Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for improving recommendation to users in data intelligence systems. In one aspect, a method includes the actions of receiving a current observation describing an interaction of a user with a data intelligence system; identifying a current user interest based on the current observation, wherein past observations of the user with the data intelligence system are clustered to form user interests in a Markov model; using the Markov model and based on the current user interest, determining a next user interest from the user interests; extracting a one past observation from the determined next user interest based on a selection criterion and a threshold, wherein the selection criterion is based on how closely the at least one past observation matches the current observation; and sending a recommendation to the user based on the past observation.

Abstract: In one embodiment, a method includes receiving, from a client system of a first user of an online social network, a request for a user-list, generating a structured query corresponding to the user-list, where the structured query is selected from a plurality of structured queries, and identifying one or more second users of the online social network that match the structured query and have a social connection or non-social connection with the first user. The social connection between the first user and the second user being based on social interactions on the online social network. The non-social connection between the first user and the second user being based on non-social interactions on a third-party application on the client system and being accessed by the online social network from the third-party application. The method includes sending, to the client system, instructions for displaying the user-list to the first user.

Abstract: A computer system receives a definition of a group that includes an indication of a hierarchal rank of group members, a plurality of content items that correspond to the group, and identifying information indicating at least one group member that corresponds to a content item. A search input by a first user is detected and an indication of a hierarchal rank for the first user is stored. Determining a weight to assign to a search result includes comparing a hierarchal rank of at least one group member that corresponds to the first content item with the hierarchal rank of the first user.

Abstract: Speech-to-text analysis, emotion analysis, and time analysis are performed on a media file by a data processing system. The media file is divided, by the data processing system, into a first fragment and a second fragment based at least in part on the speech-to-text analysis, the emotion analysis, and the time analysis. Semantic analysis is performed, by the data processing system, on the first fragment and the second fragment based on a search query for the media file. A first weight is provided for the first fragment and a second weight is provided for the second fragment based on the semantic analysis, the emotion analysis, and the time analysis. The data processing system stores the first weight in association with the first fragment and the second weight in association with the second fragment.

Abstract: Techniques for providing insights for web service providers are disclosed. In some embodiments, a system, process, and/or computer program product for providing insights for web service providers includes monitoring user activities on a web site; generating a dashboard for displaying a summary of the monitored user activities on the web site; and generating a recommendation for improving performance of the web site for display on the dashboard. For example, the recommendation can be for improving performance of search results on the web site and/or for improving content navigation on the web site.

Abstract: Systems and methods for identifying search results in response to a search query are presented. More particularly, images are selected as search results, at least in part, according to an attractiveness value associated with the images. Upon receiving a search query, a set of content is identified according to the query intent of the search query and includes at least one image. The identified set of content is ordered according an overall score determined according to relevance and, in the case of the at least one image, according to an attractiveness value. A search results generator selects items from the set of content according to their overall scores, including the at least one image, generates a search results page, and returns the search results page to the requesting party.

Abstract: The present disclosure provides a method and apparatus for determining search results, a device and a computer storage medium, wherein the method comprises: obtaining a current query, a historical query sequence of the current query and candidate search results of the current query; inputting the current query, the historical query sequence of the current query and candidate search results of the current query into a search result-sorting model, scoring the candidate search results according to the search result-sorting model, and determining search results corresponding to the current query; wherein the scores rendered by the search result-sorting model for the candidate search results are determined according to a similarity degree between an integrated vector representation of the current query and the historical query sequence of the current query and vector representations of the candidate search results.

Abstract: A method of generating a recommendation for an item is provided. A method may include receiving search results based on a query for an item. The method may also include identifying a plurality of human-curated lists from the search results, wherein each human-curated list of the plurality of human-curated lists includes a plurality of human-recommended items. Further, the method may include determining, via the at least one processor, one or more scores associated with at least one of the plurality of human-curated lists and the plurality of human-recommended items. The method may further include generating, via the at least one processor, a recommendation including at least one recommended item based on the determined one or more scores. Furthermore, the method may include displaying the recommendation for the item.

Abstract: The present disclosure is generally directed to a data processing system for customizing content in a voice activated computer network environment. The data processing system can provide an improved voice-based interface by selected response attributes based on response weightings. The selection of predetermined attributes can reduce size of response data and reduce network transmissions by providing more succinct audio-based responses.

Abstract: A method for performing data mining based on search behavior of a user is provided. One embodiment includes acquiring search behavior information of users; determining financial instruments corresponding to search behavior of the user and determining user attention information corresponding to each of the determined financial instruments. Another embodiment includes acquiring search behavior information of users; determining financial need related characteristics of the users; and grouping the users into user groups according to the determined financial need related characteristics; and determining a financial instrument corresponding to each of the user groups by matching the financial need related characteristic of each of the user groups with instrument characteristics of financial instruments.

Abstract: A kiosk configured to provide live service and sales support with a wide variety of features targeting the rural market, specifically the rural cellular mobile device market.

Abstract: Methods and systems are disclosed using linear expressions for machine learning (ML) models for ranking search results. In one example, a method for a computer trains a ML model into a decision tree for ranking search results. The decision tree is converted into a linear expression including Boolean terms. The linear expression is transmitted to one or more search computers that use the linear expression to rank search results for a search query. According to another example, a method for a computer having a search engine includes receiving a linear expression including Boolean terms representing a decision tree. The search engine processes a search query and uses the linear expression to rank search results for the search query.

Abstract: A method for dynamically modifying audio playback of a video game is provided. The method includes ascertaining a game music data associated with the video game. The game music data may include a plurality of game soundtracks classified according to predetermined criteria. The method may include collecting feedback over a period of time while the user is playing the video game. The method may continue with determining, based on the feedback, one or more replacement soundtracks based on criteria associated with the one or more replacement soundtracks. The method may further include dynamically modifying the game music data while the user is playing the video game by replacing at least one of the plurality of game soundtracks with the one or more replacement soundtracks to obtain modified game music data. The modified game music data may be provided to the user while the user is playing the video game.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating subqueries from a query. In one aspect, a method includes obtaining a query, generating a set of two subqueries from the query, where the set includes a first subquery and a second subquery, determining a quality score for the set of two subqueries, determining whether the quality score for the set of two subqueries satisfies a quality threshold, and in response to determining that the quality score for the set of two subqueries satisfies the quality threshold, providing a first response to the first subquery that is responsive to a first operation that receives the first subquery as input and providing a second response to the second subquery that is responsive to a second operation that receives the second subquery as input.

Abstract: The present invention discloses a method for combining entry structures, by which similar entry structures can be automatically combined. By this method, the efficiency in both retrieving entries and acquiring knowledge is improved. In a technical solution, this method comprises: converting an entry structure into a text; combining, in the format of text and based on a cosine similarity between entries and a cosine similarity between entries and texts, a plurality of texts into one text; and converting the combined text back to the entry structure.

Abstract: Systems, methods and computer program products for selecting dealers based on characteristics of the dealers and the user. A vehicle data system collects dealer location and historical transaction data from external data sources and generates and stores an eligibility table that identifies a set of eligible dealers for each combination of user location and vehicle make. Eligible dealers are determined from the eligibility table using a location and vehicle make identified from a user request. Scores are determined for each eligible dealer based on a dealer scoring model using a binary choice model in the form of a logistic regression of market share, inventory, close rate, price and distance, and dealers are ranked by these scores. A presentation of dealers selected by rank and by closest location to the user is generated and provided to the user via an interface running on a computing device.

Abstract: A method implemented in a database system for providing an information feed from a social networking system is provided. The method comprises receiving a request from a requesting user for an information feed of the social networking system according to configuration information customized by the requesting user. The configuration information identifies a plurality of information sources and has a set of rules for identifying which updates from the identified information sources to include in the information feed. Each information source has one or more associated updates. The method further comprising selecting a first set of updates from a plurality of updates wherein each selected update complies with the set of rules identified in the configuration information, excluding a second set of updates from the plurality of updates wherein each excluded update does not comply with the set of rules identified in the configuration information, and generating the information feed.

Abstract: A data search device includes a prediction unit that predicts a resource amount to be used in query processing scheduled to be executed, using a relationship between query processing executed in a past and a resource amount used in the query processing, and a change unit that changes an upper limit value of a resource amount usable in query processing to the predicted resource amount to be used in the query processing scheduled to be executed for which execution has been permitted.

Abstract: Embodiments of the present invention provide systems, methods, and computer storage media directed at interpreting a query related to a dataset to reduce ambiguity of the query. In embodiments, a method includes determining potential meanings of words of a natural language query related to a particular dataset. A matching graph can be generated for the words of the query and the potential meanings for the words including similarity scores for the potential meanings. The method further includes reducing ambiguity of the matching graph by assigning selected potential meanings to the words. Upon reducing ambiguity of the matching graph, the matching graph can be used to provide a response to the query. Other embodiments may be described and/or claimed herein.

Abstract: The disclosed embodiments provide a system for processing a query for a ranking of candidates for an opportunity. During operation, the system obtains a set of parameters associated with a query for a ranking of candidates for an opportunity, wherein the set of parameters include a candidate, the opportunity, and a ranking category. Next, the system uses the set of parameters to obtain a set of pre-computed scores for use in generating the ranking and determine a number of online scores required to complete the ranking. The system then generates a set of calls to produce the determined number of the online scores. Finally, the system uses the pre-computed scores and the online scores to generate a result of the query.

Abstract: Methods and apparatus to apply household-level weights to audience measurement data at a household-member level are disclosed. An example method to determine demographics of populations to measure media audiences of populations includes determining demographics for members of a first household of a sub-population. First demographics of a first member of the first household are different than second demographics of a second member of the first household. The example method includes calculating a first household-level weight for the first household based on a demographics distribution of the sub-population and aggregate demographics of a population.

Abstract: The Messaging Search and Management Apparatuses, Methods and Systems (“MSM”) transforms message, ranking request inputs via MSM components into work graphs, ML structure input data, ML structure, ranking response outputs. A work graph generation request that includes group level access control data may be obtained. A set of metadata access control carrying messages, a set of users, a set of channels, and a set of topics with access control data corresponding to the group level access control data may be determined. A user priority score for each of the other users, a channel priority score for each of the channels, and a topic priority score for each of the topics, from the perspective of each user, may be calculated. A work graph data structure may be generated that includes, for each user, data regarding the calculated user priority scores, channel priority scores, and topic priority scores.

Abstract: Methods, systems, and techniques for performing a hash-based appearance search. A processor is used to obtain a hash vector that represents a search subject that is depicted in an image. The hash vector includes one or more hashes as a respective one or more components of the hash vector. The processor determines which one or more of the hashes satisfy a threshold criterion and which one or more of the components of the hash vector qualify as a scoring component. The one or more components that qualify correspond to a respective one or more hashes that satisfy the threshold criterion and that are represented in a scoring database that is generated based on different examples of a search target. The processor determines a score representing a similarity of the search subject to the different examples of the search target.

Abstract: In accordance with various embodiments, described herein are systems and methods for use of computer-implemented machine learning to automatically determine insights of facts, segments, outliers, or other information associated with a set of data, for use in generating visualizations of the data. In accordance with an embodiment, the system can use a machine learning process to automatically determine one or more outliers or findings within the data, based on, for example, determining a plurality of combinations representing pairs of attribute dimensions within a data set, from which a general explanation or pattern can be determined for one or more attributes, and then comparing particular values for attributes, with the determined pattern for those attributes. Information describing such outliers or findings can be graphically displayed at a user interface, as text, graphs, charts, or other types of visualizations, and used as a starting point for further analysis of the data set.

Abstract: Published multimedia including both images and associated text are used to train a neural network, or other machine learning system. The neural network is trained to facilitate the identification and selection of other images for association with other text, and subsequent publishing together in multimedia. The neural network is optionally configured to receive text, or a representation thereof, and generate an image feature vector in response. Embodiments include the use of the trained neural network to select images for publication in multimedia.

Abstract: A non-transitory computer-readable recording medium stores therein a query generation program that causes a computer to execute a process including: determining, from a plurality of category schemas each being a set of attribute schemas, a main category schema to be associated with an output target table that defines acquisition target data, based on attribute names of a plurality of output items contained in the output target table; determining a subcategory schema to be associated with an un-associated item, which is an output item that is not associated with the main category schema among the plurality of output items, from category schemas that are associated with respective features of a plurality of attribute schemas contained in the main category schema; and generating queries with respect to the plurality of category schemas based on the main category schema and the subcategory schema.

Abstract: Positional communication systems and methods include receiving a definition of a first geographic area including a first commercial location by a computer system and storing the first geographic area in a memory. The location of a mobile device in the possession of a customer is determined, and if the mobile device is determined to be at a location within the first geographic area, an incentive message for the customer to leave the location and go to a second commercial location is sent to the mobile device.

Abstract: Techniques for ranking search results generated by a search engine are described. A search query initiated by a searcher is processed to identify a set of member profiles satisfying the search query. A ranking score is assigned to each member profile of the set of member profiles. A subset of the set of member profiles is identified based on a determination of an overlap between a work history specified in each of the subset of member profiles and a work history specified in a profile of the searcher. The ranking score assigned to a member profile of the subset of the set of member profiles is adjusted based on the determination. A portion of each of the set of member profiles satisfying the search query is communicated for presentation as search results in a search results interface.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for receiving, at a search engine, a search query submitted by a user to the search engine; identifying an entity that is associated with the search query; determining that the user that submitted the search query to the search engine is associated with the entity in an entity listing portal; and in response to determining that the user that submitted the search query to the search engine is associated with the entity in an entity listing portal, providing a search results webpage that includes i) search results that the search engine generated responsive to the search query and ii) an interface through which edits to data associated with the entity within the entity listing portal can be provided.

Abstract: A user device that processes an input message includes a display unit configured to display multiple apps, an input unit configured to receive an input message from a user, an input message transfer unit configured to transfer the input message to the multiple apps and receives a return value of the input message from at least one of the multiple apps capable of processing the input message, a state changing unit configured to puts the at least one app into a standby state based on the return value, and an app execution unit configured to executes the app corresponding to an additional input message, if the additional input message relevant to any one app in the standby state is input.

Abstract: A user can issue a query about a process that has a number of stages. A stage of the process is determined using the query and location data associated with the query, and a stage prediction model. A stage learning system can select a sample of query logs for a category from a database of millions or billions of users' queries. Queries can be parsed into keywords. A category can be determined from location information associated with each query and from query keywords. Queries are aligned based on location and, optionally, keywords. TF-IDF values are computed for queries and are used to determine a difference significance between aligned, adjacent queries. If aligned, adjacent queries have a substantially difference in keywords and TF-IDF, then a conversion stage is identified. Content can be presented to the user based on the category, keywords, location, and conversion stage.

Abstract: The present disclosure provides apparatus, systems, and methods for crowdsourcing domain specific intelligence. The disclosed crowdsourcing mechanism can receive domain specific intelligence as a data processing rule module. For example, a data analytics system can request a crowd of software developers to provide a data processing rule module tailored to process a particular type of information from a particular domain. When the data analytics system receives the data processing rule module from one of the software developers for the particular domain, the data analytics system can use the received data processing rule module to process information associated with the particular domain.

Abstract: A computer device, a computer-readable medium and a computer-implemented method is provided. The computer device and the computer-readable medium perform the computer-implemented method of generating unregistered internet domain names using a trained encoder function and a trained decoder function. The method includes obtaining one or more seed domain names; mapping, using a hardware processor implementing the trained encoder function, the one or more seed domain names in a low-dimensional information-rich latent space; modifying a representation of one domain name of the one or more seed domain names that is encoded in the low-dimensional information-rich latent space; and generating, using the hardware processor implementing the trained decoder function, one or more unregistered domain names for domain name registration from the representation that is modified.

Abstract: A first element is extracted from a pair including a past inquiry and a past response, wherein the first element indicates that the past response shows an understanding of the past inquiry. A model is generated used to estimate a second element in a new inquiry based on the first element, wherein the second element indicates that a new response to the new inquiry shows an understanding of the new inquiry.

Abstract: A method and system for generating summaries of posts of interleaved text are provided. The method includes embedding, by a first neural network, each post through word-to-word encoding; embedding, by a second neural network, overall content of the plurality of posts through post-to-post encoding based on the word-to-word encoding of each post; generating, by at least a third neural network, a summary of the at least one thread through word-to-word decoding based on the overall content embedding of the plurality of posts; and displaying the summary of the at least one thread to a user.

Abstract: In one embodiment, a method includes receiving an text query, identifying and ranking multiple content objects corresponding to character strings from the text query. A computing system determines the ranking based on an analysis of the association between the terms in the content objects and the n-grams in the query using term frequency-inverse document frequency (TF-IDF) scores of the terms. The computing system accesses a query-term index to determine the TF-IDF scores for the terms, wherein the query-term index includes multiple entries corresponding to multiple terms extracted from multiple prior queries. Each entry in the query-term index includes a query term, one or more related terms extracted from content objects of the online social network matching the prior query term, and a term frequency-inverse document frequency (TF-IDF) score for the respective related terms.

Abstract: A system for providing integrated monitoring and communications of diagnostic equipment is disclosed. The system may comprise a data access interface, a processor, and an output interface. The data access interface may receive heterogeneous data from a plurality of machine and sensor equipment associated with performance of a system or product. The data access interface may also to receive a user inquiry pertaining to the system and product. The processor may generate a knowledge graph based on the data associated with the system or product, as well as convert the user inquiry into a knowledge graph query by: extracting entities from the user inquiry; extracting relations from the user inquiry to identify relationships between entities; expanding the user inquiry using the knowledge graph and the entities and relations; and translating the inquiry into knowledge graph triples.

Abstract: Implementations of a system, method and apparatus described herein receive, for a configuration item in a configuration management database, status indicating an operational state and an automation state associated with the configuration item, and determine whether a conflict will occur with at least one of the operational state or the automation state of the configuration item as a result of a process affecting the configuration item. When the conflict will not occur, a change is made to at least one of the operational state or the automation state of the configuration item in accordance with the process. Upon a conclusion of the process, the change is removed. If the conflict will occur, the process not allowed to continue.

Abstract: Techniques are provided that include receiving one or more global signatures for a query image in response to an image recognition query, wherein some of the plurality of global signatures are generated using local descriptors corresponding to different cropped versions of the image. A ranking order is determined for a plurality of document images based on nearest neighbor relations between document signatures corresponding to the plurality of document images and each one of the one or more global signatures for the query image. A subset of the plurality of document images is selected based on the determined ranking order. Additional document data corresponding to the selected subset of the plurality of document images is obtained, and a search result is generated based on a geometric verification between the additional document data corresponding to the selected subset of the plurality of document images and the query image.

Abstract: A system and method for determining color variants of a fashion product is provided. A system for determining a color variant of a fashion product is provided. The system comprises a categorization module configured to receive a plurality of images corresponding to a plurality of fashion products and configured to categorize the plurality of images into a plurality of styles. The system further includes a color variant determination module configured to compute a plurality of embeddings for the corresponding plurality of images, compute a mean value for the plurality of embeddings and compare each embedding to one or more predefined constraints to determine whether the embedding belongs to a color variant set.

Abstract: An image search apparatus including: an accumulation unit that accumulates search target images; an area detector that learns search priority by analyzing first information extracted from a plurality of first areas included in a plurality of the search target images, analyzes second information extracted from a plurality of second areas included in a query image, and detects the second areas, the search priority of which is high, as conspicuous areas by using the second information; and an image search unit that searches for the search target images by using an image feature extracted from the conspicuous areas.

Abstract: An image processing engine component generation method, and search method, terminal, and system are provided. The image processing engine component includes: an online feature processing component operative to receive an incremental image, and determine image content feature information of the incremental image; an offline feature processing component operative to receive a stock image, and determine image content feature information of the stock image; a search platform component operative to receive a search request, the search request including search text information; and a processing engine component operative to receive image content feature information of the incremental image and the stock image, add the received image content feature information to index tables, and determine search results corresponding to the search text information.

Abstract: Systems, methods, and software described herein manage the initiation of applications in a system of physical nodes. In one example, a control system may identify a triggering event to initiate an application in the system based on sensor data obtained from at least a subset of the physical nodes. In response to the triggering event, the control system may identify a physical node with an image to support the execution of the application, initiate execution of the application associated with the image, and allocate or provide state information to the executing application.

Abstract: According to one embodiment, in response to a search query received from a client, a search is performed in a content database to identify a list of one or more content items based on one or more keywords of the search query. A first search is performed in an image store to identify a first set of one or more images using a first image searching method. A second search is performed in the image store to identify a second set of one or more images using a second image searching method that is different than the first image searching method. A search result is transmitted to the client, the search result having at least a portion of the content items to the client. Each content item is associated with one of the images selected from the first set of images or the second set of images.

Abstract: Techniques for question answering involve receiving, from a user, a text input expressing a question in natural language. In response to the question, a text output expressing an answer to the question may be generated. A plurality of documents comprising natural language text may be analyzed, involving mapping the question to one or more hypotheses, analyzing at least one passage of text in at least one of the documents to determine whether the passage entails at least one of the hypotheses, and in response to determining that the passage entails at least one of the hypotheses, identifying the passage as providing supporting evidence for the answer to the question. The answer and the at least one passage identified as providing supporting evidence for the answer may be presented to the user in response to the text input.

Abstract: Devices, computer-readable media and methods for affecting mental states of a first user and a second user are disclosed. For example, a processor may receive first biometric data for a first user, quantify a mental state of the first user based upon the first biometric data, receive second biometric data for a second user, and quantify a mental state of the second user based upon the second biometric data. The processor may further select a first automated action to affect the mental state of the first user and the mental state of the second user, and implement the first automated action to affect the mental state of the first user and the mental state of the second user.

Abstract: One or more client devices, systems, and/or methods of improving mobile searches are provided. A feature (e.g., “snow boots”), generated on a client device, is identified. The feature is evaluated to identify a goal (e.g., finding snow boots) associated with the feature. A mission (e.g., finding outdoor apparel), associated with the goal, is identified. A query (e.g., “outdoor stores near Akron”) associated with a second goal (e.g., find outdoor apparel store) associated with the mission is identified using a model generated utilizing a machine learning method trained using a query-goal-mission structure. A query recommendation (e.g., “outdoor apparel store”) comprising the query is presented to the user. A user satisfaction metric, associated with the query, the query recommendation, a result generated by the query recommendation, etc. is determined based upon user interaction with the query recommendation and the result. The model may be tuned based upon the user satisfaction metric.

Abstract: Systems and methods for engagement-based routing transportation services are disclosed. A method includes establishing a communication session between a server computing device, client computing device, and plurality of geographic routing entities. The method further includes receiving a request for the geographic routing system from the client computing device. The method further includes identifying ones of the geographic routing entities within a determined geographic distance of the client computing device and obtaining a plurality of scores and a plurality of geographic distances from the client computing device. The method further includes transmitting the request to one or more of the identified geographic routing entities. The method further includes receiving acceptances of the request from at least a subset of the identified geographic routing entities.

Abstract: A digital medium environment includes an asset processing application that performs editing of assets. A projection function is trained using pairs of actions pertaining to software edits, and assets resulting from the actions to learn a joint embedding between the actions and the assets. The projection function is used in the asset processing application to recommend software actions to create an asset, and also to recommend assets to demonstrate the effects of software actions. Recommendations are based on ranking distance measures that measure distances between actions representations and asset representations in a vector space.

Abstract: A computer-implemented method for processing a query is disclosed. The method includes receiving, by a server, a query from a user device over a network, the query including one or more keywords; classifying, by the server, the query into one or more categories based on the one or more keywords of the query; determining, by the server, a confidence indicator associated with each of the one or more categories from the classifying; and modifying the query to boost the one or more categories when the confidence indicator associated with the one or more categories for one or more of the keywords of the query is greater than or equal to a confidence level threshold.