Abstract: A hybrid query execution plan is generated for executing queries on database systems implementing a columnar cache. A hybrid query execution plan combines a query execution plan for querying and retrieving data from a columnar cache and a base table. A columnar cache stores cached data in column-major format, which is logically represented by the database management system in row-major format. A database management system as described herein can scan valid blocks of column data according to a column scan operation. The system can identify invalidated blocks and execute a different sub-execution plan of the hybrid query execution plan to scan corresponding rows in tables corresponding to the location of data for the invalidated blocks.

Abstract: A technique partitions a user's original query into plural smaller component queries, each of which has a common part and an instance-specific part. The technique distributes the component queries to plural processor instances of a processor. The plural processor instances transform the respective component queries into query-component responses by acting in parallel, independent of each other. The technique generates a final response based on the query-component responses, e.g., by assembling the component-query responses into the final response. The technique reduces latency because the processor instances work on parts of the user's original query at the same time, rather than as a single stream of consecutive tokens. The plural processor instances have access to a shared cache memory, and utilize relevant data that has been computed in response to previous queries.

Abstract: A method for determining a spatial relationship, includes: obtaining a calculation request for the spatial relationship, the calculation request including a polygonal region and a set of report points; determining a plurality of first coded values corresponding to the polygonal region; determining a plurality of second coded values corresponding to the set of report points and a group of report points corresponding to each of the plurality of second coded values; determining a second coded value matching each of the plurality of first coded values by sorting and matching the plurality of first coded values and the plurality of second coded values respectively; and determining a relationship between the polygonal region and each report point in the set of report points according to a group of report points corresponding to the second coded value matching each of the plurality of first coded values.

Abstract: A system for data partitioning based on discovery of discrete partitioning information. The system can receive data sets in table format from source system. The data can be stored in the source system to be partitioned and transmitted from the source system to a target system. The system can determine a respective partitioning column for each data set. The system can determine a number of partitions. The system can determine, for each data set, a respective set of discrete values from the plurality of discrete values of the respective partitioning column. The number of the discrete values of the set of discrete values can be based at least in part on the number of partitions. The system can the discrete value sets with each other. The system can determine a final set of discrete values based at least in part on the comparison.

Abstract: A query coordinator can receive a query and identify a first portion of the query to be processed by a first data processing system and a second portion of the query to be processed by a second data processing system. The query coordinator can obtain a modified query based on identifying the first portion and the second portion of the query. The query coordinator can define a query processing scheme according to the modified query and provide the query processing scheme to the second data processing system. Based on providing the query processing scheme, the query coordinator can obtain an output of the second data processing system. The query coordinator can identify a second query based on the output and provide the second query to a component of the first data processing system.

Abstract: A document repository may be searched for documents that are similar to a source document. Multiple queries may be generated based on a type of the source document, and the results may be combined in a unified response. User behavior may then be monitored, and implicit and explicit feedback may be gathered to evaluate the performance of the search. The gathered feedback may indicate how relevant each of the result documents are in comparison to the original source document. This feedback may then be used to adjust search parameters for the source document type, such that the performance of subsequent searches may be improved. A model may also be trained to classify implicit feedback using explicit feedback received from users.

Abstract: In some implementations, a redundancy system may receive, from a log storage, a plurality of log files that were generated by a plurality of monitoring programs. The redundancy system may apply a machine learning model to the plurality of log files to detect one or more redundancies. The redundancy system may determine at least one monitoring program, in the plurality of monitoring programs, that is a candidate for removal based on the one or more redundancies. The redundancy system may transmit, to a ticket system, a command to open a ticket associated with the at least one monitoring program.

Abstract: Systems, methods, and non-transitory computer-readable media implements related image search and image modification processes using various search engines and a consolidated graphical user interface. For instance, one or more embodiments involve receiving an input digital image and search input and further modifying the input digital image using the image search results retrieved in response to the search input. In some cases, the search input includes a multi-modal search input having multiple queries (e.g., an image query and a text query), and one or more embodiments involve retrieving the image search results utilizing a weighted combination of the queries. Some implementations involve generating an input embedding for the search input (e.g., the multi-modal search input) and retrieving the image search results using the input embedding.

Abstract: Example quantization-based approximate nearest neighbors (ANN) search methods and systems (e.g., search engines) are tuned to perform at the speed-recall pareto frontier. With a desired search cost or recall as input, embodiments employ Lagrangian-based methods to perform constrained optimization on theoretically-grounded search cost and recall models. The resulting tunings, when paired with the efficient quantization-based ANN implementation of the embodiments, exhibit excellent performance on standard benchmarks while requiring minimal tuning or configuration complexity.

Abstract: A method and system for generating keywords that distinguish a group of selected documents from a larger corpus. The method includes: receiving a corpus that includes a plurality of documents; selecting from the corpus, based on at least one predetermined criterion, a group of the documents; determining a first Inverse Document Frequency (IDF) score for each word in a vocabulary of the group with respect to the corpus; determining a second IDF score for each word in the vocabulary of the group with respect to the group; generating based on the first IDF score and the second IDF score, a Group Frequency Inverse Document Frequency (GFIDF) score for each word in the vocabulary of the group; and assigning based on the generated GFIDF score for each word in the vocabulary of the group, at least one keyword.

Abstract: Provided are methods and systems for searching event sequences. An example method includes receiving a request including a sequence of queries, the sequence of queries including a first query and a second query, the second query succeeding the first query in the sequence; executing the first query over a first dataset to determine a first event; executing the second query over a second dataset to determine a second event, the second event occurring after the first event; and generating, in response to the request, a further sequence including the first event and the second event. The method may include executing the first query over the first dataset to determine a third event and verifying that the third event has occurred after the second event.

Abstract: A method of enabling propagated deletion in a distributed database system comprises receiving a request to delete data in a distributed database system; causing a display of a relevant dataset and a switch between applying propagated deletion or not; receiving a first selection of a subset of records from the relevant dataset using one or more filter functions and a second selection of applying propagated deletion to the subset of records; and applying propagated deletion to the subset of records to generate a new dataset.

Abstract: Predefined patterns are detected in streaming data being transmitted in a distributed computing system that contains a plurality of computing devices wherein the plurality of computing devices are collectively configured to execute a distributed software program. A detection engine can be implemented on a computing device and can be configured to receive streaming data that is being transmitted between computing devices of the system. The detection engine can be further configured to analyze the streaming data to determine if the data matches one or more patterns that is specified by a watch point which can be converted into a happened before language expression. The streaming data can be converted into an intermediate log file that is compared against patterns expressed by a regular expression and if a match is found, an alert can be generated.

Abstract: Multiple main sequences of media items played by a plurality of different media stations are identified. Each main sequence includes one or more content items, and potentially includes syndicated content. At least one main sequence is removed from further consideration, leaving a plurality of remaining main sequences. The at least one main sequence is removed from consideration based, at least in part, on a number of content items included in the at least one main sequence. Consolidated main sequences are generated by merging one or more remaining main sequences satisfying a consolidation criterion. Each consolidated main sequence includes two or more remaining main sequences. The consolidated main sequences are compared to determine whether syndicated content was played on a particular media station.

Abstract: An information processing terminal, connected with an information processing device, to enable viewing data managed in a data management device with ease. The data management device includes: circuitry; and a memory storing executable instructions which, when executed by the circuitry, cause the circuitry to: communicably connect with an information processing terminal and an information processing device via a network; transmit access information to the information processing device, the transmitted access information being used to view data stored in the memory; receive an acquisition request from the information processing device, the received acquisition request being transmitted in response to a user operation performed on the access information in the information processing terminal connected to the information processing device, and being used to acquire the data; and transmit the acquired data to the information processing device based on the access information included in the acquisition request.

Abstract: A method for recommending results to a user from a search query is provided. The method includes receiving, in a search engine, a search query for a media file from a user, identifying a style preference of the user associated with a one or more media file attributes, based on a user-related search history, selecting, from a database, a one or more media files based on the search query and the style preference of the user, determining a style preference score for the one or more media files based on the media file attributes, and recommending to the user a top ranked media file based on the style preference score. A system including a memory storing instructions and one or more processors to execute the instructions to cause the system to perform the above method is also provided.

Abstract: A computer-implemented method for electronic record classification and machine learning inference(s)-informed automated electronic communication includes obtaining one or more electronic records comprising an unstructured component, the unstructured component comprising a textual representation of a set of instructions; extracting a first set of feature vectors comprising features related to a proposed action instruction, and a second set of feature vectors comprising features related to a proposed temporal instruction; computing a first classification inference and a second classification inference that indicates a likely proposed action instruction type and a likely proposed temporal instruction; generating, by the one or more computer processors, a structured digital artifact accessible via a graphical user interface.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for providing suggestions within a document. In one aspect, a method includes obtaining textual input provided to a document editing application by a user device, the textual input being provided to the document editing application for inclusion in a document; identifying performance measures associated with the current editing session for the document, each performance measure being based on session data obtained from the user device during a document editing session, the session data being for the textual input and prior text that was included in the document prior to the textual input; providing the performance measures as input to a suggestion model that was trained using historical performance measures identified in performance logs for historical document editing sessions of users; and throttling textual suggestions during the current editing session based on the output of the suggestion model.

Abstract: When a query is received by a stateful data processing service, the service determines, for each table scan (and associated operations) of a query, whether to select the table scan for execution by a stateless data processing service. The selected table scans are sent to the stateless data processing service for execution, and results are received by the stateful data processing service. The stateful data processing service may also execute other table scans of the query locally, against a local data cache. If the data is not present in the local data cache, then the stateful data processing service will copy the table data into the local data cache before executing the table scan. A query result based on the remote and/or local table scans may then be returned to the client.

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.