization to a thread comprises an ability to view and participate in the thread.

Abstract: Various embodiments for indexing documents in a computing device are disclosed such that a search engine can perform queries to retrieve relevant portions of a document in a computationally-efficient manner. A document structure and a document layout are extracted from a document, where the document structure defines a logical structure of content of the document, and the document layout defines a visual layout of the content of the document. A search index can be populated using the document structure and the document layout. When a search query is performed, the search results can be ranked by relevance, for example, based on a role of a user account, such that more relevant documents can be presented before less relevant documents.

Abstract: Systems and methods for classifying URLs by a hybrid classifier are provided. According to one embodiment, a hybrid classifier receives a candidate URL. A naïve Bayes classifier of the hybrid classifier classifies the candidate URL to generate a first classification of the candidate URL and a sublink classifier of the hybrid classifier classifies the candidate URL to generate a second classification of the candidate URL. The hybrid classifier determines a classification of the candidate URL based on the combination of the first and second classification.

Abstract: A system and method for asynchronously retrieving information over a network based on incremental input. The system comprises a client part, a communication protocol, and a server part. The server part receives incremental text-based input from one or more objects in the client part, and asynchronously returns matching information to the client part depending on server-defined content channels that each query one or more underlying or connected content sources, such as databases or search engines, to return matches for the latest input and for the latest values and state of other content channels.

Abstract: Described are techniques for request processing. A server receives from a client a request for first data to be returned to the client. The first data includes a first property of an object. The object is included in an object model and includes a predefined property. The first property is not a predefined property of the object and is a new property defined using an expression included in the request. The request includes any of the predefined property and a constant. First processing is performed at the server to service the request and obtain the first data. At the server in accordance with the first processing, a response to the request including the first data is generated.

Abstract: Disclosed herein are systems, methods, and software to enhance meeting technology. In an implementation, a meetings graph may be generated for a meeting between various participants in the meeting. An effectiveness of the meeting may be identified based at least in part on a comparison of the meetings graph to a model meetings graph for a model meeting. An indication of the effectiveness of the meeting may then be communicated to at least an application for surfacing in a user interface to the application.

Abstract: In one aspect, an example method involves: a computing device selecting a media program; the computing device accessing first data representing closed-captioning text that corresponds to the media program; accessing second data representing a program schedule of the selected media-program, wherein the program schedule comprises first text; the computing device selecting second text from among the first text; the computing device using the selected second-text to (i) identify a candidate misspelled-term within the closed-captioning text, and (ii) identify a candidate replacement-term for the candidate misspelled-term; and the computing device outputting, via a user interface, a prompt requesting a command to modify the closed-captioning text by replacing the identified candidate misspelled-term with the identified candidate replacement-term.

Abstract: A translation system can translate a storage request to a physical address using fields as keys to traverse a map of nodes with node entries. A node entry can include a link to a next node or a physical address. Using a portion of the key as noted in node metadata, a node entry can be determined. When adding node entries to a node, a node utilization can exceed a threshold value. A new node can be created such that node entries are split between the original and new node. Node metadata of the parent node, new node and original node can be revised to identify which parts of the key are used to identify a node entry. When removing node entries from a node, node utilization can cross a minimum threshold value. Node entries from the node can be merged with a sibling, or the map can be rebalanced.

Abstract: A method and system for activating malicious actions within electronic documents is described. In one embodiment, the method may include receiving, by a processor of a computing device, the electronic document; identifying, by the processor, an object embedded within the electronic document; identifying, by the processor, an action associated with execution of the object; executing, by the processor, the action within a context of rules associated with the object; identifying, by the processor, at least one behavior that results from execution of the action; and determining, by the processor, an existence of at least one malicious element from the identified behavior.

Abstract: Video and audio content is searchable using a text search. A search component can analyze respective items of content to identify words spoken in the items of content, and generate respective transcripts of the respective words of the items of content based on the analysis. The search component receives a text search comprising a keyword and analyzes the respective transcripts to determine whether a transcript(s) contains a word that matches or substantially matches the keyword. The search component generates a search result(s) associated with the transcript(s) that at least is a substantial match to the keyword. The search component can present a time indicator indicating a time position in proximity to where the word is located in the content of the search result(s), and presentation of the content can start from that time position. The search component can be executed in a set-top box associated with a presentation device.

Abstract: Using a bounding language to control or restrict the changes that can be made to contents of a structured document (e.g., a document encoded in the Extensible Markup Language, or “XML”), and also includes the bounding language and documents encoded according to the bounding language. A Document Type Definition (“DTD”) is defined as a “bounding DTD”, and one or more structured documents containing editing restrictions are defined according to this DTD. A processing component uses a structured document containing editing restrictions as input, and programmatically determines which fields of another structured document can be edited, which fields should be hidden, and so forth. By restricting the parts of the file that can be edited, users who need to do the editing are shielded from irrelevant details, and can carry out their task with less risk of making errors (and without needing to understand the details of the structured document markup language).

Abstract: A system, method, and apparatus for establishing communications with a secure network using a non-secure mobile device operating in a non-secure network are disclosed herein. The disclosed method involves communicating a mobile device identifier to the secure network. In one or more embodiments, the mobile device identifier is an Internet protocol (IP) address and/or a unique identification (ID) code. The method further involves verifying and/or validating, with a mobile device manager in the secure network, the mobile device identifier. Also, the method involves establishing a secure connection between the mobile device and the secure network. In addition, the method involves receiving, with the mobile device, encrypted secure data from the secure network. Further, the method involves decrypting, with the mobile device, the received encrypted secure data using previously downloaded mobile device security software.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for retrieving and using contextual data from previous conversation sessions in conversational searches. In one aspect, a method includes receiving a first query for a first user session, determining that the first query refers to one or more tags in a first repository, the first repository associating respective identifiers to respective tags, each identifier representing a corresponding user session, determining one or more particular identifiers associated with the one or more tags in the first repository, retrieving particular contextual data associated with the determined particular identifiers in a second repository, the second repository associating respective identifiers to respective contextual data associated with corresponding user sessions represented by the respective identifiers, and performing an action responsive to the first query based on the retrieved particular contextual data.

Abstract: An information processing apparatus includes a processor. The processor converts data of a hierarchical structure including a nested structure inputted in a first data format into a second data format representing the data using a plurality of tables and stores the converted data into a storing unit. The processor generates, for each information element of the inputted data, a first table including a value of the information element, and associates each of a plurality of the first tables with a corresponding another first table according to the nested structure of the data and generates a second table by integrating the first tables that satisfy a predetermined condition among the plurality of the first tables.

Abstract: Systems, methods, and machine-readable and executable instructions are provided for preventing a memory attack to a wireless access point (AP). Preventing a memory attack to a wireless access point can include receiving, with a wireless AP, a generic advertisement service (GAS) initial request from a querying station and transmitting, with the wireless AP, a GAS initial response to the querying station without querying an advertisement server based on the GAS initial request.

Abstract: A system and method for selecting point-in-time copies of data objects for restore operations. A user operates a mobile device with a touch screen, which is connected to a backup server through a network. The user selects a given data object from a graphical user interface (GUI) to restore from a backup server and selects a point-in-time for restoration. The mobile device searches point-in-time (PIT) indicator data corresponding to PIT backup copies of the given data object. If the mobile device determines there is no PIT indicator for the requested point-in-time, then the mobile device identifies one or more backups that include the given data object, automatically selects one of the one or more backups as a proposed backup for restoration, and automatically adjusts the input of the user on the graphical user interface to identify the selected one of the one or more backups.

Abstract: Electronic content such as an electronic book (e-book) that is referenced in or otherwise related to other content such as website content is identified, and a visual representation of the electronic content is generated. The visual representation is presented in conjunction with the website content, and a reader application for previewing the electronic content is launched via user interaction with the visual representation. Exiting the reader application causes a browser application that is rendering the website content to be automatically returned to a foreground state. A variety of actions are enabled via interaction with the reader application such as purchasing the electronic content, sharing the electronic content, or downloading a sample preview of the electronic content for later consumption.

Abstract: A method and system for searching documents. The method can include: identifying a first set of messages of a messaging platform referencing a document; identifying message text of the first set of messages; associating the message text with the document in a storage repository; receiving, by a computer processor, a search request including a search term provided by a client; calculating, by the computer processor, a relevance score for the document based at least on the search term and the message text; and providing, to the client and based on the relevance score, a reference to the document in response to the search request.

Abstract: A method for message selection, including: receiving a request for unviewed messages, the request identifying a context account, ranking the set of unviewed messages according to ranking criteria, where the ranking criteria is used to rank each of the set of unviewed messages based on an engagement history between the context account and an authoring account of the unviewed messages; generating, based on ranking the set of unviewed messages, a result set identifying a subset of the set of unviewed messages, and providing the result set in response to the request.

Abstract: A computer-implemented method for forming a search query includes receiving a search request for searching at least one output document based on a first input document, extract a first metadata from the first input document, a second metadata from at least one second input document, and a third metadata from synonyms of the first and second metadata, form a first set of metadata by combining the first, second and third metadata, rank a first set of categories, wherein a category associated with a highest number of input documents is ranked highest, forming a second set of categories by excluding a predefined number of categories, forming a second set of metadata by excluding at least one definition metadata from the first set of metadata, and forming a search query by combining the second set of metadata and second set of categories.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for retrieving and using contextual data from previous conversation sessions in conversational searches. In one aspect, a method includes receiving a first query for a first user session, determining that the first query refers to one or more tags in a first repository, the first repository associating respective identifiers to respective tags, each identifier representing a corresponding user session, determining one or more particular identifiers associated with the one or more tags in the first repository, retrieving particular contextual data associated with the determined particular identifiers in a second repository, the second repository associating respective identifiers to respective contextual data associated with corresponding user sessions represented by the respective identifiers, and performing an action responsive to the first query based on the retrieved particular contextual data.

Abstract: Systems and methods (e.g., utilities) for use in providing automated, lightweight collection of online, open source data which may be content-based to reduce website source bias. In one aspect, a utility is disclosed for use in extracting content of interest from at least one website or other online data source (e.g., where the extracted content can be used in a subsequent search query). In other aspects, utilities are disclosed that are operable to perform various types of analyzes on such extracted content and present graphical representations of such analyzes on a display of a client device.

Abstract: An auto-classification system and method provides dynamic user feedback in a guide that is presented to the user. The feedback presented in the guide enables the user to refine the classification model by adding or removing exemplars, creating, editing or deleting rules, or performing other such adjustments to the classification model. This technology enhances the overall transparency and defensibility of the auto-classification process.

Abstract: The present disclosure provides for automatically detecting groups of people who may be interested in different types of content (e.g., different presentations, different presentation sections, different presentation styles). Further, the present disclosure provides for automatically providing presentation filters appropriate for each group of people.

Abstract: The disclosed technology includes a health engine that monitors and modifies customer-premises equipment (CPE) devices. The health engine can detect patterns in CPE device behavior, identify problems with CPE devices, and adjust CPE device configurations proactively or reactively to address problems or prevent problems. In some implementations, the health engine can instruct a CPE device or gateway to restart, update its software or firmware, notify a user of the CPE device of an unhealthy behavior pattern in a CPE device. The health engine can modify a CPE device prior to a user using the device or when the CPE device is inactive.

Abstract: A system and method for searching using a user device associated with a display includes a controller displaying a plurality of search screens on the display device, each of the plurality of search screens perform a content search. The user device also includes a search module generating respective search results from each of the plurality of search screens and a search retention module storing the respective search results in a memory of the user device and retrieving saved search results from the memory corresponding to the previously accessed search screen when a previously accessed search screen is displayed. The controller displays the saved search results.

Abstract: Various embodiments are disclosed for optimizing the execution of a query to retrieve data associated with multiple entities from a database server system, where the entities are associated with a user of a social networking system. At least one embodiment includes analyzing parameters of the query to identify one or more databases at which to execute the query, determining whether to execute constraint type parameters on the data retrieved at a frontend or backend server of the server system, and separating the performance of privacy checks on the data retrieved from the execution of the query. The embodiment further includes executing, based on the databases identified, the query in parallel threads to retrieve individual data sets of a result set, where a particular individual data set is associated with a particular entity.

Abstract: The present invention provides a method and apparatus for processing a web page. The method comprises: identifying a plurality of source data schema elements of the web page; determining an attribute similarity between one source data schema element among the plurality of source data schema elements and one target data schema element among a plurality of target data schema elements based on attributes of the data schema elements; and mapping the source data schema element to the target data schema element for generating a target web page in response to the attribute similarity being higher than a predetermined threshold. Through this method, the data schema elements or template language elements in the web page may be transformed efficiently and conveniently, so as to transform the web page, and to enhance the transformation efficiency.

Abstract: Examples herein disclose multiple engines to produce output representative of a summary of document produced by each of the multiple engines. The examples apply a weighting mechanism to the output specific to that engine to obtain a value corresponding to that output. The examples identify specialized language if the value corresponding to that output reaches at least a particular threshold.

Abstract: In one embodiment, a first computing device accesses a first search query issued by a user and one or more first documents identified in response to the first search query; determines a first set of tokens for each of the first documents; and transmits the first documents and their associated first sets of tokens to a second computing device associated with the user for display to the user, wherein: each of the first documents is displayed together with its associated first set of tokens; and the tokens of each of the first sets of tokens are visually distinguished based on the concepts they represent and their levels of relevance with respect to the first document associated with the first set of tokens.

Abstract: Architecture that includes an index creation algorithm that utilizes available resources and dynamically adjusts to successfully scale with increased resources and be able to do so for any data distribution. The resources can be processing resources, memory, and/or input/output, for example. A finer level of granularity, called a segment, is utilized to process tuples in a partition while creating an index. The segment also aligns with compression techniques for the index. By choosing an appropriate size for a segment and using load balancing the overall time for index creation can be reduced. Each segment can then be processed by a single thread thereby limiting segment skew. Skew is further limited by breaking down the work done by a thread into parallelizable stages.

Abstract: Offline functionality-based co-ranking and clustering is carried out on a knowledge base that characterizes a heterogeneous information technology services network including a plurality of services, a plurality of providers, and a plurality of attributes. Results of the functionality-based co-ranking and clustering are stored as annotations of the services and the providers in the knowledge base, to obtain an annotated knowledge base. A service requirement is obtained from a customer requiring information technology services. The annotated knowledge base is queried, based on the service requirement; and an ordered list of at least given ones of the services, based on the querying, is returned to the customer.

Abstract: A Data Synchronization and Failover Management (DSFM) system monitors simultaneous execution of non-identical instances of a software application and may label as a particular result of the software application the earliest output corresponding to that result produced by one of the instances. The DSFM may label one of the instances as a primary instance and the other instances as secondary instances and, if the primary instance fails, may re-label one of the secondary instances that computed all of the operations associated with the last result produced prior to the failure of the primary instance, as a new primary instance.

Abstract: An online system provides a continuous feed of content items to a client device. The online system maintains an ordered set of ranked content items for sending to the client device. The online system receives requests from the client device for content items and selects a subset from an ordered set of ranked content items in response to each request. The number of content items in the subset may be based on factors such as the network connection strength of the client device. The client device re-ranks content items received from the online system based on information received during a given session. The session information may indicate the types of content items viewed during the given session, or the speed at which the user is viewing content items during the given session.

Abstract: A computer system for providing software over a network includes: a computer system for providing software over a network is provided. The system includes: a control unit configured to reside at a site, the control unit including a control unit identification (ID) that uniquely identifies the control unit to the network; a copy of the software, the software including sets of features; a license generator configured to create a features activation file containing the control unit ID and identifying at least one set of features to be activated by the control unit; a computer configured to download the features activation file to the control unit; and, the control unit configured for activating one of the sets of features according to the features activation file. A method and a computer program product are disclosed.

Abstract: There is described a multimedia processing method comprising: identifying a start and end time ranges in an input multimedia stream; comparing said ranges to an alternate multimedia stream of the same program for regions of similarity; and defining time ranges as representing chapter regions responsive to the content of both streams within the time ranges having high similarities.

Abstract: Statistical Machine Translation (SMT) based search query spelling correction techniques are described herein. In one or more implementations, search data regarding searches performed by clients may be logged. The logged data includes query correction pairs that may be used to ascertain error patterns indicating how misspelled substrings may be translated to corrected substrings. The error patterns may be used to determine suggestions for an input query and to develop query correction models used to translate the input query to a corrected query. In one or more implementations, probabilistic features from multiple query correction models are combined to score different correction candidates. One or more top scoring correction candidates may then be exposed as suggestions for selection by a user and/or provided to a search engine to conduct a corresponding search using the corrected query version(s).

Abstract: Identifying equivalent JavaScript events includes receiving source code containing two JavaScript events for equivalency analysis, extracting an HTML element containing an event from each JavaScript event and analyzing the extracted HTML elements. Responsive to a determination that the HTML elements are of a same type according to equivalency criteria B, and responsive to a determination that the HTML elements have a same number of attributes according to equivalency criteria C, a determination is made whether JavaScript function calls of each JavaScript event are similar according to equivalency criteria A. Responsive to a determination that the JavaScript function calls are similar according to equivalency criteria A, and responsive to a determination that the other attributes of the HTML elements satisfy equivalency criteria D, the JavaScript events are identified as equivalent.

Abstract: Embodiments include generating data models that may give semantic meaning for unstructured or structured data that may include data generated and/or received by search engines, including a time series engine. A method includes generating a data model for data stored in a repository. Generating the data model includes generating an initial query string, executing the initial query string on the data, generating an initial result set based on the initial query string being executed on the data, determining one or more candidate fields from one or results of the initial result set, generating a candidate data model based on the one or more candidate fields, iteratively modifying the candidate data model until the candidate data model models the data, and using the candidate data model as the data model.