Abstract: One embodiment provides a method, including: receiving, at an information handling device, user command input comprising identification of an object; receiving, at an information handling device, image input comprising at least one object; determining, using a processor, whether the object in the user command input is associated with the at least one object in the image input; and responsive to determining that the object in the user command input is associated with the at least one object, performing a function based on the determination.

Abstract: In order to solve a problem that the level of precision in inferring a crosslingual topic of documents or words in a non-parallel corpus associated in the document level is not high, a topic inferring apparatus includes: a word distribution information storage unit in which word distribution information is stored in association with each of two or more languages; a document topic distribution generating unit that acquires document topic distribution information of a multilingual document set; a segment topic distribution generating unit that acquires segment topic distribution information of each segment, using the document topic distribution information; and a word topic determining unit that determines, for each word contained in two or more documents contained in the multilingual document set, a topic of each word using the segment topic distribution information. Accordingly, it is possible to improve the level of precision in inferring a topic.

Abstract: In one aspect, a method includes receiving a first request for first supplemental content for use by a content-presentation device in performing a first content-modification operation, receiving a second request for second supplemental content for use by a tuner device in performing a second content-modification operation. The method also includes determining that the content-presentation device is connected to the tuner device, and based at least in part on the first and second requests, determining that the first and second requests are associated with the same modifiable content-segment, and, in response to the determinations, using one or more parameters as a basis to select for performance an action from a set of actions including a first action that facilitates performance of the first content-modification operation and a second action that facilitates performance of the second content-modification operation.

Abstract: Various systems and methods are provided for performing soft entity resolution. A plurality of data objects are retrieved from a plurality of data stores to create aggregated data objects for one or more entities. One or more retrieved data objects may be associated with the same entity, based at least in part upon one or more attribute types and attribute values of the data objects. In response to a determination that the one or more of the retrieved data objects should be associated with the same entity, metadata is generated that associates the data objects with the entity, the metadata being stored separately from the data objects, such that the underlying data objects remain unchanged. In addition, one or more additional attributes may be determined for the entity, based upon the data objects associated with the entity.

Abstract: An approach is provided for domain-specific chatbots that employ distributed natural language classifiers (NLCs). A NLC included in a first chatbot is executed, which generates a first intent of a question and a first confidence that the first intent is an actual intent of the question. Based on a determination that the first confidence does not exceed a threshold, the question is broadcasted to other chatbots. A second confidence that a second intent is an actual intent is received from a second chatbot in response to the broadcast, and is determined to exceed the threshold. An association among the question, the second chatbot, the second confidence, and the second intent is stored in a memory cache included in the first chatbot. A response to the question is generated based on the second intent. The response is presented to a user from the first chatbot.

Abstract: Embodiments provide a computer implemented method in a data processing system comprising a processor and a memory comprising instructions, which are executed by the processor to cause the processor to implement the method of identifying negative communications from a plurality of electronic communications, including: extracting a plurality of terms from the plurality of electronic communications, wherein the plurality of terms include subjects, action verbs, adjectives and business objects; identifying variations of each term and grouping the variations for each term to form a dictionary; generating a co-occurrence graph to illustrate relationships between different terms in the plurality of terms; assigning each term with an entity type; identifying a negative intent for each business object using the co-occurrence graph; providing one or more semantic rules, and identifying one or more negative communications having the negative intent based on the one or more semantic rules.

Abstract: A method and system are provided for implementing enhanced cognitive searches optimized to integrate deployment with development testing. An initial body of works is ingested into a system capable of answering questions. A series of queries is analyzed against the initial body of works utilizing a set of scorers utilizing criteria to form assessments, wherein each scorer uses the criteria against the query and the initial body of works to form the assessment. The assessments are analyzed to determine a usefulness of a set of entries in the initial body of works. Content are deployed using the identified selected first set of entries as deployment ready, enabling enhanced cognitive search results.

Abstract: A computer system compares databases, where each data record includes values for a plurality of fields, by applying database operations on database tables of the data records to reduce computations being performed. For each data record in a reference set, candidate matches are identified from among the data records in a target set by performing a join operation on the sets on one or more selected, normalized fields from the sets. This significantly reduced set of candidate matches can be further analyzed to determine how similar they are to the data record from the reference set.

Abstract: The described technology is generally directed towards garbage collecting content selection graphs and related data from in an in-memory content selection graph data store. When a set of content selection graphs expire, a more current content selection graph set becomes active, and the storage space (e.g., in a Redis cache) used by the expired content selection graphs is reclaimed via garbage collection. Some graphs can be replaced before use, referred to as orphaned graphs, and the storage space for any such orphaned graphs is also reclaimed during garbage collection. Also garbage collected is storage space including related data structures used to generate and validate graphs.

Abstract: A method for making a targeted offer to an audience of a population of entities (e.g., social network). The method involves retrieving, from one or more databases, a first set of information including activities and characteristics attributable to a first plurality of entities; generating a plurality of interaction associations based on at least one of selected activities criteria and selected characteristics criteria from the first set of information; and conveying to a third party one or more interaction associations to enable the third party to identify a second set of information including activities and characteristics attributable to a second plurality of entities. The second set of information has matching activities and characteristics to the activities and characteristics of the interaction associations.

Abstract: A system for adapting virtual reality (VR) or augmented reality (AR) content for learning based on a user's interests includes a personalization engine that determines a user's interests and knowledge level regarding a topic, a content modifier that modifies VR or AR content related to the topic according to the users interests and knowledge level, an object selector that selects VR and AR models used to teach the topic during a VR or AR session and modifies the VR and AR models with the modified content based on the user's interests and knowledge level; a VR engine that renders the modified VR and AR models into VR or AR images, and a VR display that displays the VR or AR images. The user's engagement is measured during interaction with the VR display and used to refine and improve an understanding of the user's interest and knowledge level regarding the topic.

Abstract: Systems, apparatus, and methods for determining unique contacts from a collection or pool of merchant data are discussed herein. Some embodiments may provide for an apparatus including circuitry configured to: access first merchant data associated with a first merchant; access second merchant data associated with a second merchant; determine a match score based the first merchant data and the second merchant data indicating a likelihood of the first merchant being the same as the second merchant; determine a match score threshold; determine whether the match score exceeds the match score threshold; and in response determining the match score fails to exceed the match score threshold, determine the first merchant as being different from the second merchant. Some embodiments may provide for techniques for machine learning with merchant data training sets to determine match scores.

Abstract: A system may generate all possible character mistakes in a first uniform resource locator associated with a first website, which may produce a set of unique and similar uniform resource locators associated with a set of similar websites. The system may execute machine vision algorithms to compare visual images of the first website and the set of similar websites, and identify a subset of similar websites, which may be undistinguishable from the first website. The system may block the subset of websites, and thereby prevent any user from accessing these fraudulent and malicious websites.

Abstract: Examples are disclosed that relate to grouping query clusters based on intent of a search query. One example provides a system for generating recommended queries related to a search query, the system including a logic subsystem, and memory holding instructions executable by the logic subsystem to receive the search query, execute the search query on an index of pre-computed query clusters to identify a baseline query cluster and one or more neighboring query clusters, form a data structure including queries from within the baseline query cluster and the one or more neighboring query clusters, traverse the data structure to form a hierarchical arrangement of queries, and send the hierarchical arrangement of queries to a computing device from which the search query was received.

Abstract: Methods, apparatus, and computer readable media are described related to utilizing a context of an ongoing human-to-computer dialog to enhance the ability of an automated assistant to interpret and respond when a user abruptly transitions between different domains (subjects). In various implementations, natural language input may be received from a user during an ongoing human-to-computer dialog with an automated assistant. Grammar(s) may be selected to parse the natural language input. The selecting may be based on topic(s) stored as part of a contextual data structure associated with the ongoing human-to-computer dialog. The natural language input may be parsed based on the selected grammar(s) to generate parse(s). Based on the parse(s), a natural language response may be generated and output to the user using an output device. Any topic(s) raised by the parse(s) or the natural language response may be identified and added to the contextual data structure.

Abstract: A social networking system suggests events for a target user based on stored data in the social networking system related to the target user and to events. The social networking system may suggest events based on the target user's affinity for, connections with, or interactions with objects in the social networking system connected to or otherwise associated with the events. For example, an event is suggested to a target user if users connected to the target user already accepted an invitation to the event. As another example, an event organized by a particular entity is suggested to the target user because of interactions between the target user and other content provided by the entity. Invitations to suggested events may be presented to the target user via a client device, allowing the target user to easily join a suggested event.

Abstract: Provided is a method for identifying problematic input fields in a webform. The method comprises obtaining a set of user interaction data for a plurality of user interactions with input fields of a webform. The user interaction data for each input field includes data corresponding to at least one aspect. The method comprises assigning an aspect value to each aspect of each user interaction with each input field. The method comprises aggregating aspect values into an aggregated aspect value for each input field, generating a score for each input field based at least in part on the aggregated aspect field, ranking the input fields by their respective scores to identify at least one particular input field in the webform as problematic, and indicating that the at least one particular input field in the webform is problematic.

Abstract: A messaging system includes features of gathering information regarding content accessed across multiple applications and/or devices and making that information available to account holders of the messaging system. The messaging system stores data related to accessed content in a plurality of indices and makes that data available to enable account holders to access previously-accessed content, even if the previous access occurred in a different software application, on a different device, or on a device with a different operating system. Account holders may also be provided with recommendations based on gathered information.

Abstract: Affinity scores are computed for a plurality of user accounts, each affinity score quantifies an affinity of a user account to a content creator of one or more digital objects hosted in a content hosting platform. Each affinity score is computed by collecting interaction data for the user account corresponding to the content creator, normalizing the collected interaction data using a total activity data of the user account on the content hosting platform and attribute data associated with the content creator, and generating the affinity score using the normalized interaction data.

Abstract: A big data analytics system includes an event monitoring controller that detects a user targeted event from an initial digital query data flow. An event conversion controller determines at least one contextual element associated with the at least one user targeted event, and automatically converts the at least one contextual element into a search criteria. An external data query builder controller automatically executes an external search query based on the search criteria. A data fusing controller automatically fuses the at least one contextual element generated from the at least one user targeted event, together with external search results generated from the external search query and the initial query data flow to generate a fused data file.

Abstract: An information processing apparatus includes a log acquirer, a first estimation model, a finder and a hint display controller. The log acquirer acquires a log at occurrence of a predetermined event. The log includes one or both of a user's input operation history and an application history. The first estimation model calculates, based on the input log, a first evaluation value for each of a plurality of hint items that are obtained by separating a plurality of hints into categories. Each of the plurality of hints is prepared as a usage hint for the user. The first evaluation value represents likelihood of each of the plurality of hint items is a hint item that includes a hint sought by the user. The finder finds, from among the plurality of hint items, one or more hint items, in each of which the first evaluation value is a first predetermined value or greater.

Abstract: A method and apparatus managing online transaction using a computer system are disclosed. According to the present invention, a target request is received by a CPU coupled to a main memory and a memory application co-processor via a memory bus. The CPU then stores the target request onto the memory application co-processor coupled to a storage class memory. The memory application co-processor then locates target contents, inside the storage class memory, where the target key-word is specified in the target request. The CPU or a coupled device then accesses the target contents associated with the target key-word inside the storage class memory directly without copying the target contents associated with the target key-word inside the storage class memory to or from the main memory.

Abstract: The present teaching relates to generating user profiles with semantic knowledge. A first information associated with a user is obtained. One or more entities are identified from the first information. The one or more entities are augmented based on second information to generate a set of augmented entities. The set of augmented entities are clustered into a set of hierarchical clusters. A set of user profiles is generated based on the set of hierarchical clusters so that the user profile is to be used to personalize content recommendation.

Abstract: Embodiments of the present disclosure disclose a method, an apparatus, a server, and a storage medium for recalling for a search. The method for recalling for a search includes: acquiring a search term inputted by a user; calculating a semantic vector of the search term using a pre-trained neural network model; and recalling, according to a pre-established index, target documents related to the semantic vector of the search term from candidate documents, the index being established based on the semantic vectors of the candidate documents, and the semantic vectors of the candidate documents being calculated using the pre-trained neural network model. The embodiments of the present disclosure may solve a problem in the existing method for recalling that the recalling accuracy is affected by failing to generalize semantics, to improve the accuracy of recalling for a search.

Abstract: A given query entity of a query database and a set of reference entities from a master database are accessed; each entity accessed corresponds to an entry in a respective database, which is mapped to a set of words that are decomposed into tokens. For each reference entity, a closest token is identified therein for each token of the given query entity, via a given string metric. A number of closest tokens are thus respectively associated with highest scores of similarity between tokens of the query entity and tokens of each reference entity. An entity similarity score is computed based on said highest scores. A reference entity of the master database is identified, which is closest to said given query entity, based on the entity similarity score. Records of the given query entity are linked to records of the master database, based on the closest reference entity identified.

Abstract: A computer-implemented method includes identifying a theme for a poetic composition. The method also includes building a datastore of text taxonomies that provide associative differences between two words in a plurality of words. The method also includes analyzing the associative differences to determine a distance from a nexus for each of the plurality of words, wherein the nexus is determined according to the theme, and wherein the distance from the nexus to one of the plurality of words describes a suitability of the one of the plurality of words for use in the poetic composition for the theme. The method also includes determining at least one associative word according to the distance from the nexus. The method also includes determining a rhyming word from the plurality of words according to the at least one associative word.

Abstract: A method may include obtaining a first byte stream from first document code and a second byte stream from second document code. The first document code has a document type and the second document code has the document type. The method may further include identifying, in the first byte stream, nonvisual noise corresponding to a custom byte code defined in a custom character encoding set. The nonvisual noise is invisible when rendering the first document code. The method may further include replacing, in the first byte stream, the custom byte code with at least one standard byte code defined in a standard character encoding set to obtain modified document code. The second document code uses the standard character encoding set. The method may further include comparing the modified document code with the second document code by comparing the first byte stream with the second byte stream.

Abstract: The present invention has been made to solve the above problem, and an object of the present invention is to provide an automatic analyzer that enables control related to a search screen display of an automatic analyzer.

Abstract: Described herein is a system and method for utilizing data flow analysis to perform data classification with respect to a source dataset and a generated derived dataset. A flow confidence for a field is calculated using an adaptive algorithm in accordance with the action performed and the derived dataset. An associated derived confidence for a particular tag is calculated in accordance with an associated confidence and the flow confidence. When the associated derived confidence is greater than or equal to a first threshold, the particular tag is copied to the derived dataset. In some embodiments, when the associated derived confidence is less than or equal to a second threshold, the particular tag is not copied to the derived dataset. Otherwise an action to be taken is identified. A response to the action is received and the adaptive algorithm is modified in accordance with the received response.

Abstract: A method of selecting content object tags for recommendation to a user includes having a taxonomy of tags for labeling content objects to be stored at a content hosting service; identifying a baseline subset of content objects based on a user context at the content hosting service; identifying a targeted subset of the baseline subset based on the user context; determining a tag score for each tag associated with the targeted subset of content objects; determining a maximum number of tags to be recommended to the user based on available space within a user interface of the user; selecting tags with the highest recommendation score from a number of different tag taxonomy branches, the number being no more than the maximum number; receiving tags selected by the user from among the recommended tags; and performing an operation on the content object corpus for the user context using the selected tags.

Abstract: There is provided a method and a system for generating clusters of documents using a combined metric parameter. A first document and a second document are received, and for a potential cluster including the first document and the second document: a first metric parameter indicative of a degree of complementariness of document content in the potential cluster is determined, a second metric parameter indicative of a degree of dilution of the document content in the potential cluster is determined. The combined metric parameter is determined based on the first metric parameter and the second metric parameter. A cluster is generated based on the combined metric parameter, where the cluster includes the first and second documents. Other document(s) or clusters may be added to the cluster by determining an updated combined metric parameter for a potential cluster and comparing the updated combined metric parameter with the combined metric parameter.

Abstract: A method of accessing a dataset to fulfill a query over an intended audience is provided. A user interface comprises input means for a user to select an intended audience dataset from a plurality of datasets which the user has permission to access. The user interface is configured to receive from the user a query defining an intended operation to be performed on entries of the intended audience dataset to generate the target audience. The query is received from the user and a supporting candidate dataset is selected. It is determined whether the selected supporting candidate dataset comprises entries which enhance the entries of the audience dataset for performing the intended operation. If so, the candidate dataset and the audience dataset are selected to contribute to performance of the operation. If not, a next candidate dataset is selected, and the determining step is repeated.

Abstract: Techniques are disclosed for toponym disambiguation. Toponym disambiguation can be performed for a set of geographic location data, such as placenames. A subset of the data and additional location information associated with the subset of the data can be initially determined. The remaining geographic location data in the set of geographic location data can be scored in order to determined additional location information for the remaining geographic location data. Additional location information for the remaining geographic location data can be determined based on calculated scores.

Abstract: Systems, methods, and non-transitory computer readable media are configured to generate multiple channel embeddings for a page of a social networking system. The multiple channel embeddings can be mapped to a shared embedding space. A page embedding for the page of the social networking system can then be generated.

Abstract: Embodiments of the present disclosure provide a method and device for managing document indexes. The method includes in response to a request for adding a target segment to a document set including a plurality of segments, determining an original segment associated with the target segment from the plurality of segments, partitioning the original segment into a first portion and a second portion, the first portion being used to update the original segment, the second portion being used as the target segment, and rebuilding indexes for documents in the target segment.

Abstract: Natural language processing includes processing unstructured data in response to a request for a conclusion to a line of inquiry to produce structured information that conforms to a computer readable information schema. Values obtained from the unstructured data are assigned to attributes of the structured information and determining a candidate conclusion to the line of inquiry based on the value-assigned attributes. A confidence score is computed that is indicative of a sufficiency of the candidate conclusion as the conclusion to the line of inquiry. One or more of the attributes are displayed to a user for modification responsive to the confidence score of the corresponding conclusion failing to meet a confidence criterion. The displayed attributes are ranked by relevance of the attribute to the corresponding conclusion. The conclusion to the line of inquiry is determined based on the modified values of the user-selected attributes.

Abstract: Disclosed is a method for suggesting revisions to a document-under-analysis (“DUA”) from a seed database, the seed database including a plurality of original texts each respectively associated with one of a plurality of final texts. The method includes tokenizing the DUA into a plurality of statements-under-analysis (“SU As”), selecting a first SUA of the plurality of SU As, generating a first similarity score for each of the plurality of the original texts, the similarity score representing a degree of similarity between the first SUA and each of the original texts, selecting a first candidate original text of the plurality of the original texts, and creating an edited SUA (“ESUA”) by modifying a copy of the first SUA consistent with a first candidate final text associated with the first candidate original text.

Abstract: The present disclosure relates to interaction based on live webcast. In an example, an interaction method includes: receiving a bullet-screen comment message transmitted by a live broadcast client; in response to identifying a reference word string from the bullet-screen comment message, comparing the bullet-screen comment message with answer data corresponding to an interaction task to obtain a matching result; and outputting the matching result.

Abstract: Disclosed is a system and method for efficient comparison of experimental research documents to identify original research work. The system when operated, provides a platform for comparing a user submitted document with documents existing in the platform and documents available in public domain. Further, the system is capable of identifying originality and the similarity of the user submitted documents, in comparison with existing documents, based on a calculated similarity index.

Abstract: A natural language understanding system that can determine an overall score for a natural language hypothesis using hypothesis-specific component scores from different aspects of NLU processing as well as context data describing the context surrounding the utterance corresponding to the natural language hypotheses. The individual component scores may be input into a feature vector at a location corresponding to a type of a device captured by the utterance. Other locations in the feature vector corresponding to other device types may be populated with zero values. The feature vector may also be populated with other values represent other context data. The feature vector may then be multiplied by a weight vector comprising trained weights corresponding to the feature vector positions to determine a new overall score for each hypothesis, where the overall score incorporates the impact of the context data. Natural language hypotheses can be ranked using their respective new overall scores.

Abstract: Deduplicated data is packed into finite-sized containers. A similarity score is calculated between files that are similarly of the deduplicated data. The similarity score is used for grouping the similarly compared files of the deduplicated data into subsets for destaging each of the subsets from a deduplication system to one a finite-sized container. The similarity score is used for grouping the similarly compared files of the deduplicated data into subsets for destaging each of the subsets from a deduplication system to one of the finite-sized containers. An indication is received by a user of which of the similarly compared files are to be grouped into the subsets for destaging each of the subsets from a deduplication system to one of the finite-sized containers. Transitive closures are used for assisting with using the similarity score for grouping the similarly compared files of the deduplicated data into the subsets.

Abstract: Systems, methods, and non-transitory computer-readable media can determine a set of potential visualizations for a data set based on a set of visualization templates. The set of potential visualizations is ranked based on ranking criteria. A set of visualization recommendations for the data set is determined based on the ranking.

Abstract: A method for generating and presenting a comment excerpt in an online publication based on a comment in a comments section is provided. The method may include determining whether a passage from the comment in the comments section is relevant to a sentence in the online publication, and in response to determining that the passage from the comment is relevant to the sentence in the online publication, extracting the passage from the comment. The method may further include determining the scope of the comment that is associated with the extracted passage, wherein determining the scope of the comment comprises determining a context associated with the extracted passage based on text surrounding the extracted passage. The method may further include, based on the determined scope of the comment, generating the comment excerpt that corresponds to the comment. The method may further include presenting the comment excerpt within the online publication.

Abstract: Message management and classification techniques are described. In one or more implementations, a message received from a sender for delivery via a user account is examined to extract one or more features of the message. A determination is then made as to whether the message corresponds to one or more categories based on the extracted features, the categories usable to enable features to be applied to the message in a user interface.

Abstract: The present disclosure involves systems, software, and computer implemented methods with facilitating the results of a search request by analyzing content that matches actions words. One example system includes operations to receiving a request via interactions with an interface from a client device, wherein the interface is associated with a repository of content and manages interactions between the client device and the repository of content based on the request from the user. Keywords are identified from the request. The keywords are matched from the request with at least one action from a repository of actions. In response to matching the keywords from the request with the at least one action, content documents are identified from the repository of content associated with the at least one action. The identified content documents are filtered based on the keywords from the request. The filtered content documents are provided to the client device.

Abstract: An exemplary point-of-interest identification system accesses data for a point-of-interest. The system generates a plurality of scores for the point-of-interest based on one or more of a plurality of predefined features found in the data for the point-of-interest and on a machine learning model. The system selects, based on the plurality of scores, a primary point-of-interest type for the point-of-interest. The system includes the selected primary point-of-interest in a dataset for the point-of-interest. Corresponding methods and systems are also disclosed.

Abstract: A content influencer scoring system may include influencer computers each associated with a respective content influencer having influencer historical performance data and legacy influencer content associated therewith. A remote server may obtain advertisement campaign data associated with an advertisement campaign and parse the advertisement campaign data for advertisement keywords. The remote server may match content influencers to the advertisement campaign data based on the advertisement keywords and, for each content influencer, generate an advertisement campaign score.

Abstract: A mechanism is provided in a data processing system for solving a textual logic problem. A logic problem solution engine in the data processing system receives an input text containing a logic problem. A parser executing in the data processing system parses the input text to identify entities in the input text. The logic problem solution engine performs cognitive processing on the input text to discover relationships between the entities. A statistical analysis component executing in the data processing system determines statistical relationships between pairs of entities in the input text to form a set of statistical relationships. The logic problem solution engine determines whether a given statistical relationship within the set of statistical relationships fits as a solution to the logic problem.

Abstract: Various aspects of the subject technology relate to systems, methods, and machine-readable media for clustering search results based on image composition. A system may, for each image in a set of images, determine a compositional vector representing one or more objects and corresponding locations within the image using a trained computer-operated convolutional neural network. The system may provide each image through a clustering algorithm to produce one or more clusters based on compositional similarity. The system may provide images from the set of images clustered by composition, in which the images include a different listing of images for each of the one or more clusters. The system may provide a prioritized listing of images responsive to a user search query, in which the prioritized listing of images includes a different listing of images for each cluster of compositional similarity based on the metadata of each image associated with the cluster.

Abstract: The present disclosure provides techniques for calculating an entity's cybersecurity risk based on identified relationships between the entity and one or more vendors. Customer/vendor relationships may impact the cybersecurity risk for each of the parties involved because a security compromise of a downstream or upstream provider can lead to a compromise of multiple other companies. For example, if organization A uses B (e.g., a cloud service provider) to store files, and B is compromised, this may lead to organization A being compromised (e.g., the files organization A stored using B may have been compromised by the breach of B's cybersecurity). Embodiments of the present disclosure further provide a technique for calculating a cybersecurity risk score for an organization based on identified customer/vendor relationships.