Abstract: A method for validating data changes made to a database is disclosed. The changes are made in the context of a transaction, and validation is performed using a rules database storing a plurality of rules. The method includes identifying a set of data entities affected by one or more data changes made in the context of the transaction. In response to an instruction to commit the transaction, data entities in the set of affected data entities are validated using rules from the rules database. The transaction is committed in dependence on the outcome of the validation.

Abstract: Techniques are described for dynamic translation of data access statements. A first set of statements is sent by an executing process to access a first data storage that supports a first query language. The first set of statements is intercepted and translated into a second set of statements configured to access a second data storage that supports a second query language. The second set of statements may be executed on the second data storage, and the result data set may be sent to the executing process that generated the first set of statements. The first and second storages may employ different data storage formats, such as relational and non-relational storage formats. The behavior of the executing process may be monitored to determine the feasibility of modifying the source code of the process to employ a different query language.

Abstract: The present disclosure provides an approach in which a domain corpus subset generator correlates documents from a document corpus to domain discernible attributes associated with domain corpus subsets. The domain corpus subset generator analyzes correlation results from the correlation and stores the documents into domain corpus subsets accordingly. In turn, a question-answer system utilizes documents included in a specific domain corpus subset to provide relevant and accurate answers to an input question.

Abstract: A method for measuring a change between two values for one or more data elements over time uses a “delta” value to indicate the significance of the change. The delta value is the difference between the two values, added to the ratio between those values multiplied by an asymptote. Experimental data sets with known values may be used to create data sets to which selection criteria are applied in order to calculate an optimal asymptote for the application of the delta values. This optimal asymptote may then be used for calculating delta values across an entire population for predictive purposes.

Abstract: Systems and methods disclosed herein provide intelligent filtering of system log messages having low utility value. In providing the filtering, the systems and methods determine the utility value of a system log message and delete the message from the system log if the message is determined to be of low utility value. As such, embodiments herein provide an system log filter, which reduces the amount of data stored in the system log based on the utility value of the message.

Abstract: Systems and methods are provided that allow a user to send a search query and a hint associated with the search query to a search system. Suggested queries are associated with hints prior to the suggested queries being provided to a user interface that allows a user to enter search information into a search field. After receiving input into the search field, the input is sent to a search system and suggested queries with their associated hints are returned. In response to selecting a suggested query, the suggested query and the associated hint are returned to the search service. The search service then provides search results based on the suggested query and the associated hint.

Abstract: A method for constructing a change database reduces the footprint of a standard longitudinal data database and increases the manageability of the data for many applications. The change database concatenates representations of a time period for the change, the direction of the change, and the value of the change into a single string. The change database is constructed by creating foundation files from formatted historical data, constructing change files from these foundation files, and iteratively creating master files that encapsulate the change data.

Abstract: A system and method for providing a file management service for collaboration is disclosed. The method for providing a file management service for collaboration, which may be implemented on a system, includes, for example, providing a client through which a user accesses a file through the file management service, further providing, by the file management service via the client, the user with collaboration capability of the file with other users, and/or controlling editing of the file by other users based on whether it is specified as being shared by the user. In one embodiment, the file management service further enables integration of a third-party service with the file management service and the third-party service is accessible for use with the file via the file management service.

Abstract: A network-based service may be provided for facilitating queries for a number of items, such as travel services. A user may submit a query including criteria for determining one or more relevant items. Based on the submitted query, the network-based service may present the user with information regarding the actions of other similar users of the network-based service, such as searches performed by the other users. Based on this information, the user may elect to supplement the current query to conform to the actions for other users. In some embodiments, actions by other users may be based at least in part on a category of the querying user. By presenting actions of similar users, a current user may be enabled to select the most relevant query terms for identifying a desired item.

Abstract: Query formation and modification techniques are described. In one or more embodiments, a query is received that is formed in a text field as a phrase in a human-readable language that includes a visual indicator that represents a missing term that is a subject of the query. Based on the query, a defined database is searched, and one or more modified versions of the phrase are presented that replace the visual indicator with a respective result of the searching. In addition, one or more options are presented that are selectable to automatically complete the phrase as a sentence in the human-readable language.

Abstract: A system and method for searching indexed content data using multiple proximity keyword searches to quickly and efficiently locate subsets of data that are either relevant or irrelevant to an issue of interest to a user.

Abstract: A processor determines a predicted quantity of updates to be made to a set of data using a history of updates. The set of data is associated with a category of replication that indicates a first frequency of data replication to be applied to the set of data. The processor determines whether the first predicted quantity of updates meets a first threshold for a frequency of replication of the set of data. The first threshold indicates an allowed number of updates and meeting the first threshold indicates that an adjustment to the frequency of replication of the set of data is to be initiated. The processor responds to a determination that the first threshold has been met by associating the set of data with a category of replication that has a second frequency of data replication to be applied to the set of data.

Abstract: A categorization service is described that automatically categorizes items of interest to a user. The user may possess an item that they wish to offer for sale using the network-based service. The user may prepare an electronic description of the item and submit this item information to a categorization service for categorization. Upon receipt, the categorization service may compare the item information to item category descriptions representing item categories of the network-based server in order to determine category recommendations for the item of interest. The recommended categories are returned to the user to enable the user to assign a category, selected from the category recommendations, to the item of interest.

Abstract: A hybrid-sharded index includes document-sharded posting lists and term-sharded posting lists. Implementations include systems and methods for updating a hybrid-sharded index. For example, a method may include receiving updates to the hybrid-sharded index and generating, at a first leaf node, replacement posting lists and change information for a respective second leaf node. The method may also include dividing the replacement posting lists into portions, a portion having associated change information and being associated with a respective one of the second leaf nodes and sending the portions to respective leaf nodes. At a particular leaf node of the second leaf nodes, the method includes merging a received portion into an updated posing list portion, swapping the updated posting list portion into memory. During the swap, the change information and the updated posting list portion are used to respond to a query with an older version of the hybrid-sharded index.

Abstract: Entity resolution in a database comprises receiving imported data comprising imported data entities each having properties each having values; receiving first user input that selects the imported data entities for resolution to existing data entities in a database; receiving second user input that specifies matching criteria for matching the imported data entities to the existing data entities, wherein each of the matching criteria comprises a matching technique; matching the imported data entities to the existing data entities using the matching criteria, resulting in creating and storing matched entity information, wherein the matched entity information is organized in matched entity data sets associated with subsets of the matching criteria that were matched; consolidating the imported data entities into the existing data entities; storing the first user input and second user input as a named criteria set for use in subsequent entity resolution operations.

Abstract: A method for providing content relevant images for an input question to a deep question answering system is disclosed. The method can include formulating, in response to receiving the input question, an answer to the input question. The method can also include identifying, based on the answer to the input question, a set of candidate images from a corpus of information. The set of candidate images may relate to the answer. The method can also include determining, based on a confidence value of the set of candidate images and a designated display specification, a first subset of the set of candidate images to provide in a display area. The method can also include selecting the first subset to be provided in the display area.

Abstract: One or more processors apply rules to a first query to generate a modified query such that processing of the modified query is optimized for a first set of processing conditions. One or more processors measure a degree of latency experienced during processing of the modified query under a second set of processing conditions. One or more processors generate other rules to be applied to queries based, at least in part, on the measured degree of latency.

Abstract: Embodiments include methods, systems and computer program products for offloading multiple processing operations to an accelerator. Aspects include receiving a database query from an application, performing an analysis on the query, and identifying a plurality of available accelerators. Aspects further include retrieving cost information for one or more templates available on each of the plurality of available accelerators, determining a query execution plan based on the cost information and the analysis on the query, and offloading one or more query operations to at least one of the plurality of accelerators based on the query execution plan.

Abstract: A method, apparatus, and computer program product are provided for routing files within a document management system. The document management system may comprise or may otherwise communicate with an Optical Character Recognition (OCR) system and Full Text Search (FTS) indexing system. Files may be fed to a file routing apparatus with different priorities based on an originating source. The files may be routed for OCR processing and/or FTS indexing based on file and/or document types, licensing, and/or the like. Images and corresponding machine-readable text may be archived. Text may be FTS indexed so that the associated files may be identified and accessed by various client systems.

Abstract: According to an aspect, a computer implemented method for managing a database in real-time is provided. The method includes scanning for predetermined data in a predetermined workload stream during real-time operation of the database, and determining a workload priority for located predetermined data according to one or more workload rules. A storage priority of located predetermined data is determined according to a physical location of the located predetermined data. The located predetermined data are moved during the real-time operation of the database to a different physical location if the storage priority and the workload priority do not match.