Abstract: A learning device calculates a feature of each data included in a pair of datasets in which two modalities among a plurality of modalities are combined, using a model that receives data on a corresponding modality among the modalities and outputs a feature obtained by mapping the received data into an embedding space. The learning device then selects similar data similar to each target data that is data on a first modality in a first dataset of the datasets, from data on a second modality included in a second dataset of the datasets. The learning device further updates a parameter of the model such that the features of the data in the pair included in the first and the second datasets are similar to one another, and the feature of data paired with the target data is similar to the feature of data paired with the similar data.

Abstract: Generating a first signature value corresponding to a first dataset stored in a first data region having a first access permission level. Determining a signature anonymization value. Truncating the first signature value based on the signature anonymization value. Generating a secure search query for identifying, based on the truncated first signature value, second signature values corresponding to entries of a second dataset stored in one or more second data regions having one or more second access permission levels. Receiving a response to the secure search query including the plurality of second signature values. Determining the entries of the second dataset from the second signature values. Comparing the respective entries of the second dataset to the at least a portion of the first dataset. Generating one or more notifications based on the comparisons indicating a potential match between a particular respective entry of the second dataset and the at least a portion of a first dataset.

Abstract: The disclosed embodiments include a method for providing a cloud computing semantic layer operation. The method can comprise receiving a description of a semantic layer operation, the description specifying performance of the semantic layer operation by the semantic layer using a first cloud operation performed on a first cloud computing platform; receiving a first resource usage indication corresponding to the first cloud operation, wherein the resource usage indication comprises a direct or indirect indication of least one of a storage amount, a compute amount, a number of transactions, a bandwidth, a number of application programming interface calls, or an application or database type associated with the first cloud operation; receiving a request that indicates performance of the semantic layer operation; providing an estimated resource usage indication; receiving confirmation of the request; and communicating with the cloud provider to perform the semantic layer operation.

Abstract: The disclosed embodiments include a method for providing cloud computing semantic layer. The method comprises receiving a description of a semantic layer operation performed by a semantic layer, the description including multiple implementations of the semantic layer operation, the multiple implementations differing in at least one of cloud operations used or cloud computing platforms used; receiving a request that indicates performance of the semantic layer operation; receiving first resource usage indications corresponding to the cloud operations, wherein the first resource usage indications comprise direct or indirect indications of least one of a storage amount, a compute amount, a number of transactions, a bandwidth, a number of application programming interface calls, or an application or database type associated with each of the cloud operations: selecting a first one of the multiple implementations; and communicating with the first one of the cloud providers to perform the semantic layer operation.

Abstract: The disclosed embodiments include a method for providing cloud computing semantic layer. The method comprises receiving a description of a semantic layer operation performed by a semantic layer, the description including multiple implementations of the semantic layer operation, the multiple implementations differing in at least one of cloud operations used or cloud computing platforms used; receiving a request that indicates performance of the semantic layer operation; receiving first resource usage indications corresponding to the cloud operations, wherein the first resource usage indications comprise direct or indirect indications of least one of a storage amount, a compute amount, a number of transactions, a bandwidth, a number of application programming interface calls, or an application or database type associated with each of the cloud operations; selecting a first one of the multiple implementations; and communicating with the first one of the cloud providers to perform the semantic layer operation.

Abstract: In accordance with an embodiment, described herein are systems and methods for determining or allocating an amount, quantity, or number of compute instances or virtual machines for use with extract, transform, load (ETL) processes. In an example embodiment, a particular (e.g., optimal) number of virtual machines (VM's) can be determined by predicting ETL completion times for customers, using historical data. ETL processes can be simulated with an initial/particular number of virtual machines. If the predicted duration is greater than the desired duration, the number of virtual machines can be incremented, and the simulation repeated. Actual completion times from ETL processes can be fed back, to update a determined number of compute instances or virtual machines. In accordance with an embodiment, the system can be used, for example, to generate alerts associated with customer service level agreements (SLA's).

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for a source independent query language. One of the methods includes receiving a first executable statement, the first executable statement referencing a second set of fields in a table in a relational database, the first executable statement having instructions to cause the database to perform operations on data in the table. The method includes generating a second executable statement based on the first set of fields and the first executable statement. The method includes determining a mapping between the first set of fields and the second set of fields. The method includes specifying a derived table using the corresponding values and the mapping. The method includes generating instructions to cause the database to perform the operations on the derived table. The method includes sending the second executable statement to the database.

Abstract: Techniques support graph pattern matching queries inside a relational database management system (RDBMS) that supports SQL execution. The techniques compile a graph pattern matching query into a SQL query that can then be executed by the relational engine. As a result, techniques enable execution of graph pattern matching queries on top of the relational engine by avoiding any change in the existing SQL engine.

Abstract: In accordance with one disclosed method, a computing system may receive, via a first version of an application presenting an interface in a first language, a first query entered in a second language different from the first language. The computing system may search resources accessible to a second version of the application to identify at least a first resource corresponding to the first query, the second version of the application being in the second language. Based at least in part on a result of the searching, an indication of the first resource may be returned to the interface.

Abstract: An information processing apparatus has an inference unit with a first function of inputting first text data including a set of a first question text and a first answer text, and second text data including a set of a second question text and a second answer text; and a second function of, in a case in which a first synonym relationship in which a word a in the first text data is capable of being substituted with a word b is established and a second synonym relationship in which the word a in the second text data is capable of being substituted with a word c is established. A similarity between the word b and the word c is calculated from a first similarity between the first answer text and the second answer text and a second similarity between the first question text and the second question text.

Abstract: A software and hardware facility persistently maintains metrics on directories at different levels within a tree of a filesystem. The facility hierarchically aggregates attributes of files contained by directories and stores them as metric values in inode date of each directory within a tree. The stored metric values represent summed or otherwise aggregated data from the descendant directories and files. The metric values are utilized to provide representative compositional samplings of data in the filesystem tree structure. Based on the metric value in its inode data, each directory is weighted in a manner that determines the likelihood that each file sample is taken from that directory.

Abstract: Herein, a computer stores metadata that defines an analytic view that is based on organizing operation(s) that respectively specify organizing column(s) of organizing table(s), including a join operation that specifies a dimension column of a dimension table and a join column of a fact table. An original query is received that does not reference the analytic view. The original query specifies at least one organizing operation that is identical, somewhat similar, or compatible to an organizing operation on which the analytic view is based. Based on that at least one organizing operation that the original query specifies and the metadata that defines the particular analytic view, the original query is determined to satisfy rewrite criterion(s). In response to determining that the original query satisfies the rewrite criteria, a rewritten query that references the analytic view is generated based on the original query that does not reference the analytic view.

Abstract: A method, computer program product and computer system is disclosed that generates a set of distributed representation vectors from a dataset of textual and non-text data. In one method, a computer system receives a dataset, cleans the received dataset, parses the cleaned dataset to identify known classes of data, extracts data elements from the dataset based on the known classes of data, organizes the extracted data elements into one or more records, compiles a dictionary of unique data elements and associated codes from the one or more records, creates a set of training pairs using permutations of the codes that correspond to data elements within each record, and computes a distributed representation vector for each of the data elements in the dictionary using the set of training pairs.

Abstract: A method and device are provided for providing human network information. Information associated with user contacts is received from at least one external server. A first schema of the information associated with user contacts is converted into a specific schema. Relationship information related to a first user is generated based on the information associated with the user contacts converted into the specific schema. The relationship information related to the first user includes data related to the first user and data related to a relationship between the first user and other users.

Abstract: Some embodiments provide an automated method for processing a query. The method receives a query from a client, where the query references an object in a source database, the object having two sets of properties. The method generates a full set of properties by retrieving one set of properties from the target database and retrieving the other set of properties from a metadata storage. Based on the full, combined set of properties, the method generates a new query for execution on the target database.

Abstract: The disclosed embodiments include a method for providing cloud computing semantic layer. The method comprises receiving a description of a semantic layer operation performed by a semantic layer, the description including multiple implementations of the semantic layer operation, the multiple implementations differing in at least one of cloud operations used or cloud computing platforms used; receiving a request that indicates performance of the semantic layer operation; receiving first resource usage indications corresponding to the cloud operations, wherein the first resource usage indications comprise direct or indirect indications of least one of a storage amount, a compute amount, a number of transactions, a bandwidth, a number of application programming interface calls, or an application or database type associated with each of the cloud operations; selecting a first one of the multiple implementations; and communicating with the first one of the cloud providers to perform the semantic layer operation.

Abstract: The disclosed embodiments include a method for providing a cloud computing semantic layer operation. The method can comprise receiving a description of a semantic layer operation, the description specifying performance of the semantic layer operation by the semantic layer using a first cloud operation performed on a first cloud computing platform; receiving a first resource usage indication corresponding to the first cloud operation, wherein the resource usage indication comprises a direct or indirect indication of least one of a storage amount, a compute amount, a number of transactions, a bandwidth, a number of application programming interface calls, or an application or database type associated with the first cloud operation; receiving a request that indicates performance of the semantic layer operation; providing an estimated resource usage indication; receiving confirmation of the request; and communicating with the cloud provider to perform the semantic layer operation.

Abstract: A first information processing apparatus includes a record set acquiring unit that acquires a record set, a record identification information acquiring unit that acquires record identification information, a concealing unit that generates a concealment record set from the record set using conversion information and generates concealment record identification information based on the record identification information and the conversion information, and a first transmitter that transmits the concealment record set and the concealment record identification information to a second information processing apparatus. The second information processing apparatus includes a second receiver that receives the concealment record set and the concealment record identification information from the first information processing apparatus and a data analyzer that analyzes each record in the concealment record set identified based on the concealment record identification information using a learning model or a numerical model.

Abstract: Certain aspects of the present disclosure provide techniques for generating a user experience for a software program product based on a knowledge engine. Techniques for generating the user experience include a UI builder tool providing a set of tabular UI views and receiving in each tabular UI view corresponding input data for generating a calculation graph, a completeness graph, a client UI view, and a constants table. Based on the input data, the UI builder tool and knowledge engine can generate a set of artifact files. The knowledge engine can generate and/or execute the calculation graph(s), completeness graph(s), and constants table(s) as defined in the corresponding artifact files. The UI builder tool can generate an instance of the client UI view. With the generated calculation graph(s), completeness graph(s), constants table(s), and an instance of the client UI view, the user experience can be provided to a computing device.

Abstract: Disclosed herein are database metadata transfer systems and methods thereof. The database metadata transfer methods may include receiving a command to transfer preproduction metadata to production metadata; reading preproduction database metadata; generating a data file for making the change, the data file comprising SQL commands or insert statements; and executing a transfer from preproduction metadata to production metadata based on the SQL commands or insert statements. In some embodiments, the database may be a call center database and the data file may an Extensible Markup Language (XML) file.

Abstract: Systems and methods are provided for determining an alias file that defines aliases for one or more datasets. At least one Structured Query Language (SQL) query that includes one or more aliases defined in the alias file can be determined. The at least one SQL query can be translated to a corresponding executable SQL expression. The corresponding executable SQL expression can be executed.

Abstract: Systems, methods and computer program products for using searchable aggregate indices associated with non-aggregated value storage. In one method, a search system stores metadata values for each of a plurality of objects in a storage unit. The metadata values are stored in corresponding value storage locations that are associated with an identifiable metadata fields. An aggregate index is provided which includes a dictionary of terms that are contained in metadata values associated with a designated set of the metadata fields. The aggregate index is searched for one or more specific search terms, and one or more of the metadata values are retrieved from the value storage locations in response to the search, where the individual metadata fields associated with the retrieved metadata values are identified.

Abstract: Methods and systems are used for providing a domain specific language (DSL) for a cloud hosted system with metric-based state. As an example, a first DSL query request including first DSL query input based on a DSL is received. The first DSL query input is translated to a first set of query instructions based on the first DSL query input, translation rules metadata, and the DSL. The first set of query instructions is executed to retrieve, for each cloud system of a plurality of cloud systems, metrics data associated with an application state of an application deployed on each cloud system stored in a cloud deployment repository. A first set of response output is computed by matching the retrieved metrics data from the cloud deployment repository to the first DSL query input. A first DSL query response including the first set of response output is transmitted to a client system.

Abstract: The invention provides a vectorized representation method of a software source code. The vectorized representation method is an AST-based neural network which is a hierarchical vector representation method comprising the following implementation steps: step 1-1, converting an original software source code into an AST at the lowest layer, and then further dividing the AST according to source code statements to acquire a smaller statement tree sequence, wherein statement trees in the statement tree sequence are different in sequence, and the statement tree sequence is consistent with an original statement sequence; step 1-2, encoding the statement trees into statement vectors e1, e2, . . .

Abstract: A data query method and a data query system are provided. A data query server loads, based on a preset configuration interface, configuration metadata sent by a central cluster server for a target application programming interface (API), to a target storage region being located in a database server and associated with the target API, and loads queryable data associated with the target API to the target storage region. Then, the central cluster server sends a query instruction to the data query server based on query metadata in a data query request sent by a user terminal for the target API. After that, the data query server queries corresponding target query data in the target storage region and sends the target query data to the user terminal through the central cluster server. In this way, a data query service can be provided in a form of an API interface.

Abstract: A method for managing semantic information on an M2M/IoT platform is provided. The method for managing semantic information according to an embodiment of the present invention stores semantic data in the first attribute of an M2M resource and updates a part of the semantic data stored in the first attribute. Accordingly, efficient management of semantic information on an M2M/IoT platform is possible, and particularly a partial update of the semantic information can be performed.

Abstract: An e-commerce voice response determination system is provided and generally includes a server and a customer computing device. The server may determine, based on a user query, a plurality of product descriptions. The server may determine a similarity value between the user query and each of the product descriptions based on the similarity between each word of the user query and each word of the product descriptions. The server may determine the product description with the most similarity to the user query, and provide a response to the customer computing device based on that product description. In some embodiments, the server may determine whether the user query matches a predefined query. If the user query matches the predefined query, the server provides a response to the customer computing device based on a predefined response to the predefined query.

Abstract: Disclosed is a system for searching for an object based on color-sentiment, and a method thereof. The system for searching for an object based on color-sentiment includes: a search requestor terminal inputting, by a search requestor, at least one color of multiple colors to search for an object, and receiving multiple object search results based on sentiment information matched with the color; and an object search server transmitting, to the search requestor terminal, the multiple object search results based on an object search information database storing the sentiment information matched with the color.

Abstract: A domain-independent framework parses and interprets compound natural language queries in the context of a conversation between a human and an agent. Generic grammar rules and corresponding semantics support the understanding of compound queries in the conversation context. The sub-queries themselves are from one or more domains, and they are parsed and interpreted by a pre-existing grammar, covering one or more pre-existing domains. The pre-existing grammar, extended by the generic rules, recognizes all compound queries based on any queries recognized by the pre-existing grammar. Use of the disclosed framework requires little or no change in the domain-specific NLU handling code. The framework defines a generic approach to propagating context data between sub-queries of a compound query. The framework can be further extended to propagate intra-query context data in, out and across query components.

Abstract: A software and hardware facility persistently maintains metrics on directories at different levels within a tree of a filesystem. The facility hierarchically aggregates attributes of files contained by directories and stores them as metric values in each directory within a tree. The stored values represent summed or otherwise aggregated data from the descendant directories and files. The metric values represent aggregated attributes such as checksums, MIN and/or MAX aggregates, sameness bits, bloom filter aggregates, queryable user tags, moving average aggregates, b-tree aggregates, etc.

Abstract: In some examples, formatting semi-structured data in a database includes obtaining semi-structured data in a column store database, assigning the semi-structured data to a semi-structured column with a key and value format, and formatting the semi-structured data with an optimal search format.

Abstract: A business intelligence (BI) server is described that supports data and schemas stored in multiple languages. The BI server implements a lookup table and lookup function that allows users to work with queries in different languages. When the user logs in, a session object is created for the user, which maintains the state information. A session variable specifies the language currently being used by the user. The BI server can inspect this session variable to determine the language of the user and perform the lookup translations as necessary. For example, if the language used by the session is different from the language of the base table storing the necessary information, the BI server can perform a translation by invoking a lookup function. The execution of the lookup can include performing a join operation of the base table with the lookup table to yield a translated value requested by the query.

Abstract: A one-sided outer-join-based query is rewritten using a materialized view (MV), the definition of which includes the join but does not include a filter predicate from the query. The rewritten query nullifies data from the include-matching table, included in the MV, that does not satisfy the filter predicate. To improve accuracy of the query results, certain rows are removed from intermediate results of the query. To facilitate revising the query results for accuracy, the MV includes a unique column from the include-all table and also an indicator column that indicates whether a given row of the MV is an inner-join-type row or an anti-join-type row. The rewritten query adjusts the indicator values, in the indicator column, of MV rows that do not satisfy the filter to reflect a modified anti-join-type indicator value. Based on the modified indicator values and unique columns from include-all tables, accuracy of query results is attained.

Abstract: Generating a first signature value corresponding to a first dataset stored in a first data region having a first access permission level. Determining a signature anonymization value. Truncating the first signature value based on the signature anonymization value. Generating a secure search query for identifying, based on the truncated first signature value, second signature values corresponding to entries of a second dataset stored in one or more second data regions having one or more second access permission levels. Receiving a response to the secure search query including the plurality of second signature values. Determining the entries of the second dataset from the second signature values. Comparing the respective entries of the second dataset to the at least a portion of the first dataset. Generating one or more notifications based on the comparisons indicating a potential match between a particular respective entry of the second dataset and the at least a portion of a first dataset.

Abstract: A method of computing a join pattern agnostic aggregate function includes loading source code and parsing the source code to identify different database query operations. In response to the identification of an aggregate function in one of the different database query operations in which records in multiple tables are joined together in a table join, a multiplicity of primary keys are retrieved, each corresponding to a different one of the multiple tables and also an object of one of the tables referenced by the identified aggregate function. An existence of a fan out condition associated with the table join is then computed. On condition that an existence of the fan out condition is not computed a non-fan out sensitive implementation of the aggregate function is invoked with respect to the object. But, otherwise, a different, fan-out sensitive implementation of the aggregate function is invoked with respect to the object.

Abstract: The technology disclosed relates to formulating and refining field extraction rules that are used at query time on raw data with a late-binding schema. The field extraction rules identify portions of the raw data, as well as their data types and hierarchical relationships. These extraction rules are executed against very large data sets not organized into relational structures that have not been processed by standard extraction or transformation methods. By using sample events, a focus on primary and secondary example events help formulate either a single extraction rule spanning multiple data formats, or multiple rules directed to distinct formats. Selection tools mark up the example events to indicate positive examples for the extraction rules, and to identify negative examples to avoid mistaken value selection. The extraction rules can be saved for query-time use, and can be incorporated into a data model for sets and subsets of event data.

Abstract: Techniques for providing application contextual information. One or more sets of database context identifiers corresponding to events that occur within the database are generated by the database. The one or more sets of database context identifiers have at least one application context field. A session identifier corresponding to a session to be monitored is sent from the application to the database. Information to be stored in the database with the session identifier is sent to the database. Database logs and application logs are correlated using at least the session identifier.

Abstract: Systems, methods, and computer-executable instructions for using program synthesis for a database query includes analyzing a database query, written in a first language, to identify a first sub-query. A first synthesis template is constructed for the first sub-query. A first program, synthesized in a second language different from the first language, is synthesized using the first synthesis template. The functionality of the first program is equivalent to the first sub-query. The database query is rewritten to call the first program. The rewritten database query is executed.

Abstract: Database compiler with environment support may be provided. Database operators in a database compiler may be extended to include environment as an input parameter in addition to input data for native environment support. Environment operators may be added to the database compiler, which perform one or more of accessing and modifying the environment. Environment operators may include a return-environment operator, a map-with-environment operator and a compose-with-environment operator.

Abstract: Adapting database queries from a common interface into conditions specific to database storage formats, optimization of a data model, and virtualization of query results from combined databases.

Abstract: This invention is a system for integrating data sets organized in one organization type with data sets organized in a second organization type so that data queries submitted to be processed in the manner of the first organization type can be translated into queries usable by the data set in the second data organization type and the results returned to satisfy the first query.

Abstract: Disclosed herein are system, method, and computer program product embodiments for supporting different data sources provided in developing applications. An embodiment operates by a processor receiving the application descriptor, which includes a first data source for providing a first portion of data for the application, the first data source providing the first portion of data as a first data type that is incompatible with the application. The processor can then determine whether the first data source provides the first portion of data as the first data type and request based on the determination, the first portion of data from the first data source. The processor can thereafter receive a modified first portion of data in the second data and create a model for the application based thereon.

Abstract: A method for providing database management system support to a source application configured to operate with a source database management system Is provided. The method comprises detecting database management system communications issued by the source application to the source database management system; and responsive to said database management systems communications, impersonating the source database management system to the source application.

Abstract: A computing system uses neural networks to translate natural language queries to database queries. The computing system uses a plurality of machine learning based models, each machine learning model for generating a portion of the database query. The machine learning models use an input representation generated based on terms of the input natural language query, a set of columns of the database schema, and the vocabulary of a database query language, for example, structured query language SQL. The plurality of machine learning based models may include an aggregation classifier model for determining an aggregation operator in the database query, a result column predictor model for determining the result columns of the database query, and a condition clause predictor model for determining the condition clause of the database query. The condition clause predictor is based on reinforcement learning.

Abstract: In one embodiment, a method includes accessing a first set of objects, identifying one or more configuration files based on attributes of the first set of objects and attribute types specified by the configuration files, receiving from a user an input of a query constraint corresponding to one of the identified configuration files, identifying a second set of objects matching the first query constraint, and sending to the user search results corresponding to the second set of objects.

Abstract: Embodiments of the present invention provide systems and methods for data management. Synopsis tables have been found to be more effective for maintaining a high level of system performance while answering analytical queries. Synopsis tables, which contain MAX, MIN, and Bloom filter columns, may be modified by dropping ineffective data content within these columns and regenerating dropped data when beneficial. By automatically modifying data, database queries may be optimized.

Abstract: Embodiments of the present invention provide systems and methods for data management. Synopsis tables have been found to be more effective for maintaining a high level of system performance while answering analytical queries. Synopsis tables, which contain MAX, MIN, and Bloom filter columns, may be modified by dropping ineffective data content within these columns and regenerating dropped data when beneficial. By automatically modifying data, database queries may be optimized.

Abstract: Embodiments of the present invention provide a method and apparatus for managing opinion data. In an embodiment, there is provided a method for managing opinion data. The method comprises: acquiring an opinioned sentence from one or more text sources; extracting an opinion tuple based on the opinioned sentence, the opinion tuple at least containing an opinion word and an opinion target. The method further comprises: storing in association the opinioned sentence, opinion tuple and source information corresponding to the opinioned sentence, wherein the source information is associated with the text source from which the opinioned sentence is acquired. The opinioned sentence, the opinion tuple and the source information acquired from the text sources are stored in association by using for example an XML storage format so that the stored opinion data are associated to a certain degree, and the stored opinion data are easily extended and modified.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating a target sequence from an input sequence. In one aspect, a method comprises maintaining a set of current hypotheses, wherein each current hypothesis comprises an input prefix and an output prefix. For each possible combination of input and output prefix length, the method extends any current hypothesis that could reach the possible combination to generate respective extended hypotheses for each such current hypothesis; determines a respective direct score for each extended hypothesis using a direct model; determines a first number of highest-scoring hypotheses according to the direct scores; rescores the first number of highest-scoring hypotheses using a noisy channel model to generate a reduced number of hypotheses; and adds the reduced number of hypotheses to the set of current hypotheses.

Abstract: According to one embodiment, a method, computer system, and computer program product for acronym resolution is provided. The present invention may include receiving documents; identifying explicit expansions within the documents; receiving an input from a user; retrieving passages relevant to the received input from the documents; for each acronym within the one or more relevant passages, determining whether the acronym corresponds with explicit expansions within the relevant passages; for each of the acronyms that do not correspond with explicit expansions, determining whether the acronym corresponds with implicit expansions within the relevant passages; and for each of the acronyms that do not correspond with implicit expansions, determining whether the acronym corresponds with acronyms within a universal acronym list, and transmitting the one or more resolved acronyms to a question answering system.