Abstract: Systems and methods are described herein for generating a search query using flexible autocomplete suggestions. A text input is received, and a plurality of portions of the text input, each corresponding to a different search parameter, are identified. For each of the identified portions, at least one suggested alternate text is retrieved based on the corresponding attribute, and a user interface element is generated for display in which the original text of the respective portion is displayed, along with the suggested alternate texts for that portion, which are selectable by the user. Upon receiving selection of a suggested alternate text in at least one user interface element, a search query is generated based on each portion for which no alternate has been selected, and each selected alternate text. Search request are retrieved in response to the search query, and the results are generated for display to the user.

Abstract: An online system receives a search query from a user. In response to the request, the online system obtains search results matching the search query and identifies a set of attributes describing a context of the search query. The online system generates a data structure that includes a plurality of search coefficients. The search coefficients are selected based on the identified set of attributes. Some of the search coefficients have conflicting values. The online system traverses the data structure to identify the search coefficients having conflicting values. For each search coefficient having conflicting values, the online system resolves conflicts and determines a value of the search coefficient. The online system ranks search results based on the resolved values of the search coefficients.

Abstract: A method including analyzing the query data to formulate one or more groups of synonym candidates, wherein each synonym in the one or more groups of synonym candidates is categorized in one or more synonym candidate categories. Reformulating the first search query by removing at least one synonym candidate from the first group of synonym candidates to form a first set of synonyms such that each synonym in the first set of synonyms is categorized in the synonym candidate category that matches the search query category of the search query, wherein the at least one synonym candidate removed from the first group of synonym candidates is not categorized in the synonym candidate category.

Abstract: A method generates data visualizations. A device receives a first user input to specify a natural language command directed to a data source. The natural language command includes a first keyword that is a gradable adjective that modifies a second keyword in the natural language command. The device identifies a numerical data field, from the data source, whose field name co-occurs with the first keyword in natural language utterances in a predefined stored corpus of natural language utterances, based on quantifying co-occurrence using a co-occurrence metric. The device computes a numerical range for values of the numerical data field based on correlation of sentiment polarity of the first keyword and the field name of the numerical data field. The device then generates a data visualization for the data source, based on filtering rows of data from the data source according to the numerical range for the numerical data field.

Abstract: A shard of a selected document is identified. A hierarchy of documents including the selected document are to be tagged with metadata in a data store of a search engine so that the related documents are queryable by the metadata using the search engine. The shard specifies a location of the related documents within the data store of the search engine. For each related document, a tagging operation is generated within a script, the tagging operation specifying the metadata, the shard, an identifier of the related document, and an identifier of the selected document. The script including the generated tagging operation for each related document is transmitted to the search engine to execute to tag the related documents in the data store with the metadata and with the selected document as a source of the metadata.

Abstract: A method for presenting content items includes receiving, by a first user device, a request for a video item hosted by a content platform, and providing a graphical user interface comprising a first portion having a first media player to playback the requested video item and a second portion having a second media player to playback an additional video item of a plurality of additional video items. The requested video item comprises a plurality of portions, each associated with a corresponding portion of the additional video item based on a set of actions that were performed by a second user interacting with the corresponding portion of the additional video item. The method further includes in response to an interaction of the first user with respect to a portion of the additional video item, causing the playback of the requested video item to pause.

Abstract: Systems and methods are disclosed for associating a plurality of Internet-enabled devices with a common user profile for targeting Internet content or advertising. One method includes: receiving, from a plurality of Internet-enabled devices, a plurality of requests for electronic content or advertising; extracting, from each of the plurality of requests, a source IP address and a unique identifier associated with the respective Internet-enabled device; for each source IP address for which requests were received over a predetermined time period from a number of Internet-enabled devices below a threshold number of devices, identifying each possible pair of devices from which requests were received; and for each possible pair of devices, calculating a probability that the pair of devices are owned or operated by a common user.

Abstract: Systems, methods, and computer readable media related to generating query variants for a submitted query. In many implementations, the query variants are generated utilizing a generative model. A generative model is productive, in that it can be utilized to actively generate a variant of a query based on application of tokens of the query to the generative model, and optionally based on application of additional input features to the generative model.

Abstract: One or more computer processors match a query pattern to a received query; context information related to the received query; retrieve a set of query records including the same context information as the obtained context information from an index knowledge base, wherein each query record in the set of query records include context information related to a respective history query, the query pattern, an index type associated with the query pattern, and performance information relating to the query pattern and the index type; determine that a subset of the retrieved query records includes one or more query patterns equivalent to the matched query pattern; select a query pattern and an associated index type from the subset of query records based on associated performance information in the set of query records; and perform the received query by applying the selected query pattern and the associated index type.

Abstract: A method, a device, and a non-transitory storage medium are described, which provide a natural language-based content system with corrective feedback and training service. The natural language-based content system with corrective feedback and training service may collect data based on interaction with search results from users. The natural language understanding model may generate feedback data based on the collected data, and use the feedback data to further train the natural language understanding model and update search and discovery logic for searching and discovering contents. The feedback data may categorize errors based on the interaction, and identify differences between search queries received during a search session with a user.

Abstract: A method of information retrieval is provided. The method comprises receiving a query from a user and parsing the query in real-time as the user enters the query. The parsed query is interpreted dynamically based on a defined schema of a knowledge base, and a number of query interpretations is displayed in real-time as the user enters the query. When a selection of one of the query interpretations is received from the user information is retrieved from the knowledge base according to the selected query interpretation.

Abstract: Providing access to usage reports on a cloud-based data warehouse including maintaining, by a management module, a metadata table on the cloud-based data warehouse, wherein the metadata table comprises usage reports for a plurality of organizations; receiving, by the management module, a request for the metadata table from an administrator account for a first organization of the plurality of organizations; granting, by the management module, the administrator account permission to access a filtered portion of the metadata table, wherein the filtered portion of the metadata table is generated by filtering the metadata table by an organization identifier of the first organization; and providing, by the management module, the filtered portion of the metadata table to the administrator account.

Abstract: Aspects include receiving a request to perform an image classification task in a target domain. The image classification task includes identifying a feature in images in the target domain. Classification information related to the feature is transferred from a source domain to the target domain. The transferring includes receiving a plurality of pairs of task-irrelevant images that each includes a task-irrelevant image in the source domain and in the target domain. The task-irrelevant image in the source domain has a fixed correspondence to the task-irrelevant image in the target domain. A target neural network is trained to perform the image classification task in the target domain. The training is based on the plurality of pairs of task-irrelevant images. The image classification task is performed in the target domain and includes applying the target neural network to an image in the target domain and outputting an identified feature.

Abstract: Systems and methods are described herein for interpreting natural language search queries that account for contextual relevance of words of the search query that would ordinarily not be processed, including, for example, processing each word of the query. Each term or phrase is associated with a respective part of speech, and a frequency of occurrence of a combination of adjacent terms or phrases public domain is determined. A relevance of each term is then determined based on its respective type of term and frequency of occurrence in the public domain. The natural language search query is then interpreted based on the importance or relevance of each term.

Abstract: Systems and methods are described herein for interpreting natural language search queries that account for contextual relevance of words of the search query that would ordinarily not be processed, including, for example, processing each word of the query. Each term or phrase is associated with a respective part of speech, and a frequency of occurrence of a combination of adjacent terms or phrases public domain is determined. A relevance of each term is then determined based on its respective type of term and frequency of occurrence in the public domain. The natural language search query is then interpreted based on the importance or relevance of each term.

Abstract: Systems and methods are described herein for generating a search query using flexible autocomplete suggestions. A text input is received, and a plurality of portions of the text input, each corresponding to a different search parameter, are identified. For each of the identified portions, at least one suggested alternate text is retrieved based on the corresponding attribute, and a user interface element is generated for display in which the original text of the respective portion is displayed, along with the suggested alternate texts for that portion, which are selectable by the user. Upon receiving selection of a suggested alternate text in at least one user interface element, a search query is generated based on each portion for which no alternate has been selected, and each selected alternate text. Search request are retrieved in response to the search query, and the results are generated for display to the user.

Abstract: The present disclosure is directed to generating cross-platform electronic tokens. First information of a first user associated with a first identity platform is accessed. Second information of a second user associated with a second identity platform is accessed. The second identity platform is different from the first identity platform. The first information and the second information are analyzed to determine whether the first user and the second user are the same person. In response to a determination that the first user and the second user are the same person, a cross-platform electronic token is generated to be used for both the first identity platform and the second identity platform.

Abstract: A method generates data visualizations. A device receives a first user input to specify a natural language command directed to a data source. The natural language command includes a first keyword that is a gradable adjective that modifies a second keyword in the natural language command. The device identifies a numerical data field, from the data source, whose field name co-occurs with the first keyword in natural language utterances in a predefined stored corpus of natural language utterances, based on quantifying co-occurrence using a co-occurrence metric. The device computes a numerical range for values of the numerical data field based on correlation of sentiment polarity of the first keyword and the field name of the numerical data field. The device then generates a data visualization for the data source, based on filtering rows of data from the data source according to the numerical range for the numerical data field.

Abstract: Techniques for response generation are provided. One computer-implemented method comprises: receiving, by an electronic device operatively coupled to a processing unit, a query from a user. The computer-implemented method also comprises generating, by the electronic device, a dialog candidates set based on user context, wherein the user context comprises the query. The computer-implemented method can also comprise generating, by the electronic device, a response based on the dialog candidates set.

Abstract: Systems and methods are disclosed for associating a plurality of Internet-enabled devices with a common user profile for targeting Internet content or advertising. One method includes: receiving, from a plurality of Internet-enabled devices, a plurality of requests for electronic content or advertising; extracting, from each of the plurality of requests, a source IP address and a unique identifier associated with the respective Internet-enabled device; for each source IP address for which requests were received over a predetermined time period from a number of Internet-enabled devices below a threshold number of devices, identifying each possible pair of devices from which requests were received; and for each possible pair of devices, calculating a probability that the pair of devices are owned or operated by a common user.

Abstract: The methods and systems are designed to utilize an integrated combination of just in time, just in place, and just on device actions connected to an image recognition process to reduce or remove the risk of a User utilizing fake or stolen credentials in order to get verified.

Abstract: A relationship graph system disclosed herein provides a system and method for inferring relationships between various entities based on analysis of a plurality of observations including elements related to two different entities. The method comprises representing each of the plurality of observations using a relationship edge between two nodes, each of the two nodes representing an element related to one of the two different entities, assigning a score to each relationship edge based on metadata related to the observation represented by the relationship edge, combining a plurality of relationship edges between the elements related to the two different entities to generate an aggregate relationship edge between the two different entities, and inferring a relationship between the two different entities based on length of the aggregate relationship edge between the two different entities.

Abstract: A machine learning system builds and uses computer models for identifying or predicting intensity of emotional reactions elicited by a particular video. Such computer models may also determine which particular emotional reaction corresponds to certain times during the video, and whether these reactions are positive or negative for a particular user. The computer models can also predict emotional reactions likely to be elicited by new videos based on learned correlations between video features and elicited emotional reactions.

Abstract: A computer-implemented method includes scanning data maintained on multiple social networks, where, scanning includes identifying a first set of data that is associated with a protected social entity. Scanning data maintained on one or more additional platforms, where, scanning includes identifying a second set of data that is associated with the protected social entity. Scanning data is performed on a continuous basis without user initiation. One or more characteristics of the first set and second set of identified data are determined, and a reference to the identified data, that indicates the characteristic, is generated for each of the one or more characteristics. The one or more generated references of the first set of identified data is compared to the one or more generated references of the second set of identified data. A correlation score is determined based on the comparison, and a threat level indicator is generated based on the determined correlation score.

Abstract: Methods and systems for network-based user authentication are provided. In one embodiment, a method is provided that include receiving a request to connect to a local network. The request may be received from a user device and may include a user device identifier. The method may then include allowing the user device to connect to the local network and create a partial identity of a user of the user device. The partial identity may include a plurality of identification factors associated with the user. The method may then include monitoring network activities of the user device and updating the partial identity based on the monitored network activities.

Abstract: Tagged sound file libraries are searched using an iterative search application. A searcher specifies a multi-term search query which is directed to the sound libraries. Responsive audio files from the libraries are ranked by confidence levels for the occurrence of sounds corresponding to searched terms, weighted by user-determined search term priorities. High-ranking candidate audio files are presented to the user who is able to indicate a preference or a disfavoring of the candidates. Based on the feedback, the application adjusts the priorities of the individual search terms in the search query, adds or removes search terms from the query, and performs a new search with a revised set of search terms. The results are ranked using the adjusted priorities. The iterative search locates files with desired sounds that conform to the initial search or that have evolved from it under the user's direction.

Abstract: A method performed by a physical computing device communicatively coupled to a computer network for creating an aggregate report of a presence of a user of the physical computing device on the network includes periodically searching the network with the physical computing device for content related to the user authored by a source other than the user, the searching being based on a unique set of identifiers associated with the user, dynamically updating the unique set of identifiers based on contextual trends within content returned as a result of the searching, and aggregating content published by the user to the network with the content returned as a result of the searching to form the aggregate report of the presence of the user on the network.

Abstract: In various embodiments, training objects are classified by human annotators, psychometric data characterizing the annotation of the training objects is acquired, a human-weighted loss function based at least in part on the classification data and the psychometric data is computationally derived, and one or more features of a query object are computationally classifies based at least in part on the human-weighted loss function.

Abstract: A method for presenting a graphical user interface (GUI) of a software application connected to a server, the software application comprising at least a database, is provided. The method receives, at the server, user input comprising a search; generates, at the server, a set of search results based on the search, the set of search results being associated with entities of the software application, each of the entities comprising a respective database object containing predefined categories of data, the database comprising at least the database object; determines a priority order of the search results, based on user interaction history with the entities; and presents, via the GUI, the search results according to the priority order.

Abstract: Techniques for dynamic display of visualized units received from plural sources are described. Visualized units are scheduled for display for a time period and rendered on the display for the time period. The visualized units can be shared according to user preferences, picked according to a user selection into personal, shareable sets of visualized units. The visualized units can be annotated, mixed from different sources and focused.

Abstract: In some embodiments, a client application at a computing device can receive a request to generate a digital focus area based on input criteria. In response to the request, the client application can search a local storage and/or a remote storage for content related to the input criteria to yield a content collection, the remote storage being associated with a user account registered at a content management system. The client application can generate a workspace for the digital focus area, the workspace containing selected content from the content collection. The client application can pin, to a desktop displayed on the computing device, an interface object associated with the digital focus area, the interface object providing access to the workspace. The client application can then monitor events associated with the input criteria, and update the digital focus area based on one or more of the events associated with the input criteria.

Abstract: Systems, methods, and media for translating requests to acquire or change goods and services by receiving a request in a service catalog. The received request is received in a format that is not suitable for a target vendor. The received request is then translated into a format suitable for the target vendor. The translated received request is then transmitted to the vendor using the translated format.

Abstract: A dependency is created between a first search query and a second search query. The first search query defines a first data processing pipeline and the second search query defines a second data processing pipeline that extends the first data processing pipeline. A modification is detected to the first data processing pipeline defined by the first search query. Based on the modification to the first data processing pipeline being detected, the dependency is enforced such that the second data processing pipeline is modified to extend the modified first data processing pipeline. The modification to the first data processing pipeline can include a first set of pipelined commands corresponding to the first search query being modified, and the dependency can be enforced by causing a second set of pipelined commands corresponding to the second search query to be modified to include the modified first set of pipelined commands.

Abstract: Described are methods, systems and computer readable media for keyed row data selection and processing.

Abstract: A method and server for training a machine learning algorithm (MLA) for determining a query-completion suggestion for a partial query is disclosed. The method comprises receiving and parsing past queries into n-grams. Each one of the n-grams being associated with respective n-gram features, the n-gram features being indicative of a pair-based co-occurrence of n-grams in the past queries. The method also comprises, for a given n-gram of a given past query: selecting at least one candidate n-gram from the n-grams based on the pair-based co-occurrence; generating respective feature vectors for the given n-gram and the at least one candidate n-gram; generating a training set for the given n-gram comprising an input portion and a label portion; and training the MLA based on the training set to determine a predicted group-based co-occurrence of at least one in-use candidate n-gram and at least one in-use n-gram.

Abstract: Systems and methods are disclosed for associating a plurality of Internet-enabled devices with a common user profile for targeting Internet content or advertising. One method includes: receiving, from a plurality of Internet-enabled devices, a plurality of requests for electronic content or advertising; extracting, from each of the plurality of requests, a source IP address and a unique identifier associated with the respective Internet-enabled device; for each source IP address for which requests were received over a predetermined time period from a number of Internet-enabled devices below a threshold number of devices, identifying each possible pair of devices from which requests were received; and for each possible pair of devices, calculating a probability that the pair of devices are owned or operated by a common user.

Abstract: The present disclosure describes a method and device for presenting information. An example method includes receiving an information presentation request sent from a first terminal and acquiring a first network address of the first terminal. At least one account according to the first network address, a first period to which a current time moment belongs, and a prestored correspondence between a network address, a period and an account are determined. Presentation information matching with account information of the at least one account is selected and the presentation information is sent to the first terminal.

Abstract: Disclosed are systems, methods, and non-transitory computer-readable media for expanding search queries using query term weighting. A search system receives a search term comprising multiple keywords and executes a search query based on the search term. In response to determining that the number of search results is less than a threshold number, the search system determines, based on historical query reformulation data, weight values for the keywords included in the search term. The weight values indicate a probability that each respective keyword would be removed when reformulating the search term. The search system generates a reformulated search term based on the weight values and executes a second search query based on the reformulated search term. The search system causes presentation of at least a portion of the search results on a client device of the requesting user.

Abstract: The system receives user-selected fields from a set of candidate fields. Based on the user-selected fields, the system recommends at least one candidate field from the set of candidate fields. The system may select the at least one candidate field for recommendation based, at least in part, on a set of one or more prior queries. Responsive to receiving a user selection of a recommended candidate field, the system adds the recommended candidate field to the set of user-selected fields to obtain an updated set of user-selected fields. The system executes the query based on the updated set of user-selected fields.

Abstract: Embodiments of the present technology relate managing database query concurrency. A method of the present technology can include receiving a query, generating a first query plan that can be used to execute the query in system memory without any system memory constraints, and estimating a system memory cost for executing the query in the system memory using the first query plan. The method can also include placing the query in a queue if available system memory does not satisfy the estimated system memory cost. The method can further include conditionally selecting the query from the queue, conditionally generating a second query plan for the query that can be used to execute the query in the system memory in compliance with a system memory constraint, and conditionally executing the query in the system memory.

Abstract: One or more embodiments allow a user to search a gallery of digital content. In particular, a user can interact with a digital content system to search for, and identify, one or more digital content items (e.g., photos, videos, audio) within a collection of digital content. For instance, the digital content system can maintain tokens with respect to a collection of digital content and associate the tokens with digital content items within the collection of digital content. The digital content system can also provide a gallery of digital content items within a view area of a graphical user interface. Upon receiving a search query, the digital content system can identify a token and identify digital content items corresponding to the token. The digital content system can further provide a new or modified gallery within the view area of the graphical user interface based on the identified digital content items.

Abstract: The present disclosure is directed to training and using machine learning models to determine user intent from a search query, for example via a semantic parse that identifies particular catalog fields for items in an electronic catalog that would satisfy the user's current mission as reflected in their search query intent. The determined intent can then be used to filter recommendations and/or pre-select attribute-value input fields on detail pages displayed after the user navigates away from the search results page, until the mission is complete.

Abstract: Users of a social networking system initiate conversations between other users and additional users who may not be registered to use the social networking system. Each additional user is identified by an item of contact information, and conversation messages are sent to an additional user using the contact information. The conversations are associated with the item of contact information and saved in the social networking system. A user of the social networking system may subsequently claim the item of contact information. The claiming user is given access to the associated conversations if the social networking system can verify that the item of contact information belongs to the claiming user.

Abstract: Disclosed are various embodiments for a social networking behavior-based identity system that employs social networking data that a user has elected to share through an opt-in procedure. An assertion of a user identity is received from a client. It is determined whether the assertion of the user identity specifies a correct security credential. Social networking data identifying a circle of friends is received. It is determined whether the user identity belongs to a user at the client based at least in part on a reputation of one or more members of the circle of friends and whether the assertion of the user identity specifies the correct security credential.

Abstract: A method, an apparatus, and a storage medium are provided for searching for an object. The method is applied in a terminal device and includes: in response to a searching request with respect to an object to be searched for, acquiring a scene picture for surrounding environment through an image capturing apparatus of the terminal device; detecting whether a suspected object corresponding to the object to be searched for exists in the scene picture; and when a suspected object exists in the scene picture, highlighting the suspected object in an augmented reality (AR) manner.

Abstract: Techniques and mechanisms are disclosed for generating and causing display of graphical interfaces which enable an interactive and flexible search results visualization process. Based on results data identified in response to execution of a search query, an interface element is displayed which enables users to select a field contained in the results data, also referred to herein as a “dimension” or “facet,” and for which a “faceted” visualization of the results data can be dynamically generated and displayed. As used herein, a faceted visualization refers to a graphical interface including display of at least two separate data visualizations generated based on a selected facet data dimension, where each separate data visualization corresponds to a distinct value of the selected facet dimension.

Abstract: Systems and methods are disclosed for associating a plurality of Internet-enabled devices with a common user profile for targeting Internet content or advertising. One method includes: receiving, from a plurality of Internet-enabled devices, a plurality of requests for electronic content or advertising; extracting, from each of the plurality of requests, a source IP address and a unique identifier associated with the respective Internet-enabled device; for each source IP address for which requests were received over a predetermined time period from a number of Internet-enabled devices below a threshold number of devices, identifying each possible pair of devices from which requests were received; and for each possible pair of devices, calculating a probability that the pair of devices are owned or operated by a common user.

Abstract: In some embodiments, a client application at a computing device can receive a request to generate a digital focus area based on input criteria. In response to the request, the client application can search a local storage and/or a remote storage for content related to the input criteria to yield a content collection, the remote storage being associated with a user account registered at a content management system. The client application can generate a workspace for the digital focus area, the workspace containing selected content from the content collection. The client application can pin, to a desktop displayed on the computing device, an interface object associated with the digital focus area, the interface object providing access to the workspace. The client application can then monitor events associated with the input criteria, and update the digital focus area based on one or more of the events associated with the input criteria.

Abstract: Internet advertising to users of web browser personal computer systems is a very large and mature industry. However, many new digital devices such as cellular telephones, table computer systems, and video game console are now presenting an even larger internet advertising market. Although the techniques used for targeting advertisements to web browsers on personal computers are sophisticated, the techniques for accurately targeting internet advertisements to these new digital devices are limited. To improve the quality of targeting advertisements on new digital devices a set of techniques for accurately pairing digital identities is disclosed. Once various digital identities are linked, all of the accumulated digital profile information from these linked digital identities may be used to accurately select advertisements for all of the linked digital devices.

Abstract: Systems and methods for mapping IP addresses to an entity include receiving at least one domain name associated with the entity. Embodiments may further include determining one or more variations of the at least one domain name based on analysis of domain name data collected from a plurality of domain name data sources that mention a variation of the at least one domain name. Some embodiments may also include identifying one or more IP addresses pointed to by the one or more variations of the entity's domain name based on analysis of IP address data collected from a plurality of IP address data sources. Additional embodiments include assigning weights to each of the identified one or more IP addresses and creating a mapping of IP addresses to associate with the entity based on analysis of the weighted one or more IP addresses.