Abstract: Techniques for machine learning inferencing endpoint discovery in a distributed computing system are discloses herein. In one example, a method includes searching a database containing machine learning endpoint records having data representing values of execution latency or prediction accuracy corresponding inferencing endpoints deployed in the distributed computing system. The method also includes generating a list of inferencing endpoints matching the individual target values and determining whether a count of the inferencing endpoints in the generated list exceeds a preset threshold. In response to determining that the identified count does not exceed the preset threshold, the method includes instantiating one or more additional inferencing endpoints in the distributed computing system based on the individual target values in the received query.

Abstract: A method of processing database queries according to an embodiment of the disclosure includes the steps of acquiring an input query of a database, generating a join graph consisting of vertexes corresponding to the tables of the database and edges corresponding to join information among the tables, dividing the join graph into a plurality of subgraphs, and for the input query, generating an execution plan of a target query including N-ary join operators by using the plurality of subgraphs, optimizing the execution plan of the target query based on a cost model, acquiring a table sort algorithm of the N-ary join operators included in the target query based on an available size of a GPU or CPU memory, determining a sort order of one or more join columns included in the tables of the N-ary join operators, and based on the sort order of the one or more join columns, performing an operation for the plurality of subgraphs.

Abstract: A database management system for processing large volumes of data in a key-value store database is provided. The system may be configured to receive a plurality of filled fillable request forms where each request form may include a request including a plurality of field labels and a plurality of fillable text fields corresponding to each of the plurality of the field labels. The system may be configured to extract each set of inputted data from each fillable text field. The system may be configured to store, in the key-value store database, for each request form, each of the plurality of field labels and the corresponding set of inputted data as a combination key-value pair. The combination key may be equal to a WIP ID number, form ID number and field ID number. The corresponding value may be equal to the set of data of the corresponding field ID number.

Abstract: A method and apparatus of a device that transforms data is described. In an exemplary embodiment, the device receives data from the plurality of data sources, wherein the plurality of data sources has different data formats. In addition, for each of the plurality of data sources, the device selects a first set of columns for transformation for that data source and expands the selected first set of columns into a set of rows, wherein each of the rows has an identical data format. Furthermore, the device combines the plurality of sets of rows having an identical data format into a transformed data set.

Abstract: Aspects include a method that includes receiving a structured query language (SQL) database command at a processing device configured to manage a table space of a relational database, the database command including a merge statement. The method also includes identifying a group of target records from the database command, and associating each target record with one of a plurality of portions of the table space, and in response to determining that the group of target records is associated with more than one portion of the table space, sorting the target records into a plurality of sub-groups, each sub-group associated with a different portion of the table space. The method further includes generating a sub-task associated with each sub-group, the sub-task prescribing aspects of performing the merge operation using an associated sub-group respective sub-group, and performing all of the sub-tasks on the memory space in parallel.

Abstract: A search engine identifies external data records that describe similar entities and may each conform to a different data format or source schema. The engine derives mappings capable of translating data values between differently formatted attributes of two source schemas and uses these mappings to identify degrees of similarity between attributes and schemas. When the search engine receives a search request, the engine translates submitted search criteria into values of a first schema's attributes and then uses the mappings to map those values onto selected attributes of other schemas. The search engine then uses each schema's selected attributes to select external data records formatted in that schema. Each selected record is assigned a match score that is weighted by the similarity of the record schema's selected attributes to the search criteria. Records are then retrieved in order of decreasing match score.

Abstract: Aspects include a method that includes receiving a structured query language (SQL) database command at a processing device configured to manage a table space of a relational database, the database command including a merge statement. The method also includes identifying a group of target records from the database command, and associating each target record with one of a plurality of portions of the table space, and in response to determining that the group of target records is associated with more than one portion of the table space, sorting the target records into a plurality of sub-groups, each sub-group associated with a different portion of the table space. The method further includes generating a sub-task associated with each sub-group, the sub-task prescribing aspects of performing the merge operation using an associated sub-group respective sub-group, and performing all of the sub-tasks on the memory space in parallel.

Abstract: A device and a method are provided to automatically generate test case for embedded software. This invention is in software test field, including symbolic execution kernel module, path selection module, solver, debugger, concrete execution kernel module and debugger agent module. The tested software and test cases are uploaded from the host system to the embedded system through debugger and debugger agent. The concrete execution kernel module starts the tested software. The symbolic execution kernel module captures the run-time information of the tested software through the debugger. When the tested software operates on the symbol source, the symbolic execution kernel module marks the symbol source, tracks the symbol propagation, generates path condition and sends the path condition to path selection module. This invention can automatically generate test cases for embedded software, which doesn't need the source code of the tested software and can be conveniently used for commercial software.

Abstract: A method and system for interfacing between a caller application and a service module. A request for performing a transaction from the caller application is received. The request includes at least one caller application attribute describing the request. A service module data structure is built pursuant to the received request. The service module data structure includes a generic service document and at least one service module attribute. Each service module attribute is stored in a relational table of the service module data structure. The request is serviced within the service module data structure, which results in instantiating the generic service document. The generic service document is returned to the caller application.

Abstract: A message sending system, the system including: a member terminal of a member; a main server connected to the member terminal through a wired or wireless network; and an emergency contact terminal having a predetermined relationship with the member terminal, information of the emergency contact terminal is registered in the main server, wherein the member terminal is configured to upload, to the main server, a list of mourners comprising a list of persons to be invited as mourners and a message to be delivered to each of the mourners, and wherein the main server is configured to: determine, through the member terminal and the emergency contact terminal, the time of death of the member having the member terminal, and in response to determining the time of death of the member, send a notice of the member's death to mourner terminals registered in the list of mourners.

Abstract: A pedestrian detection system including a camera device and a cloud server. The cloud server includes an entrance node and a compute node unit including a plurality of compute nodes. The entrance node is configured to receive image data from the camera device, to divide the image data into a plurality of subordinate task areas, and to extract a percent of the subordinate task areas. The compute node unit is configured to receive the percent of the subordinate task areas from the entrance node, and processes simultaneously the percent of the subordinate task areas. Therefore, an efficiency of processing image of pedestrian can be improved.

Abstract: Embodiments of the present invention address deficiencies of the art in respect to method and process management, and provide a data processing system, method, and computer program product for process contributions in a method architecture. In accordance with an embodiment of the present invention, a data processing system configured for process contributions can include a method management tool configured for composing a process utilizing the method architecture. The system further can include one or more source method elements combinable to define an activity for the process. Finally, the system can include one or more process contributions defining variations in the activities of the process.

Abstract: A method of facilitating the entry of text into a new message generated by a messaging application of a handheld electronic device is provided that includes receiving a command to reply to or forward a received message that is included within a message thread, scanning the message thread, identifying a plurality of words in the message thread, generating one or more word lists that include the identified words, and displaying the one or more word lists. The method further includes receiving a selection of a selected one of the words from the one or more word lists, and entering the selected one of the words into the new message. Also provided is an improved handheld electronic device that implements the described method.

Abstract: Techniques are described for creating a first data abstraction model for a first database. Embodiments analyze the first database to determine a first set of structural characteristics, and analyze a second database to determine a second set of structural characteristics. The analyzed second database is associated with a second data abstraction model. The first set of structural characteristics is compared with the second set of structural characteristics to identify one or more similarities between the two sets of structural characteristics. Embodiments then create the first data abstraction model for the first database, based on the identified similarities and the second data abstraction model.

Abstract: Systems and methods for photograph mapping are disclosed herein. In one embodiment a first digital image and at least one user-generated datum is received from at least one user. The first image is geographically organized according to the at least one datum. The first image is associated with at least one location and at least one direction. The first image is provided from a first person perspective to a user in response to a request.

Abstract: A system is provided for managing data sets in a cloud processing and/or federated environment. In an embodiment, the system described herein may be used in connection with cloud processing of big data sets. The term “big data,” as used herein, may be generally defined to describe data sets so large and complex that they become difficult to work with using on-hand database management tools. The system described herein enables the persistent storage of semantic technology statements for big data sets for processing in a cloud processing and/or federated environment.

Abstract: In one embodiment, a method determines a job to dispatch for transcoding a source file. The source file is stored on a node in a distributed file system in which a plurality of nodes are storing a plurality of source files. The method then determines location information for the source file in the distributed file system where the location information identifies the node in which the source file is stored. Also, a node in the plurality of nodes is determined in which to dispatch the source file based on the location information for the source file. Then, the job is dispatched to the determined node such that transcoding of the source file is performed on the determined node.

Abstract: Systems and methods are provided for providing schema mapping and data generation. According to one exemplary method, a first mapping formula is obtained for transforming values associated with a source attribute into at least one value associated with a conceptual attribute. The source attribute values comprise a plurality of duplicate values, and the first mapping formula performing a common operation on the duplicate values. The method includes generating mapping data for mapping the conceptual attribute to a target attribute of a target model, and based on the mapping data and the first mapping formula, generating a second mapping formula for transforming the conceptual attribute value into at least one value associated with the target attribute.

Abstract: Legacy computing system archive files may have unique data structures designed for long-term data retention purposes and not particularly suited for data mining, querying, and other data collection functions needed to prepare certain types of reports. A report specific template may be created to facilitate the extract of specific legacy system data necessary to complete report that is stored in legacy system archive files. Once the data objects containing information relevant to the report have been identified, a hierarchical list of tables and fields of each table associated with each identified data object may be obtained from an external data source for further selection. Copies of the selected table and/or field data may then be extracted, compared, and merged with other data, such as other field and table data, obtained from other sources.

Abstract: Disclosed are method and apparatus for identifying members of a social network who have a high likelihood of providing a useful response to a query. A query engine examines the personal pages of a set of members and automatically gleans semantic information relevant to the query. From the automatically-gleaned semantic information, a score indicative of the likelihood that the member may provide a useful response is calculated.

Abstract: Encoding and/or decoding of messages. On the encoding end, a composite encoder encodes message from an internal format that is used by internal system components into an external format. However, the composite encoder may encode the outgoing messages into different external formats on a per-message basis. For incoming message, a composite decoder decodes incoming messages from any one of a plurality of external formats into the internal format also on a per-message basis. A per-message report mechanism permits internal system components and the encoding/decoding components to communicate information regarding the encoding or decoding on a per message basis. XML messages can be converted automatically into Binary messages.

Abstract: Techniques are described for creating a first data abstraction model for a first database. Embodiments analyze the first database to determine a first set of structural characteristics, and analyze a second database to determine a second set of structural characteristics. The analyzed second database is associated with a second data abstraction model. The first set of structural characteristics is compared with the second set of structural characteristics to identify one or more similarities between the two sets of structural characteristics. Embodiments then create the first data abstraction model for the first database, based on the identified similarities and the second data abstraction model.

Abstract: A data model may be designed based views of a pre-existing data model. Metadata associated with the pre-existing data model may be presented to the user to support the user's design effort. The metadata may inform and otherwise guide the user's design choices to reduce the likelihood of inconsistencies between the new data model and the pre-existing data model. The metadata may also be used to identify and warn the user of potential inconsistencies between the new data model and the pre-existing data model.

Abstract: A method and system for generating schema to Java descriptors. In one embodiment, a second object having customized schema to Java mapping information is created. The second object is placed within a first object. Via the first object, schema to Java generation is initiated.

Abstract: In one embodiment the present invention includes a computer-implemented method of improving performance of schema matching. The method includes storing a matching topology that maps a first data structure to a second data structure. The method further includes identifying a first portion of the matching topology that is a candidate for improvement as determined according to an incidence graph. The method further includes rewriting the first portion of the matching topology with a second portion according to the incidence graph.

Abstract: A query processing method intersects two or more unsorted lists based on a conjunction of predicates. Each list comprises a union of multiple sorted segments. The method performs lazy segment merging and an adaptive n-ary intersecting process. The lazy segment merging comprises starting with each list being a union of completely unmerged segments, such that lookups into a given list involve separate lookups into each segment of the given list. The method intersects the lists according to the predicates while performing the lazy segment merging, such that the lazy segment merging reads in only those portions of each segment that are needed for the intersecting. As the intersecting proceeds and the lookups are performed, the intersecting selectively merges the segments together, based on a cost-benefit analysis of the cost of merging compared to the benefit produced by reducing a number of lookups.

Abstract: Methods, systems, and computer-readable storage media for receiving a plurality of source hierarchical schemas, processing the source hierarchical schemas to generate a merged graph, which is comprising a plurality of merged nodes, each merged node being provided based on one or more nodes from at least two of the source hierarchical schemas, determining and resolving using conflicts in the merged graph to generate a conflict-free merged graph, wherein resolving comprises splitting one or more merged nodes into respective sub-sets of merged nodes, defining and processing a constraints satisfaction problem (CSP) to generate a plurality of mediated hierarchical schemas (MHSs), each MHS being a solution to the CSP, identifying an MHS of the plurality of MHSs as an optimum MHS, wherein a canonical hierarchical schema (CHS) is provided as the optimum MHS, and storing the CHS as a computer-readable document in a computer-readable memory.

Abstract: Resolving conflicting graph mutations in a distributed computing system. Graph data for at least a partition of a graph is stored in a worker system of a distributed computing system. The graph represents relationships among a set of tangible items that model a real-world condition having an associated problem. A plurality of conflicting mutation requests are received to mutate the graph. A conflict between the mutation requests is resolved with a conflict resolution function that lacks direct access to the graph data. The graph data is updated responsive to a result generated by resolving the conflict using the conflict resolution function.

Abstract: Dynamic blocking determines which pairs of records in a data set should be examined as potential duplicates. Records are grouped together into blocks by shared properties that are indicators of duplication. Blocks that are too large to be efficiently processed are further subdivided by other properties chosen in a data-driven way. We demonstrate the viability of this algorithm for large data sets. We have scaled this system up to work on billions of records on an 80 node Hadoop cluster.

Abstract: In some example embodiments, a system and method are illustrated to link a plurality of scenes from at least one multimedia file. The system and method include loading a plurality of multimedia files including a first and second multimedia file. The first multimedia file includes a first scene, and the second multimedia file includes a second scene. The system and method include linking the first scene with the second scene. The linking includes connecting a first durational data for the first scene with a second durational data for the second scene. The first durational data identifies a first start location and a first end location for the first scene. The second durational data identifies a second start location and a second end location for the second scene. The system and method further include generating a linked durational data file as a result of the linking.

Abstract: Techniques are disclosed for generating statistical views in a database system. In one embodiment, a request is received to execute a database workload. One or more constraints pertaining to executing the database workload is retrieved. The database workload is evaluated to generate multiple statistical view candidates. The statistical view candidates are refined based on the one or more constraints. One or more statistical views are then generated based on the refined statistical view candidates.

Abstract: A system in which an information processing apparatus, relay server apparatus, and management server apparatus are connected via the network, wherein the said information processing apparatus possesses a unit to send a request to merge files managed by the said management server to the said relay server apparatus, and the said relay server apparatus possesses a unit to receive the said merge request from the said information processing apparatus, to determine whether the merged file created based on the said merge request received using the said receiving unit is stored in the memory area of the said relay server apparatus, and to retrieve the said merged file from the said memory area if it is determined that the merged file is stored in the said memory area using the said determination unit solves the problem.

Abstract: An information terminal which has a storage section which stores a guide and a specific character string, and a control section. The guide indicates a possibility that an acquisition act of a different electronic document associated with an electronic document displayed on a display section is permitted only if an authentication condition stipulated at the site having the different electronic document is met The control section searches the electronic document displayed on the display section for the specific character string, and counts character strings matching the specific character string when a condition check operation is performed through the operation section. The control section further controls the display section to display the guide if an acquired count is equal to or greater than a predetermined number.

Abstract: Embodiments of the disclosure relate to file merging based on file metadata. An exemplary embodiment concerns a system for merging two files that may comprise first and second data structures respectively associated with first and first files. Each data structure may include a starting address and an ending address of the associated file, and file attributes. The system may include a logic component for respectively referencing the starting address and ending address of a second file in the starting address and ending address of a first file, and a logic component for linking the last record in the first file with the first record of in the second file.

Abstract: A Double Blinded Privacy-Safe Distributed Data Mining Protocol is disclosed, among an aggregator, a data consumer entity having privacy-sensitive information, and data source entities having privacy-sensitive information. The aggregator does not have access to the privacy-sensitive information at either the data consumer entity or the data source entities. The aggregator formulates a query without using privacy-sensitive information, and sends the query to the data consumer entity. The data consumer entity generates a list of specific instances that meet the conditions of the query and sends the list, encrypted, to the data source entities either directly or through the aggregator. The data source entities match the list against transactional data, de-identify the matched results, and send them to the aggregator. The aggregator combines results from data source entities and sends the combined result to the data consumer entity.

Abstract: In various embodiments, a system receives a multidimensional schema entity of a first type and converts the multidimensional schema entity to a second type. The system receives user input and converts the multidimensional schema entity to the second type based on the input received from the user. In various embodiments, the system creates multidimensional schema entities automatically. In various embodiments, a method for converting multidimensional schema entities from one or more types to one or more other types is described. In various embodiments, a multidimensional schema entity is created automatically based on input from two other multidimensional schema entities. In various embodiments, two multidimensional schema entities are merged in one multidimensional schema entity. In various embodiments, multidimensional schema entities are used to generate a report. Queries extract data from the multidimensional schema entities and load it in the report. The report is presented on a graphical user interface.

Abstract: A method of consolidating records from databases is disclosed. The method comprises initially making a first identification of a candidate match between records using a first search technique. The method further comprises making a second identification of a candidate match between records using a second search technique, and displacing the first identification of a candidate match between records based on a probability that the second identification is correct.

Abstract: Techniques are disclosed for generating statistical views in a database system. In one embodiment, a request is received to execute a database workload. One or more constraints pertaining to executing the database workload is retrieved. The database workload is evaluated to generate multiple statistical view candidates. The statistical view candidates are refined based on the one or more constraints. One or more statistical views are then generated based on the refined statistical view candidates.

Abstract: Legacy computing system archive files may have unique data structures designed for long-term data retention purposes and not particularly suited for data mining, querying, and other data collection functions needed to prepare certain types of reports. A report specific template may be created to facilitate the extract of specific legacy system data necessary to complete report that is stored in legacy system archive files. Once the data objects containing information relevant to the report have been identified, a hierarchical list of tables and fields of each table associated with each identified data object may be obtained from an external data source for further selection. Copies of the selected table and/or field data may then be extracted, compared, and merged with other data, such as other field and table data, obtained from other sources.

Abstract: In various embodiments, approaches for identifying missing search phrases are described. Search terms are behaviorally associated with an item in the electronic repository. If a search of the electronic repository fails to surface the item, the at least one search term can be identified as a missing search phrase for the item.

Abstract: An approach is provided for managing projection and injection operations on information spaces with respect to their information content. An information space projection module receives a query to project a first information space from a second information space. In response to the query, the module extracts a subset of information content from the second information space by using a partitioning function. The module also extracts a subset of rules from the second information space by using the partitioning function. The module then creates the first information space using the extracted subset of information content, and the extracted subset of rules while maintains a link between the first and the second information spaces. An information space injection module enables further injection of the first information space back into the second information space.

Abstract: In accordance with particular embodiments, a method includes receiving data from disparate data sources comprising data associated with different people. At least one of the data sources comprises data associated with at least one common person. The method also includes extracting at least one record from each of the disparate data sources. Each extracted record is associated with a corresponding person and comprises at least one feature associated with the corresponding person. The method additionally includes distributing the records to computers configured to identify matches between the records based on at least one matching feature associated with two or more records. The method further includes merging the at least one feature from each of the records of each corresponding match into merged records, each merged record corresponding to a different match. The method also includes generating at least one fused view associated with the at least one common person and comprising a merged record.

Abstract: Disclosed are method and apparatus for identifying members of a social network who have a high likelihood of providing a useful response to a query. A query engine examines the personal pages of a set of members and automatically gleans semantic information relevant to the query. From the automatically-gleaned semantic information, a score indicative of the likelihood that the member may provide a useful response is calculated.

Abstract: A system and method which may allow a Web application to manage cookies and prevent important data in cookies from being arbitrarily deleted. Cookie data may be separated into a number of tiers according to their importance. When a request to write new data to a cookie is received, the tier to which the new data belongs may be determined and compared to the tier(s) of existing data in the cookie, and existing data may be deleted from the cookie to free room for the new data only when the existing data is not more important than the new data.

Abstract: A computer-implemented system may include one or more storage devices storing first metadata defining an instance of a first metaobject, second metadata defining an instance of a solution supplement metaobject, wherein the second metadata comprises annotations to the first metadata, and wherein the first metaobject and the solution supplement metaobject are instances of a meta-metaobject. Also stored is executable program code of a service to determine merged metadata based on the first metadata and the annotations to the first metadata. A processor is included to execute the executable program code.

Abstract: An apparatus and method for merging the data of a first container and a second container into a target container that includes at least one block having a plurality of extents to store data objects. The apparatus may include a storage sever coupled to a plurality of storage devices. The storage server is configured to merge multiple data objects of the plurality of containers that store a data object per each container into a target container that stores multiple data objects within the target container.

Abstract: Embodiments disclosed herein provide an implementation defined segments (IDS) subsystem which allows new data segments to be added to an identity hub after deployment. A set of metadata tables are utilized to describe IDS, each of which is a data structure encapsulating a single row from a master data record residing in the identity hub. Once a segment (an object) is described, the identity hub can use the information to define persistent storage for the object in the database for any relational database management system, create internal structures to hold the data and process business rules and demographic comparisons against the data object, describe the data object to remote clients, and allow the clients to query the identity hub at runtime about what data objects exist, what fields and data types they contain, and additionally how they might be displayed or formatted on various clients.

Abstract: The present invention provides a reproducing apparatus for reproducing a plurality of pieces of content data stored in a content data storage unit according to a play list defining reproduction order of the content data. The apparatus includes a play list storage unit, a request information generating unit, and a play list reconstructing unit. The play list storage unit stores the play list having a play list generating rule and identifying information for identifying the content data matching the play list generating rule. The request information generating unit generates reconstruction requesting information for requesting reconstruction of the play list. The play list reconstructing unit reconstructs at least a part of the play list stored in the play list storage unit into a play list having the play list generating rule corresponding to an apparatus use condition in response to the reconstruction requesting information generated by the request information generating unit.

Abstract: Automated form filling for mobile devices that provides reliable and accurate information to the forms, is easy to use, and is scaleable. A server-side approach is utilized, in which online applications for a mobile device are invoked on a server through a server-side proxy/cache. The proxy scans the content that is generated by the application for transmission to the mobile device to find forms that may be embedded in the content. When a form is encountered, fields of the form are filled with stored information based on automatically generated mapping information.

Abstract: A method for constructing a processing request so that an information processing application satisfying the processing request can be assembled, includes: inputting a processing request, wherein the processing request includes a goal that is represented by a graph pattern that semantically describes a desired processing outcome; and assembling a processing graph that includes at least one component that satisfies the desired processing outcome, wherein the processing graph is assembled by associating the graph pattern that semantically describes the desired processing outcome with a graph pattern that semantically describes an applicability condition of the at least one component or a graph pattern that semantically describes an inclusion effect of the at least one component.