Abstract: Systems and techniques are generally described for detecting evasive terms in item listings. In some examples, a list of keywords associated with a first category of specified items may be received. In some examples, text data associated with a first item may be received. A first plurality of tokens representing a corrected version of the text data may be generated using spell correction techniques. In some cases, first flag data indicating that the text data constitutes evasive text may be generated based at least in part on the generation of the first plurality of tokens. The first flag data and a numerical representation of the first plurality of tokens may be input into a binary classifier trained to generate label data indicating whether text data associated with a given item is evasive or non-evasive. The binary classifier may generate label data indicating that the text data associated with the first item is evasive.

Abstract: One example method includes identifying dissimilar items in a data set. A data set may be walked one or more times and the nodes or vertices of the data set may be scored based on the number of times the nodes are touched during the walks. Scores below a threshold score are determined to be dissimilar nodes in the data set. This allows a diverse set of nodes to be identified. A dissimilar data set may be used to prevent unintentional bias in algorithmic training.

Abstract: Technologies are described here for, among other things, improving search query relevance by executing a query on a search engine, retrieving search-page-data generated from executing the query, the search-page-data including document-titles and universal resource locators (URLs), each document-title being a title of a document associated with a URL, determining relevant-entity-words in the query from an entity relevance score for matching search terms in the query, Domain-URLs, and Domain-Titles, determining relevant-intent-words in the query from an intent-word relevance score based on a number of times a search term appears in the query and the URLs relative to other search terms in the query and the URLs, comparing each of the determined relevant-entity-words and each of the determined relevant-intent-words with a plurality of stored past-user queries, retrieving the plurality of stored past-user search queries including the relevant-entity-words and the relevant-intent-words, and transmitting a set of qu

Abstract: A system and method for optimizing industrial assembly process in an industrial environment is disclosed. A system operates on artificial intelligence (AI) based conversational/GUI platform, where it receives user commands related to industrial assembly process improvement queries. By analyzing received user commands, system identifies type of industrial assembly process mentioned by extracting relevant keywords or other attributes. Using trained AI-based classification table, system determines performance attributes associated with identified type of process. The system leverages various sources such as domain knowledge, organization-specific knowledge bases, data from tools/internet-based services, and statistical measurements from industrial environment.

Abstract: A response generating apparatus (10) includes an acquiring unit (40) that acquires input information that is a trigger for generating a response with respect to a user and context information that is information indicating a situation of the user and a response generating unit (50) that generates, based on the context information acquired from the user, a response associated with the input information.

Abstract: Within an advertising buying platform, advertising buyers are able to direct advertising content to TV/OTT/VOD inventory that matches targeted market segments. The advertising buyers select from a list of available market segments, and the system pairs the market segments to an anonymous dataset of media consumers (using a database of hashed online IDs). The dataset of hashed online IDs is then mapped to external OEM IDs (such as from a Smart TV) via the functions of a hashed identity mapping method. This permits the advertiser to target advertising content to a population of TV viewers. Also described herein are methods of bidding on programmatic TV content.

Abstract: A system for generating similarity data for different datasets in a cloud data platform. A first dataset of a plurality of datasets on the cloud data platform is identified, where the first dataset is associated with a first user of the cloud data platform. A semantic type for each feature the first dataset is identified, and each semantic type for the first dataset is compared with existing data of the first user. Semantic types for each feature of each dataset are identified, and each semantic type for the first dataset is compared to each semantic type of each dataset. Overlap requests are generated to output overlap datasets between the first dataset and each of the plurality of datasets. A results dataset is generated by applying the overlap requests to a joined dataset comprising data from the first dataset and data from each of the plurality of datasets.

Abstract: A method, an apparatus, an electronic device and a storage medium for training a user click model, which relate to the artificial intelligence field, are disclosed. is the method may include: collecting a plurality of pieces of behavior data from a log database of users on a network, each piece of behavior data including a user's feedback information regarding resources in the network within a preset time period; generating a plurality of resource access features respectively corresponding to the plurality of pieces of behavior data, based on a pre-created header resource library and the plurality of pieces of behavior data; and training the user click model with the plurality of resource access features. The technical solution provides a lossless modeling manner which, compared to the existing modeling manners, may effectively optimize the precision and improve the accuracy of the user click model.

Abstract: One or more computing devices, systems, and/or methods are provided. In an example, an internet resource identification item associated with one or more internet resources may be identified. User activity information associated with a plurality of events may be analyzed to determine a plurality of sets of text associated with the internet resource identification item, wherein each set of text of the plurality of sets of text is associated with an event, of the plurality of events, associated with an internet resource of the one or more internet resources. A plurality of term representations may be determined based upon the plurality of sets of text. A user intention-based representation associated with the internet resource identification item may be generated based upon the plurality of term representations. A content item may be selected for presentation via a client device based upon the user intention-based representation.

Abstract: Consolidation of disparate data source queries is provided. An application intake component can receive an application package file, and process the file to identify one or more actions and entities for the application. A natural language processor component can receive an input audio signal to identify an action and an entity on which to perform the action. An application delivery component can identify applications that can perform the action and applications that have access to the entity, and then select an application that is configured to perform the action on the entity. The application delivery component can provide an indication of the application selected by the system to a client device.

Abstract: Systems, methods, and other embodiments associated with defining an adaptive polymorphic object agnostic conversational interaction model are described. In one embodiment, a computer-implemented method includes separating a conversational user input query into component structural entities using a semantic processor; constructing a data query based at least on the component structural entities, metadata describing one or more of the component structural entities, a query paradigm, and a current contextual state; transmitting the data query to a data provider for execution; and presenting the data query results through a user interface.

Abstract: Techniques are described herein for generating, editing, and optimizing queries using neural networks. In some embodiments, the techniques include training a neural network using a set of performant database queries to automatically learn patterns between different sequences of tokens in performant queries. Once trained, the neural network may receive an incomplete query as input, where the incomplete query includes one or more query tokens. The trained neural network may then perform next token prediction to project a set of one or more additional query tokens that may follow the one or more query tokens in the incomplete query to form a completed, performant query.

Abstract: A device implementing a system for limiting the scope of a search includes a processor configured to, receive, by a first application, first user input including at least a portion of a first search term, and to determine that a second application corresponds to the first search term. The processor is further configured to display a graphical element for activating a search filter that limits search results to content of the second application, and to receive second user input including selection of the graphical element. The processor is further configured to send, to the second application, a search request including at least one of the first search term or a second search term, to receive, from the second application, a completion suggestion or search result based on a search performed on the content of the second application, and to provide the completion suggestion or search result.

Abstract: A method includes: repeatedly executing a first processing configured to generate, in response to input of conversion examples of values set, a conversion program for converting values of records, convert the values of the records by executing the conversion program, and display a conversion result; and executing a second processing configured to select one or more second records from the records, each of the one or more records being a record on an upper side of a first record, the first record being a record for which a conversion example is added in second or subsequent input, determine whether a value of each second record is changed in a latest conversion result from a previous conversion result, and in response that a value of the second record is changed, cause the first processing to highlight a value of the second record in display of the latest conversion result.

Abstract: Example embodiments of the present disclosure include a system comprising a computer-readable storage medium storing at least one program and a computer-implemented method for providing a dynamic content delivery search system. In example embodiments, entry of a first keyword of a search string comprising a plurality of keywords is detected. The first keyword is identified after entry of the first keyword and prior to entry of a second keyword of the search string. A first set of matching results that match the identified first keyword is determined. Entry of the second keyword is detected and the second keyword identified. Using the second keyword, the first set of matching results is refined. The resulting search results are presented to a user.

Abstract: An online concierge system may determine recommended search terms for a user. The online concierge system may receive a request from a user to view a user interface configured to receive a search query. The online concierge system retrieves long-term activity data including previous search terms entered by the user while searching for items to add to an online shopping cart. For each previous search term, the online concierge system retrieves categorical search terms corresponding to one or more categories to which the previous search term was mapped. The online concierge system determines a set of nearby categorical search terms and sends, for display via a client device, the set of nearby categorical search terms as recommended search terms.

Abstract: A method of propagating annotations of content items to a search query is disclosed. A strength of a correspondence between a search query and a listing of an item on a network-based publication system is determined. The strength of the correspondence is based on an analysis of a set of actions by a set of users who submitted the search query. A set of annotations is generated. The set of annotations is propagated to a search engine and used to enhance search results.

Abstract: A network device (e.g., a user's mobile phone) may be used to make a telephone call to a landline telephone associated with a client device (e.g., a business's device). If the telephone call is terminated, either the network device or the client device may generate a text message to the other and establish a communication session. The client device may present a variety of options to the network device of service requests that may be completed by text message. The network device may transmit the service request and the service request may be fulfilled by the client.

Abstract: The subject matter of this specification generally relates to preventing the distribution of forbidden network content. In one aspect, a system includes a front-end server that receives content for distribution over a data communication network. The back-end server identifies, in the query log, a set of received queries for which a given forbidden term was used to identify a search result in response to the received query even though the given forbidden term was not included in queries included in the set of received queries. The back-end server classifies, as variants of the given forbidden term, a term from one or more queries in the set of received queries that caused a search engine to use the given forbidden term to identify one or more search results in response to the one or more queries and prevents distribution of content that includes a variant.

Abstract: Methods, systems and programming for providing query suggestions based on user feedback. In one example, a prefix of a query is first received. An input including a prefix of a query is received from a user in a search session. A plurality of query suggestions are fetched based on the prefix of the query. Rankings of the plurality of query suggestions are determined based, at least in part, on the user's previous interactions in the search session with respect to at least one of the plurality of query suggestions. The at least one of the plurality of query suggestions has been previously provided to the user in the search session. The plurality of query suggestions are provided in the search session based on their rankings as a response to the input.

Abstract: A method of generating content information that matches at least one stored keyword is described. At least one keyword associated with content information is stored. At least one previously searched keyword in a search record is matched with the at least one stored keyword associated with the content information. First-category mapping data is generated based on a first mapping between the matched at least one stored keyword and the at least one previously searched keyword. Second-category mapping data is generated based on the content information and the at least one stored keyword. A received target keyword is determined to be included in the first-category mapping data. In response to the received target keyword, which is included in the first-category mapping data, circuitry of a terminal searches for the content information associated with the target keyword in the second-category mapping data and displays the content information.

Abstract: A resource link engine may aggregate, from one or more communication services, data including a plurality of messages exchanged between a plurality of users. The resource link engine may compute, for each resource link included in the plurality of messages, a first relevance score of the resource link for a user and/or a second relevance score of the resource link for a group of users including the user. The resource link engine may identify, based on the first relevance score and/or the second relevance score, one or more resource links relevant to the user. In response to detecting the user interacting with a browser at a device, the resource link engine may provide, to the device, the resource links identified as being relevant to the user. Related systems, methods, and articles of manufacture, including computer program products, are provided.

Abstract: A method of content production (e.g., content editing) using content modeling to facilitate content production. In one embodiment, an automated process is configured to render content. For a given content portion, and as the given portion is being rendered, the portion is processed to generate a content model. With respect to a concept expressed in or otherwise associated with the content, the system compares the content model with a target content derived model to generate a relevancy score. The target content derived model is generated by (a) identifying a set of target content portions in which the concept is expressed, (b) generating from each content portion an associated target content model; and (c) performing a vector operation on the associated target content models. Preferably, each associated target content model is built using an Artificial Intelligence (AI)-based content analysis. The relevancy score is used to generate a content production recommendation.

Abstract: Methods, systems, and computer readable media related to generating a combined search query based on search parameters of a current search query of a user and search parameters of one or more previously submitted search quer(ies) of the user that are determined to be of the same line of inquiry as the current search query. Two or more search queries may be determined to share a line of inquiry when it is determined that they are within a threshold level of semantic similarity to one another. Once a shared line of inquiry has been identified and a combined search query generated, users may interact with the search parameters and/or the search results to update the search parameters of the combined search query.

Abstract: A resource search method and a related product. The method comprises: receiving a search request sent by a mobile terminal, the search request including a target search word; searching in a preset application library based on a first search algorithm and the target search word to obtain a first search result set; when a conversion rate of the target search result set is less than a first preset threshold, searching in the preset application library based on a second search algorithm and the target search word to obtain a second search result set; performing merge according to the first search result set and the second search result set to obtain a third search result set; and pushing the third search result set to the mobile terminal and instructing the mobile terminal to display the third search result set.

Abstract: A messaging system automatically populates a stream of messages using only a seed selected by the requesting account holder. In one embodiment, the seed includes the streams of one or more of the “top” accounts of the messaging system . Here, “top” is according to any one of a number of different metrics stored in the messaging system. With knowledge of the seed, the messaging system automatically populates a stream for the requesting account holder, without requiring any other input. As a result, an account holder is provided with a fully functioning stream with very little effort or knowledge required on their part.

Abstract: A system includes a network interface configured to access an enterprise data warehouse and an enterprise abbreviation list that maps enterprise terms to abbreviations. A processing device can communicate with the network interface and a memory device that stores instructions to analyze the enterprise data warehouse to determine name attribute scores based on occurrences of the enterprise terms and the abbreviations in the enterprise data warehouse. Execution of the instructions can also generate a scoring summary of phrases, apply fuzzy reasoning logic to identify one or more relationship patterns and weights for the phrases including at least one shared word to produce training data for a data model associated with the enterprise data warehouse, and update the data model with a new abbreviated field name associated with a new field name based on identifying a closest match of the new field name with the training data.

Abstract: A neural network machine translation method comprises: obtaining a to-be-translated source sentence; converting the source sentence into a vector sequence; determining candidate objects corresponding to the vector sequence according to a prefix tree which is pre-obtained and built based on a target sentence database, and determining a target sentence as a translation result according to the candidate objects.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for determining a plurality of sibling search queries for an input search query. In one aspect, a method comprises: receiving an input search query that satisfies a context template comprising a sequence of one or more words and a wildcard, wherein a wildcard represents variable data, wherein the input search query satisfies the context template and comprises a target word sequence that corresponds to the wildcard in the context template; and determining a plurality of sibling search queries for the input search query, wherein each sibling search query satisfies the context template and comprises a sibling word sequence that corresponds to the wildcard in the context template.

Abstract: A messaging system automatically populates a stream of messages using only a seed selected by the requesting account holder. In one embodiment, the seed includes the streams of one or more of the “top” accounts of the messaging system. Here, “top” is according to any one of a number of different metrics stored in the messaging system. With knowledge of the seed, the messaging system automatically populates a stream for the requesting account holder, without requiring any other input. As a result, an account holder is provided with a fully functioning stream with very little effort or knowledge required on their part.

Abstract: Systems and methods for recommending a media asset relating to a character unknown to a user are provided herein. The systems and methods may receive a first media asset viewed by a user, determine a first character in the first media asset, and include the first character in the character viewing profile for the user. Further, the systems and methods may receive a second media asset, including the first character, determine that a second character appears in the second media asset, and determine that the second character is not in the character viewing profile. The systems and methods may recommend a third media asset to the user for the second character.

Abstract: A system to collect and store in a special data structure arranged for rapid searching massive amounts of data. Performance metric data is one example. The performance metric data is recorded in time-series measurements, converted into unicode, and arranged into a special data structure having one directory for every day which stores all the metric data collected that day. The performance metric data is collected by one or more probes running on machines about which data is being collected. The performance metric data is compressed prior to transmission to a server over any data path. The data structure at the server where analysis is done has a subdirectory for every resource type. Each subdirectory contains text files of performance metric data values measured for attributes in a group of attributes to which said text file is dedicated. Each attribute has its own section and the performance metric data values are recorded in time series as unicode hex numbers as a comma delimited list.

Abstract: Methods, systems, and computer program products for software-defined database replication for a plurality of databases. Labels are applied to each of the plurality of databases and linking rules including conditions relating to the labels are used to define database replication links.

Abstract: Methods, systems and apparatus for identifying advertisements previously presented when processing related search queries related to a submitted search query, and providing the advertisements identified from the related search queries for presentation on a search results page for the submitted search query.

Abstract: Embodiments relate to configuring artificial-intelligence (AI) decision nodes throughout a communication decision tree. The decision nodes can support successive iteration of AI models to dynamically define iteration data that corresponds to a trajectory through the tree.

Abstract: There is provided a computer-implemented method of providing a set of alternative search results, the method including the steps of: (i) receiving a search request; (ii) searching a database to provide a first set of a plurality of search results, each search result including a plurality of attributes; (iii) identifying a best search result as the initially highlighted result; (iv) providing the highlighted search result for display; (v) receiving a selection of an attribute in the highlighted search result; (vi) providing the set of alternative search results, in which the selected attribute is varied with respect to the highlighted search result, and (vii) providing the set of alternative search results for display. Related computer program products, servers and systems are also provided.

Abstract: A cyber attack evaluation method includes transmitting, based on first domain information included in cyber attack information, an inquiry about whether a first IP address associated with the first domain information is stored to a first management server configured to store associations between domain information and IP addresses, transmitting another inquiry about a first answer history related to the first domain information to a second management server configured to store answer histories of the first management server by monitoring communication of the first management server, the answer histories being related to the associations between the domain information and the IP addresses, and outputting, based on an answer of the inquiry acquired from the first management server and the first answer history acquired from the second management server, a result of diagnosing threat content of a cyber attack related to the first domain information.

Abstract: System and method to determine a reduced cohort criteria, the method including: defining N selection criteria to select a cohort from among a universe of patient data; querying a patient database, by use of a processor, and by use of the N selection criteria, in order to define a full patient population; selecting a subset of size M of the N selection criteria, to produce a subset criteria; selecting a permutation of the subset criteria, to produce a permuted subset criteria in a predetermined order; for each member of the permuted subset criteria: querying the patient database by use of the member of the permuted subset criteria to produce a respective interim patient population; combining all respective interim patient populations to produce a partial patient population; and calculating, by a processor, a coverage figure of merit that compares the partial patient population to the full patient population.

Abstract: A method and system provide for entering text into a text field by determining complete-text terms, displaying the complete-text terms, and receiving a selection of a complete-text term as text to be entered into the text field. In some cases, a complete-text term can include all prior text entries used to determine the complete-text terms. In other cases, a complete-text term can include only a portion or none of the prior text entries used to determine the complete-text term. The determination of the complete-text terms can be based on a vehicle identifier. The determination of the complete-text terms can be based on one or more complete-text term selection-limiters, such a location, use, or diagnostic trouble code associated with the vehicle. The system can include a vehicle service tool that requests a set of complete-text terms and receives the set from a complete-text term storage device or an off-board processor.

Abstract: Methods and systems are provided for narrowing a scope of a search on a computing device to provide relevant search results to the user. Prior to receiving a search query from the user in a search box of a desktop taskbar, scope options are communicated for presentation from which the user can select. These scope options enable the user to select from web-based searches and local searches. A selection is received form the user of one of the scope options. Suggested search results are generated based on this scope option selection. At least one suggested search results is communicated for presentation on a display of the computing device.

Abstract: Media unit retrieval methods, systems and computer program products are provided that allow a user to search for an item by iteratively presenting media units such as images representing items to the user and receiving user input consisting of selections of the presented media units (including possibly the empty selection). Features, or attributes, a user is interested in, for example semantic features, are inferred from the interaction and media units are retrieved for presentation based on similarity with user-selected media units, through sampling of a probability distribution describing the intent or interests, or combinations of approaches. Accordingly, the user-experience is akin to a conversation about what the user is looking for. Retrieval may be based on both selected and unselected media units and the selection may comprise making a selection with a single action. Further, a database of media units can capture similarity relationships for efficient media unit retrieval.

Abstract: Methods and devices for determining whether a local version of content is stored on an electronic device associated with a user account on a backend system are described herein. In a non-limiting embodiment, the backend system may track and monitor the content stored on the electronic device using the associated user account. If an individual speaks an utterance requesting a particular content item, the backend system may determine, prior to sending the content to the electronic device, whether a local version is stored within the electronic device's memory. If so, the backend system may instruct the electronic device to output the local version, thereby reducing the amount of bandwidth consumed. The backend system may further be capable of predictively generating and then caching certain audio data to the electronic device.

Abstract: Recommending applications is disclosed. An application graph is disclosed that represents applications. Each node of the graph corresponds to an application and edges relate applications that can handle the same file type. When an input application is identified, the graph can be used to recommend other applications that may be a suitable replacement for the input application. The recommendation is based on the graph and its links and on characteristics of the organization.

Abstract: Methods, systems, and computer programs are presented for improved search methods based on career transition embeddings. One method includes an operation for generating career transition vectors for members of an online service, each career transition vector comprising identifiers associated with the career transitions of each member. The method further includes operations for performing a similarity analysis of the career transition vectors to generate an embedding vector for each identifier, detecting access of a first member to a job search user interface, and selecting one or more top embedding vectors based on one or more embedding vectors of the first member. One or more search starters associated with the one or more top embedding vectors are generated, and the one or more search starters are presented on the job search user interface.

Abstract: A method for resource search and related products are provided. The method is applicable to a server and includes the following. A search request is received from a mobile terminal, where the search request includes a target search term. Search for a target search result set from a preset application library according to the target search term. The target search result set is pushed to the mobile terminal and the mobile terminal is instructed to display the target search result set, when the target search result set has a conversion rate greater than a first threshold value, where the target search result set is displayed with preset display contents, and the preset display contents at least include a conversion rate.

Abstract: Moments of interest of an event may be automatically detected, encoded as snippets, and distributed to users subscribed to a channel of this event. In various embodiments, a moment of interest of an event is to be automatically detected, e.g., based on a near real-time cue on a social network or features in the event transmission. Further, a snippet can be encoded from the transmission of the event based on the detected moment of interest. The snippet may be put on a market place to be bid by various content distributors. Eventually, the snippet may be distributed to an interested user, e.g., via a channel subscribed by the user. Other embodiments may be described and/or claimed.

Abstract: Various aspects of the subject technology relate to systems, methods, and machine-readable media for relevance feedback with faceted search interface. A system provides a first search query to an image search engine for initiating an image search, where the first search query indicates a first search term. The system obtains image metadata from an image collection, and obtains a listing of query refinement terms associated with the first search query from the query refinement data structure. The system determines a listing of search results containing images with one or more facets associated with the listing of query refinement terms. The system then receives user input indicating selection of at least one of the one or more facets of an image in the listing of search results. The system provides a second search query using the first search term and a second search term indicating the selected one or more facets.

Abstract: Certain aspects of the present disclosure provide techniques for generating a replacement sentence with the same or similar meaning but a different sentiment than an input sentence. The method generally includes receiving a request for a replacement sentence and iteratively determining a next word of the replacement sentence word-by-word based on an input sentence. Iteratively determining the next word generally includes evaluating a set of words of the input sentence using a language model configured to output candidate sentences and evaluating the candidate sentences using a sentiment model configured to output sentiment scores for the candidates sentences. Iteratively determining the next word further includes calculating convex combinations for the candidate sentences and selecting an ending word of one of the candidate sentences as the next word of the replacement sentence. The method further includes transmitting the replacement sentence in response to the request for the replacement sentence.

Abstract: A system includes a processor configured to wirelessly receive data indicating vehicle-feature usage for an individual vehicle. The processor is also configured to aggregate received data to form a feature-usage customer profile defining feature preferences. The processor is further configured to select vehicles associated with a customer-classification, including predefined feature-usage characteristics, the customer-classification determined based on a correlation between the predefined feature-usage characteristics and the aggregated data in the feature-usage profile defining feature preferences. The processor is also configured to compare the aggregated data to the selected vehicles to determine a vehicle having features preferred by a customer as indicated by the aggregated data in the feature-usage profile and recommend the determined vehicle to the customer.

Abstract: Embodiments describe systems and methods for identifying temporal demand for queries and using metadata to modify autocomplete results. In one embodiment, a record of historical queries is stored by a system and analyzed to identify periodic or repeated events where demand for autocomplete results associated with one or more categories deviates from normal demand. A temporal model based on this record is used to adjust autocomplete search results during subsequent time periods associated with the repeated events.