Abstract: An electronic message comprising text generated by a first user can be identified and at least a second user can be identified as a recipient. At least one n-gram contained in the text of the electronic message can be identified by performing natural language processing on the text. A first user profile of the first user and a second user profile of the second user can be accessed. Based on the first user profile and the second user profile, at least one shared user context of the first user and the second user relating to the at least one n-gram can be determined. The at least one shared user context of the first user and the second user relating to the at least one n-gram can be presented to at least the second user.

Abstract: Aspects herein provide third-party application authors with a resolver chaining platform that simplifies the task of creating customized resolvers to gather information from user input while also allowing those authors to chain their custom resolvers with generic resolvers provided by the platform's host and that relate to commonly used parameter types. In particular, the present disclosure provides a method and system for authoring and using these resolver chains made up of a combination of custom and generic resolvers.

Abstract: Described herein are systems, methods, and techniques by which a processing unit can build an end-to-end dialogue agent model for end-to-end learning of dialogue agents for information access and apply the end-to-end dialogue agent model with soft attention over knowledge base entries to make the dialogue system differentiable. In various examples the processing unit can apply the end-to-end dialogue agent model to a source of input, fill slots for output from the knowledge base entries, induce a posterior distribution over the entities in a knowledge base or induce a posterior distribution of a target of the requesting user over entities from a knowledge base, develop an end-to-end differentiable model of a dialogue agent, use supervised and/or imitation learning to initialize network parameters, calculate a modified version of an episodic algorithm, e.g., the REINFORCE algorithm, for training an end-to-end differentiable model based on user feedback.

Abstract: A computing device can receive at least one document including text segments. The computing device can determine a characteristic pattern associated with a first text segment of the text segments and determine an association between the first text segment and a first individual class of a model based at least in part on the characteristic pattern. The computing device can determine a collection of classes of the model associated with the at least one document based at least in part on the association, and determine an actionable item associated with the at least one document based at least in part on the collection and at least one grammar pattern. The computing device can present or transmit an indication of the actionable item. In some examples, the computing device can arrange actionable items in an order based at least in part on occurrences of the actionable items from multiple documents.

Abstract: Multi-turn cross-domain natural language understanding (NLU) systems and platforms for building the multi-turn cross-domain NLU system are provided. Further, methods for using and building the multi-turn cross-domain NLU system are provided. More specifically, the multi-turn cross-domain NLU system supports multi-turn bot/agent/application scenarios for new domains without having to select a task definition and/or define a new schema during the building of the NLU system. Accordingly, the platform for building the multi-turn cross-domain NLU system that does not require the builder to select a task and/or build a schema for a selected task provides an easy to use, cost effective, and efficient service for building a NLU system. Further, the multi-turn cross-domain NLU system provides a more versatile NLU system than previously utilized NLU systems that were trained for and limited to a selected task and/or domain.

Abstract: A mechanism is provided for improving predicate parses (or logical representations of a passage) using semantic knowledge. In response to encountering an ambiguous decision point during a syntactic analysis of a portion of natural language content, a candidate meaning of the ambiguous decision point is generated. Characteristics of the ambiguous decision point are evaluated based on a semantic knowledge base to determine a semantic meaning associated with the ambiguous decision point. A determination is made as to whether the semantic meaning supports or refutes the candidate meaning. In response to determining that the semantic meaning refutes the candidate meaning, the candidate meaning of the ambiguous decision point is overridden based on the semantic meaning to include the semantic meaning as a final meaning for the ambiguous decision point. The portion of natural language content is then processed based on the final meaning for the ambiguous decision point.

Abstract: Embodiments for semantic hierarchical grouping of short text fragments by a processor. Sub-terms are extracted from a plurality of input text fragments according to a lexical sub-term hierarchy. Each of the sub-terms in the lexical sub-term hierarchy are matched with concepts based on an ontology of concepts representing a domain knowledge. The input text fragments are automatically grouped into a hierarchy of concepts based on the matching and a semantical relationship between each concept and matching sub-term.

Abstract: Systems and methods for extracting facts from natural language texts. An example method comprises: receiving an identifier of a token comprised by a natural language text, wherein the token comprising at least one natural language word references a first information object; receiving identifiers of a first plurality of words representing a first fact of a specified category of facts, wherein the first fact is associated with the first information object of a specified category of information objects; identifying, within the natural language text, a second plurality of words; and responsive to receiving a confirmation that the second plurality of words represents a second fact associated with a second information object of the specified category of information objects, modifying a parameter of a classifier function that produces a value reflecting a degree of association of a given semantic structure with a fact of the specified category of facts.

Abstract: A method may include receiving, by a device, an input sample of textual content. The method may include identifying, by the device, a comparison sample that is semantically similar to the input sample. The comparison sample may be identified based on a similarity score, of the comparison sample and the input sample, satisfying a semantic similarity threshold. The method may include identifying, by the device, a plurality of output samples of textual content based on acceptance information corresponding to the plurality of output samples and the comparison sample. The acceptance information may be determined based on a user input regarding similarity or relevance of the plurality of output samples and the comparison sample, and the user input may be received before the input sample is received.

Abstract: A method for message communication, includes: acquiring a communication message sent by a current contact in a communication application, the communication application having a messaging function; determining a message type which matches the communication message; acquiring at least one candidate word for replying to the communication message, in accordance with the message type; and displaying the at least one candidate word.

Abstract: An aspect includes receiving a description of a display screen generated by a software program executed in a first mode to interact with a user in first language. The description is in a serialized format that includes screen capture parameters and an offset location in a first dictionary corresponding to a screen element in the display screen. The first dictionary includes screen elements utilized by the software program when executing in the first mode to generate the display screen. A target language is identified, and a second dictionary of screen elements utilized by a deserialization program in the target language is accessed. The screen element is located at the offset location in the second dictionary. The description of the display screen is combined with the second dictionary, and the display screen is recreated in the target language based on the combining.

Abstract: A terminology marker system integrates a terminology analytical component for quantifying the amount of linguistic noise found in the translation output as measured against a dictionary; further, correlating the noise measured on a continuous basis enables the analytical component to build terminology predictive models used in a feedback loop to upstream components of the supply chain to improve future translation of new content. The system also provides a smart terminology assessment component for assessing linguistic assets and improving the quality of those assets to assist in translation. The system also provides a smart terminology evaluation component that is able to analyze MT output to make smart decisions on reducing the amount of post editing corrections needed for delivering a persistent level of translation quality.

Abstract: A multi-language translation converter comprising: a main body including a plurality of walls forming an interior volume and a front surface; at least one microphone member; at least one speaker member; at least one computer circuit board; and multi-language translation software. The multi-language translation software is encoded upon the at least one computer circuit board and is adapted to decipher language retrieved by the at least one microphone member and provide a translation in a pre-chosen language through text upon the liquid crystal display screen and audio through the at least one speaker member.

Abstract: Various approaches to provide ideogram translation are described. A communication application initiates operations to translate ideogram(s) upon detecting a message created by a sender that includes ideogram(s). A translation of the ideogram(s) is generated based on a content of the ideogram(s) and contextual information associated with the message. The contextual information includes a sender context, a recipient context, or a message context. The translation is provided to the recipient for display.

Abstract: Various embodiments for managing migration between server systems are presented. Indications of a plurality of migration scripts to be executed during a migration of a database system from a source server system to a target server system may be received. An execution order for the plurality of migration scripts may be determined based upon characteristics of a plurality of database objects to be migrated using the plurality of migration scripts and characteristics of the plurality of migration scripts. Execution of a first migration script of the plurality of migration scripts may be initiated according to the determined execution order. The first migration script may cause copying of a database object of the plurality of database objects from the source server system to the target server system.

Abstract: Methods and systems for log management include pre-processing heterogeneous logs and performing a log management action on the pre-processed plurality of heterogeneous logs. Pre-processing the logs includes performing a fixed tokenization of the heterogeneous logs based on a predefined set of symbols, performing a flexible tokenization of the heterogeneous logs based on a user-defined set of rules, converting timestamps in the heterogeneous logs to a single target timestamp format, and performing structural log tokenization of the heterogeneous logs based on user-defined structural information.

Abstract: Embodiments include methods, and a file system maintenance manager, and computer program products for performing file system maintenance. Aspects may include: surveying, by a file system maintenance manager, available compute nodes, and determining an amount of file system maintenance work to be performed in an unprocessed work chunk pool. The aspect may include dispatching work chunks to the available compute nodes for performing file system maintenance. The aspect may also include monitoring status changes of the compute nodes, and adjusting the work chunks dispatched to each available compute node according to the status changes of the compute nodes. The aspect may further include detecting capacity and performance of each of compute nodes, classifying the compute nodes available into high speed, medium speed, and low speed categories, and dispatching unprocessed work chunks to each of compute nodes dynamically, according to the capacity and performance of the compute nodes.

Abstract: Aspects include creating a profile for a user from user account data and a user activity log, searching an indexed repository for keywords matching elements from the user account data and the user activity log, and retrieving documents from the indexed repository that are identified based on the searching. Aspects further include creating a custom portal for the user that includes the documents.

Abstract: In some implementations a computer-accessible medium includes a multimedia-document integration module that includes a heterogeneous distinction identifier between a first multimedia-document and a second multimedia-document and includes an integrator of the first multimedia-document and the second multimedia-document into an integrated multimedia-document, a data capture module that includes a data-extractor of the integrated multimedia-document, a query module that includes a query-generator encapsulated in a corresponding number of enquiry/inquiry transmissions, the enquiry/inquiry transmissions being short-message-service text-messages, a communication-subsystem that includes a transmitter of the enquiry/inquiry transmissions to an external device and receiver of an acknowledgement transmission or a negative-acknowledgement transmission from the external device, the acknowledgement/negative-acknowledgement transmission being a short-message-service text-message, and, a variance analytic module that includ

Abstract: According to embodiments, reconfiguration of lock masters for a cluster of nodes is optimized using coordinated hash indexes to index into the master hash table and into local hash tables stored on the nodes of the cluster. A DBMS uses a hybrid hash index, a portion of which represents a corresponding master hash index, to index into both the master hash table and into a local hash table for a given lock. The hash index used to store lock metadata in a particular local hash table bucket, on a particular node, encodes the lock master index, for a master hash table, to which the locks in the local hash table bucket correspond. Only the portions of the local hash table on the lock master that correspond to the index of the master hash table bucket need to be scanned in order to perform needed tasks for lock master reconfiguration.

Abstract: A multi-token matching scheme efficiently detects multi-token keyword matches in content received by a database system. A keyword list may include multi-token keywords. Head tokens from the multi-token keywords are converted into a head hashmap. Token chains identifying tokens and delimiters in the multi-token keywords are generated and linked to the head hashmap. Tokens in the content are compared with the head hashmap. The matching tokens and following segments in the content are compared with the token chains. Portions of the content matching the token chains are identified as keyword matches. The matching scheme may use reverse delimiter tries to detect content matches for multi-token keywords with leading delimiters.

Abstract: A system and method for providing multiple user access to a playlist is described. A master user device and a guest user device are connected to the same network. A guest user device may search for playlists available in the network, and select one of the available playlists that is created by a master user. The guest user may then modify the selected playlist by adding a media file to the playlist and selecting a position in which the media file is to be output in the playlist. The guest user device may add the media file to the playlist by sending a cloud link to the master user device. The media file to be added to the playlist must satisfy one or more rules associated with an eligibility criteria before the media file can be added to the playlist.

Abstract: Methods, systems, and media for presenting recommended content based on social cues are provided. In accordance with some embodiments, a method for presenting recommended content is provided comprising: receiving a query associated with a user; generating a list of relevant media content items based on the query; selecting items for presentation to the user based on social relevance scores based on social connections of the user that have consumed a relevant item and contextual relevance scores for each item; causing the selected items to be presented to the user; and indicating that a particular item was consumed by a particular social connection.

Abstract: An information processing apparatus includes a use determination processing portion, a first acquisition processing portion, and a time notification processing portion. The use determination processing portion determines whether or not its associated information processing apparatus is in use. The first acquisition processing portion, when a use request of its associated information processing apparatus is received in a state where the use determination processing portion has determined that its associated information processing apparatus is in use, acquires a planned use time of its associated information processing apparatus via an input operation performed on an operation portion of its associated information processing apparatus. The time notification processing portion notifies a first information processing apparatus that is a transmission source of the use request, of the planned use time acquired by the first acquisition processing portion.

Abstract: Systems, devices, and methods include generating, for at least one search term of a query, a first expanded set of search terms based on at least one related term of the at least one search term obtained from a knowledge base, generating a second expanded set of search terms by applying a statistical model to the search terms of the first expanded set based on terms included in an electronic document corpus stored in an electronic data storage, and generating a third expanded set of search terms based on search terms in the first and second expanded sets. For each search term of the third expanded set, a term score is determined based on occurrence of the search term in the electronic document. The term scores are combined to create a document score for the electronic document.

Abstract: Method and systems for parallel scoring an ensemble model are provided. Aspects include loading data into a first distributed data structure having a plurality of partitions, each partition having loaded data in the form of a set of pairs of data formed of a record to be scored and a partial score for that record. A component model in the ensemble model is selected and processing of the records carried out in parallel across the partitions including updating the partial score for each record. In response to a partial score for a record not meeting an accuracy threshold, the method retains the record in the first distributed data structure to be scored by a subsequent component model. In response to the partial score for a record meeting the accuracy threshold, the method moves the record and updated partial score to an output result data structure to provide a final score.

Abstract: Users are most likely to search for documents “around” them, such as documents which are created, read, or modified by other users they work closely or collaborate with, documents on sites the user frequents, and/or documents trending around them, in addition to documents they've created or interacted with themselves. Therefore, embodiments are directed to relevance ranking of query results based on external context. A search service may receive a query and an external context list, determine a relevance value for each entry in the external context list, and perform the query, where a search index comprising items may be returned in response to performance of the query. The search service may then match entries in the external context list against associated properties of the items, compute a score for the items based on the matching, and rank results of the query based on the score computed for the items.

Abstract: In one embodiment, a method includes receiving a search query inputted by a first user, wherein the search query comprises one or more n-grams; calculating a needle-confidence score for the search query that is calculated by a needle-intent classifier based on at least the n-grams of the search query and a language model analysis of the n-grams, and wherein the needle-confidence score represents a probability that the search query was intended as a needle search; classifying the search query as a needle search if the calculated needle-confidence score is above a threshold confidence score; and generating a plurality of search-result modules, each search-result module comprising one or more search results matching the search query, wherein one of the search-result modules is a social module, and wherein the number of search results in the social module is based on the classification of the search query as a needle search.

Abstract: Calculating a failure intensity value for a group of search sessions is provided. In one example, a method for calculating a failure intensity value for a group of search sessions is provided. The method includes obtaining, by a computing device comprising a processor device, the group of search sessions, where each search session of the group of search sessions includes user input that identifies a search query. The method further includes classifying each search session of the group of search sessions into one or more result categories, where each result category is indicative of an outcome of a search session. The method additionally includes calculating the failure intensity value for the group of search sessions based on a combination of weighted values of the group of search sessions, where the weighted values are based on the one or more result categories of the group of search sessions.

Abstract: A method, apparatus and computer program product are provided for ranking channel pages of a host system. Content channels may comprise content relating to a particular category or having a commonality. Explicit data provided by user input, and/or implicit data derived or received from a user device or third party may be used to generate explicit and/or implicit parameters. Based on the parameters, content channels may be ranked and displayed in an order based on the ranking, and/or a home content channel may be identified. The host system may therefore target a user with relevant content.

Abstract: There is provided a method of providing a ranked list of computer-implemented activities to a user. As a performed activity is received, a plurality of lists, which each of the lists defines a ranking of activities, is accessed. The rankings of different lists are calculated with respect to different time horizons. The positions in the lists are associated with scores which are updated in view of the received activity. Also, the rankings of the lists are updated in view of the received activity. The lists are then combined into a single list. The combination is based on the updated lists and the scores associated with the positions in the lists.

Abstract: Method and systems for parallel scoring an ensemble model are provided. Aspects include loading data into a first distributed data structure having a plurality of partitions, each partition having loaded data in the form of a set of pairs of data formed of a record to be scored and a partial score for that record. A component model in the ensemble model is selected and processing of the records carried out in parallel across the partitions including updating the partial score for each record. In response to a partial score for a record not meeting an accuracy threshold, the method retains the record in the first distributed data structure to be scored by a subsequent component model. In response to the partial score for a record meeting the accuracy threshold, the method moves the record and updated partial score to an output result data structure to provide a final score.

Abstract: Prioritizing items based on user activity includes determining a user interest based on a current user activity and prioritizing items in a list presented in a display based on the user interest.

Abstract: A system and method utilizes information collected and stored in a database, and dynamically and visually builds up an interior location map with information on occupants within the structure or building; the map is accessible upon request by internet and mobile device users.

Abstract: A system and method for orienting the presentation of a virtual environment with respect to multiple users in a shared virtual space is provided. The multiple users may be physically present in different physical spaces. For each of the multiple users, the system may detect physical constraints associated with the respective physical space, and may determine a longest, unobstructed physical path in the physical space based on an orientation of the user in the physical space and the associated physical constraints. A presentation of the virtual environment to the multiple users in the shared virtual space may then be oriented with respect to each of the multiple users so as to maximize interaction amongst the multiple users in the shared virtual space.

Abstract: An embodiment of the disclosure provides a method of integrating data across multiple data stores in a smart cache in order to provide data to one or more recipient systems. The method includes automatically ingesting diverse data from multiple data sources, automatically reconciling the ingested diverse data by updating semantic models based on the ingested diverse data, storing the ingested diverse data based on one or more classification of the data sources according to the semantic models, automatically generating scalable service endpoints which are semantically consistent according to the classification of the data sources, and responding to a call from the one or more recipient systems by providing data in the classification of the data sources.

Abstract: Techniques are disclosed for managing a large set of data items, such as electronic data interchange (EDI) transactions or e-mail messages. A data item in a data set is received. The data item is exchanged between a first entity and a second entity. A relationship measure is determined between the first entity and the second entity relative to hierarchical information between the first entity and the second entity. An importance measure of the data item is determined based on the relationship measure. The importance measure is assigned to the data item. Further, the data items may be aggregated using pattern matching techniques relative to the importance factor, the aggregation of which may be used for intelligent archiving or purging decisions.

Abstract: Embodiments relate to a computer system, computer program product, and method to process complex files, and specifically, to support read and write requests of a multi-object file. Upon receipt of a file, a computer system parses the file into two or more logical objects. Each logical object has an associated or inherent characteristic. Each of the logical objects is matched to a storage tier in a multi-tier storage array. Each logical object is then assigned to a tier based on one or more object characteristics in the matched storage tier, and stored in a decomposed format. In addition, an identification of each logical object, and the object assignment, is recorded in an index.

Abstract: A method of adding informative content to an electronic knowledge base includes generating a plurality of database tables from a plurality of websites, each database table based on a corresponding website, and individually processing each database table to identify potentially informative content from each database table. The method further includes identifying as informative content, the potentially informative content that is included in a first predetermined number of the database tables of the plurality of database tables, the first predetermined number being greater than one, and adding the informative content to the electronic knowledge base.

Abstract: A method includes determining a first content is being accessed, wherein the first content includes a first message posted by a first user in a platform for communication between a plurality of users. The method initializes a sentiment filter for the first content based on a set of parameters. Responsive to identifying a first word, the method determines whether the first word of the first message exceeds a negative sentiment threshold. Responsive to determining the first word of the first message exceeds the negative sentiment threshold, the method identifies one or more parameters for displaying the first message posted by the first user, wherein the one or more parameters alter how the first word and the first message are displayed with respect to the exceeding of the negative threshold. The method displays the first message posted by the first user according to the identified one or more parameters.

Abstract: Methods, systems and programing for providing query suggestions. In one example, a search query is received from a user. At least one rewritten search query is generated by replacing at least some text in the search query with a synonym of the at least some text. The synonym of the at least some text is obtained based on a plurality of previously submitted search queries. A plurality of query suggestions is obtained based on the search query and the at least one rewritten search query. At least a portion of the plurality of query suggestions is provided to the user.

Abstract: In one embodiment, a method includes accessing multiple posts of an online social network, each post being associated with a first entity of the online social network. The method also includes generating multiple candidate keywords by extracting one or more n-grams from content of the accessed posts and calculating a rank for each candidate keyword. The method also includes storing one or more candidate keywords in association with the first entity, each stored candidate keyword having a calculated rank higher than a threshold rank and sending, to a second entity of the online social network, one or more suggested queries associated with the first entity for display on the client system, each suggested query comprising one or more of the stored candidate keywords associated with the first entity.

Abstract: Systems and methods for query large database records are disclosed. An example method includes: obtaining a first search query including a first keyword; accessing a relational database that stores a mapping between one or more keywords and a data record location associated with a distributed file system (DFS). The data record location identifies a location on the DFS at which a data record matching the one or more keywords is stored. The method also includes, determining, using a relational database, a first data record location based on the first keyword; identifying a first data record based on the first data record location; and providing the first data record as a matching record responsive to the first search query.

Abstract: The present disclosure discloses a cloud file transmission method, a terminal, and a server. The method includes receiving a request for obtaining a file, where the request includes a file identifier of the file; obtaining saved first information by using the file identifier and an established first association relationship between the file identifier and the first information, where the first information is generated by performing calculation on raw data of the file, and the raw data is data that can distinguish the file from another file; sending the first information to cloud, so that the cloud finds a saved file according to the first information and an established second association relationship between the file and the first information; and receiving the file sent by the cloud. In this manner, a terminal can accurately obtain a corresponding file from cloud without being limited by an operation.

Abstract: An identifying data set is defined including semi-identifying attributes with semi-identifying attribute values corresponding to an entity. Descriptive records corresponding to the semi-identifying attributes are created. Based on the descriptive records, retention records with retention attributes are created. A composed retention attribute is generated based on applying cryptographic hash algorithms to the retention attributes stored in a database. A retention attribute description is generated with a category and payload corresponding to the composed retention attribute. The semi-identifying attributes with corresponding positions are stored in the retention attribute. Hashes corresponding to the retention attributes are computed based on a composing mechanism that includes logic to compute the hashes. The hashes are combined to generate the composed retention attribute. When a request is received to permanently delete the entity from the database, the entity may be permanently deleted from the database.

Abstract: Representations of hierarchical structures identified by source and target file storage directories are displayed on a display device. Inconsistencies are identified between the hierarchical structures identified by the source and target file storage directories. Graphical interconnections are displayed that extend at least partially between locations of at least some of the identified inconsistencies. Refactoring action indicia are displayed that are user selectable and are associated with defined actions performed by a file directory controller to modify the hierarchical structure identified by the target file storage directory to resolve at least some of the inconsistencies identified between the hierarchical structures identified by the source and target file storage directories.

Abstract: Storage virtualization techniques allow files and directories to be stored remotely, for example, by a cloud storage provider, but in a manner that appears to a user or application running on a local computing device as if the files are stored locally—even though the data of those files and directories may not be resident on the local computing device. That is, the contents of files and directories that may exist in the cloud look and behave as if they were stored locally on a computing device. The level of hydration of placeholders may be monitored and operations may be performed to reduce the impact of excessive hydration on the resources of the computing device.

Abstract: Embodiments for, in a shared storage environment, managing data replication between first and second sites of a distributed computing environment by one or more processors. Metadata is pre-seeded from the first to the second site as an assembled metadata map. Data blocks corresponding to the pre-seeded metadata not currently stored at the second site are determined by the second site using the metadata map within a deduplication environment. A transfer request for the data blocks is returned by the second site to the first site.

Abstract: Efficient sharing of documents via logical tagging is provided. A document tagging system allows a user to tag a document with a label that is associated with a virtual folder of documents in an online document system, and to provide one or more users with access to the virtual folder. Providing access to the virtual folder associated with the label provides access to the document and other documents tagged with the label. The document and other documents tagged with the label can be shared with other users without attaching documents to emails or creating duplicate copies of the documents. A user interface element is provided for allowing the user to enter or select one or more labels with which to tag a document. Various documents can be associated with a label regardless of the document type. Further, the documents may be stored across various repositories.

Abstract: A data replication system and process is disclosed. A device receives files respectively via a network from a remotely disposed computing device and partitions the received files into data objects. The device creates hash values for the first data objects and stores the data objects on remotely disposed storage systems at location addresses. The first device stores in records of a storage table, for each of the data objects, the hash values and corresponding location addresses. The device receives an indication to replicate a portion of the stored data objects and replicates the stored data objects by copying the indicated data objects stored on the first remotely located storage system to the second remotely located storage system, and after replicating the indicated data objects, copies or recreates the one or more hash values and the one or more location addresses and the one or more key-to-location (e.g.