Abstract: A multi-class classification system performs multi-class classification using an adapted, multi-level binary classification model in conjunction with pre-classification and post-classification processing. Original, multi-class training data is transformed into coarse-level adapted training data and fine-level adapted training data. In the coarse-level adapted training data, objects from the original training set are classified into either a majority classification set (representing the largest original classification set) or a non-majority classification set representing a combination of the remaining classification sets from the original training data. In the fine-level adapted training data, objects from the non-majority classification set retain their original classifications from the original training data.

Abstract: Content packages (including, e.g., visualizations, models) are replicated from one landscape to another that is restricted. The restricted landscape may be governed by regulations relating to privacy or other concerns. For package replication, a poll-pull mechanism is applied. Specifically, the first landscape creates a replicate package job targeting to the restricted landscape. The restricted landscape periodically polls and checks for any replicate package job for targeted for it. If any targeted replicate package jobs are discovered by polling, the restricted landscape will pull the job details to create a new replicate package job in its own (restricted) landscape. Under this approach, even though the replicate package job was initiated in the first landscape, it will be executed in the restricted landscape, thereby conforming to any governing regulations.

Abstract: A networked system receives, from a device of a user, a selection of an item from a list for monitoring. The network system also receives, from the device of the user, a condition that triggers the reporting of news for the selected item. The network system monitors for news for the selected item. A determination is made as to whether the condition that triggers the reporting of the news is satisfied. In response to determining that the condition that triggers the reporting of the news is satisfied, the networked system transmits a notification to the user indicating the news for the selected item.

Abstract: Methods and systems for providing computer-assisted guided review of unstructured data to generate a structured data output based on customizable template rules. In embodiments, an unstructured file is received, and a predefined template is selected. The predefined template includes a plurality of fields, each field corresponding to a field of the structured report. The predefined template also defines extraction rules for each field of the predefined template, and the extraction rules define parameters for identifying unstructured data relevant to the associated field. The extraction rules are applied to the unstructured file to identify data relevant to the field associated with the corresponding extraction rule, and the data identified as relevant is confirmed. Confirming the relevant data includes determining to refine the relevant data based on a condition, and modifying the extraction rule associated with the field to refine the relevant data.

Abstract: A search assistant engine is described that integrates with a data intake and query system and provides an intuitive user interface to assist a user in searching and evaluating indexed event data. Additionally, the search assistant engine provides logic to intelligently provide data to the user through the user interface such as determining fields of events likely to be of interest based on determining a mutual information score for each field and determining groups of related fields based on determining a mutual information score for each field grouping. Some implementations utilize machine learning techniques in certain analyses such as when clustering events and determining an event templates for each cluster. Additionally, the search assistant engine may import terms or characters from user interaction into predetermined search query templates to generate tailored search query for the user.

Abstract: A method includes performing a search of an inverted index structure indexing values of a column to generate an in-range indexed value set by identifying all indexed values of the inverted index structure falling within a range corresponding to a range-based filter. A set of characteristics of the in-range indexed value set are identified based on performing the search of an inverted index structure. When the set of characteristics compare favorably to the set of index-usage requirements, output is generated based on performing a plurality of searches to the inverted index structure based on the in-range indexed value set. When the set of characteristics compare unfavorably to the set of index-usage requirements, the output is generated without performing any searches to the inverted index structure.

Abstract: A document search system that enables efficient document search regardless of the ability of a user is achieved. Document search is performed using a document search system in which database document data is stored. After first document data and second document data are input to the document search system, the document search system extracts a plurality of terms from the first document data. The extraction of the terms is performed using morphological analysis, for example. Next, the extracted terms are weighted on the basis of the second document data. For example, texts included in a document represented by the second document data are classified into first and second texts. Among the terms extracted from the first document data, the weight of the term included in the first text is set larger than the weights of the other terms. The classification of the texts can be performed in accordance with a rule basis or using machine learning.

Abstract: An identification of a specification that identifies one or more data sources is received. The one or more data sources are respectively associated with one or more database queries. Each of the one or more database queries is associated with a different embedded screen. An end-user application that is configured to generate selectable user interface elements for the one or more different embedded screens is generated. Generating the end-user application is based on the specification. In response to selection of a particular selectable user interface element of the selectable user interface elements, an embedded screen associated with the particular selectable user interface element is identified, and a user interface including the identified embedded screen is provided.

Abstract: Systems and techniques for patent and prior art analysis are described herein. An issued independent claim of the identified patent is electronically retrieved. A pre-allowance published independent claim of the identified patent is electronically retrieved. The issued independent claim is automatically compared to the pre-allowance published independent claim using at least one unique keyword. A search query string is formulated comprising the at least one unique keyword. A search query is executed to a prior art database comprising a set of prior art documents using the search query string to identify a result prior art document set. An owner of each prior art document of the result prior art document set is identified. The owner is automatically ranked based on a number of keywords from the search query string owned by the owner. A user interface is output to a user comprising a visualization of the rank of the owner.

Abstract: Replicating multiple storage systems utilizing coordinated snapshots, including identifying a replica dataset stored across two or more target storage systems, wherein the replica dataset is a replication target for a source dataset stored across two or more source storage systems; identifying two or more local replicated checkpoints that are replicated from the two or more source storage systems to the two or more target storage systems, wherein two or more local source checkpoints for the two or more local replicated checkpoints are associated with a coordinated source checkpoint for the source dataset; and determining, based on the two or more local replicated checkpoints, a coordinated target checkpoint for the replica dataset.

Abstract: The disclosed methods and systems provide computer-assisted guided review of unstructured data to generate a structured data output based on customizable template rules. In embodiments, an unstructured file is received, and a predefined template is selected. The predefined template includes a plurality of fields, each field corresponding to a field of the structured report. The predefined template also defines extraction rules for each field of the predefined template, and the extraction rules define parameters for identifying unstructured data relevant to the associated field. The extraction rules are applied to the unstructured file to identify data relevant to the field associated with the corresponding extraction rule, and the data identified as relevant is confirmed. Confirming the relevant data includes determining to refine the relevant data based on a condition, and modifying the extraction rule associated with the field to refine the relevant data.

Abstract: A method comprises receiving a request to combine data from a plurality of data collections based on one or more constraints, and analyzing the request using one or more machine learning algorithms to determine an order by which respective ones of the plurality of data collections are to be searched. The respective ones of the plurality of data collections are searched according to the determined order, and an output of combined data from the searching is generated in response to the request.

Abstract: In one technique, a definition of a materialized view is identified. Based on the definition, multiple candidate partitioning schemes are identified. A query is generated that indicates one or more of the candidate partitioning schemes. The query is then executed, where executing the query results in one or more partition counts, each corresponding to a different candidate partitioning scheme of the one or more candidate partitioning schemes. Based on the one or more partition counts, a candidate partitioning scheme is selected from among the plurality of candidate partitioning schemes. The materialized view is automatically partitioned based on the candidate partitioning scheme.

Abstract: An index is provided of content at timestamps of frames in a video. A search request is received with search content. A determination is made of matching content in the index, associated with a timestamp, matching the search content. A computer addressable link to the timestamp in the index associated with the matching content is returned to the search request. Selection of the computer addressable link causes the video to play from the timestamp.

Abstract: Methods, systems, and computer-readable storage media for receiving a query from a digital assistant of an enterprise system, retrieving data that is responsive to the query from a data management system, inputting a first few-shot prompt to a LLM, and determining, in response to the first few-shot prompt, that a graphical representation of the data is to be generated, and in response: inputting a second few-shot prompt and a third few-shot prompt to the LLM, receiving code from the LLM responsive to the third few-shot prompt, and executing the code to render the graphical representation with the digital assistant.

Abstract: The retrieval of files can be facilitated in real-time in the general context of backup and recovery. A processor can receive a request to retrieve a particular volume of data from an object storage database. A virtual volume can be presented to an emulator for creation of a virtual machine representing the particular volume of data. A request can be received to access a particular file from the particular volume of data at a specific point in time. In response to the request to access the file, the data stream can be paused, the particular file can be fetched, and the particular file can be transmitted to the emulator.

Abstract: A computer-implemented method that includes: receiving, at a middleware server, an input stream of query requests each requesting an electronic information service provided by an external database of an external trusted party, wherein the query requests are issued by internal enterprise applications; determining, by the middleware server, a first service validity period for a query request from the input stream, wherein the first service validity period is pre-determined; determining, by the middleware server, a gap between a current date for the query request and a corresponding past date when the query request was last successfully answered by the external database; and in response to determining that the gap is smaller than the first service validity period, fetching a corresponding query result from a staging table on the middleware server, wherein the corresponding query result was received when the query request was last successfully answered by the external database.

Abstract: Example solutions for executing a query in a declarative graph query language include receiving the query for data in a database and determining if one or both of i) a pattern in the query, and ii) the data in the database render the query suitable for being executed, at least in part, in parallel. If either condition indicates that the query is suitable for being executed, at least in part, in parallel, one or more fork operations and join operations are injected into a query plan, and the query is executed according to the query plan. Some examples further determine whether executing the query in parallel is computing resource-efficient, and only executes computing resource-efficient queries in parallel.

Abstract: A system, method, and computer-readable medium are disclosed that may generate staging data independently from the execution of control scripts, which process the staging data, improving the quality, efficiency, and speed of data processing systems. The staging data can be independently loaded, validated, and utilized across multiple control scripts, reducing redundancy in the loading of data and the overhead of executing separate data processing for each control script. The control scripts can be automatically validated, such as by verifying expected output data ranges. Additionally, the complexity of the control scripts can be reduced as the loading of data is not performed by the control scripts. The controls scripts can generate a variety of output data, such as an indication of impacted accounts, and provide notifications based on the output data. A variety of machine learning classifiers may automatically generate the staging data and validate the staging data and/or output data.

Abstract: As a part of an embedded system environment, an overlay filesystem package manager obtains package files for a filesystem operation. the package files are stored in a temporary memory location. a composite layer presenting upper and lower filesystems of an overlay filesystem, is accessed to compare the package files with an associated files in the composite layer. When a difference is detected between a package file and an associated file, the package file is copied or moved to the upper filesystem. If the package file is not different from the associated file in the composite layer, it is not copied and deleted from the temporary memory location.